October 2009 - developers - lists.mariadb.org

[Maria-developers] Updated (by Monty): Improve the show statistics (60)
by worklog-noreply＠askmonty.org 22 Oct '09

22 Oct '09

----------------------------------------------------------------------- WORKLOG TASK -=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=- TASK...........: Improve the show statistics CREATION DATE..: Thu, 22 Oct 2009, 17:42 SUPERVISOR.....: Monty IMPLEMENTOR....: Bothorsen COPIES TO......: CATEGORY.......: Server-RawIdeaBin TASK ID........: 60 (http://askmonty.org/worklog/?tid=60) VERSION........: WorkLog-3.4 STATUS.........: Assigned PRIORITY.......: 60 WORKED HOURS...: 0 ESTIMATE.......: 0 (hours remain) ORIG. ESTIMATE.: 0 PROGRESS NOTES: -=-=(Monty - Thu, 22 Oct 2009, 18:38)=-=- High Level Description modified. --- /tmp/wklog.60.old.17258 2009-10-22 15:38:47.000000000 +0000 +++ /tmp/wklog.60.new.17258 2009-10-22 15:38:47.000000000 +0000 @@ -1,12 +1,17 @@ In the MariaDB 5.2 branch, some new parts to show statistics have been added. Monty wants these to be improved upon. -1) Change "show table_statistics" to "show table statistics" +1) +Change "show [table|index|user|client]_statistics" to +"show [table|index|user|client] statistics" -2) Add a "flush statistics" that flushes all statistics to the current "flush -table statistics" +Do the same change to the flush command -3) Add where and like clauses to show statistics. +2) +Add a "flush statistics" as a synonym for +flush table_statistics, index statistics..." + +3) Add where and like clauses to show XXXX_statistics. (only like: match most important column) This will be done as a patch to the 5.2 tree. DESCRIPTION: In the MariaDB 5.2 branch, some new parts to show statistics have been added. Monty wants these to be improved upon. 1) Change "show [table|index|user|client]_statistics" to "show [table|index|user|client] statistics" Do the same change to the flush command 2) Add a "flush statistics" as a synonym for flush table_statistics, index statistics..." 3) Add where and like clauses to show XXXX_statistics. (only like: match most important column) This will be done as a patch to the 5.2 tree. Interesting files for this patch are: sql_show.cc, sql_yacc.yy and lex.h. ESTIMATED WORK TIME ESTIMATED COMPLETION DATE ----------------------------------------------------------------------- WorkLog (v3.5.9)

1 0

[Maria-developers] New (by Bothorsen): Improve the show statistics (60)
by worklog-noreply＠askmonty.org 22 Oct '09

22 Oct '09

----------------------------------------------------------------------- WORKLOG TASK -=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=- TASK...........: Improve the show statistics CREATION DATE..: Thu, 22 Oct 2009, 17:42 SUPERVISOR.....: Monty IMPLEMENTOR....: Bothorsen COPIES TO......: CATEGORY.......: Server-RawIdeaBin TASK ID........: 60 (http://askmonty.org/worklog/?tid=60) VERSION........: WorkLog-3.4 STATUS.........: Assigned PRIORITY.......: 60 WORKED HOURS...: 0 ESTIMATE.......: 0 (hours remain) ORIG. ESTIMATE.: 0 PROGRESS NOTES: DESCRIPTION: In the MariaDB 5.2 branch, some new parts to show statistics have been added. Monty wants these to be improved upon. 1) Change "show table_statistics" to "show table statistics" 2) Add a "flush statistics" that flushes all statistics to the current "flush table statistics" 3) Add where and like clauses to show statistics. (only like: match most important column) This will be done as a patch to the 5.2 tree. Interesting files for this patch are: sql_show.cc, sql_yacc.yy and lex.h. ESTIMATED WORK TIME ESTIMATED COMPLETION DATE ----------------------------------------------------------------------- WorkLog (v3.5.9)

1 0

[Maria-developers] New (by Bothorsen): Improve the show statistics (60)
by worklog-noreply＠askmonty.org 22 Oct '09

22 Oct '09

----------------------------------------------------------------------- WORKLOG TASK -=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=- TASK...........: Improve the show statistics CREATION DATE..: Thu, 22 Oct 2009, 17:42 SUPERVISOR.....: Monty IMPLEMENTOR....: Bothorsen COPIES TO......: CATEGORY.......: Server-RawIdeaBin TASK ID........: 60 (http://askmonty.org/worklog/?tid=60) VERSION........: WorkLog-3.4 STATUS.........: Assigned PRIORITY.......: 60 WORKED HOURS...: 0 ESTIMATE.......: 0 (hours remain) ORIG. ESTIMATE.: 0 PROGRESS NOTES: DESCRIPTION: In the MariaDB 5.2 branch, some new parts to show statistics have been added. Monty wants these to be improved upon. 1) Change "show table_statistics" to "show table statistics" 2) Add a "flush statistics" that flushes all statistics to the current "flush table statistics" 3) Add where and like clauses to show statistics. (only like: match most important column) This will be done as a patch to the 5.2 tree. Interesting files for this patch are: sql_show.cc, sql_yacc.yy and lex.h. ESTIMATED WORK TIME ESTIMATED COMPLETION DATE ----------------------------------------------------------------------- WorkLog (v3.5.9)

1 0

[Maria-developers] New (by Bothorsen): Provide an " ORDER BY FIRST_JOIN.column" feature (59)
by worklog-noreply＠askmonty.org 22 Oct '09

22 Oct '09

----------------------------------------------------------------------- WORKLOG TASK -=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=- TASK...........: Provide an "ORDER BY FIRST_JOIN.column" feature CREATION DATE..: Thu, 22 Oct 2009, 12:33 SUPERVISOR.....: Bothorsen IMPLEMENTOR....: COPIES TO......: CATEGORY.......: Server-RawIdeaBin TASK ID........: 59 (http://askmonty.org/worklog/?tid=59) VERSION........: WorkLog-3.4 STATUS.........: Un-Assigned PRIORITY.......: 60 WORKED HOURS...: 0 ESTIMATE.......: 0 (hours remain) ORIG. ESTIMATE.: 0 PROGRESS NOTES: DESCRIPTION: Question: How much effort would be required to provide an "ORDER BY FIRST_JOIN.column" feature? We often do self-join queries and want to order them by index, i.e.: create table t1 (a int, b int, unique key 'by_a_b' (a, b))); select * from t1 x inner join t1 y on (x.b = y.b) where x.a=1 and y.a=2 order by (x|y).b desc limit 10; but it is only ordered by index if you order by the first table in the optimized join order, which could be either x or y depending on which is more selective. we don't want to force the join order because often one is far more selective than the other, but the only way to know what table to order by then is to explain the query first. if there was some way to tell mysql to just order by the first join in the optimized order, that would help us. sometimes we do this across tables too. both of these can be solved by rewriting the query to say "using" instead of "on" in which case we don't have to specify the table name of the column, just "order by b desc". but, we also want to be able to do exclusions, in which case an "on" is required and therefore we run into ambiguous column names: Monty answered: MySQL has an optimization where it knows that if x.b = y.b is used then it can replace x.b with y.b and y.b with x.b in the WHERE part MySQL however doesn't do it for the ORDER BY part and I don't think that should be very hard to do. Question continues: create table t2 (c int, b int, unique key 'by_c_b' (c, b))); select * from t1 x inner join t1 y using (b) left join t2 on (t2.c = 3 and t1.b = t2.b) where x.a=1 and y.a=2 and t2.c is null order by (x| y).b desc limit 10; if we were ordering ascending, i think we could just leave off the order by entirely in this case and it would happen to work since it's reading in index order. but the combination of requiring an "on" and descending sort leaves us unable to use these tricks. if we had either an "ORDER BY FIRST_JOIN.column" or some way to tell the server that 'b' is in fact joined as being equivalent across tables (except where it may be null from left joining) and that we shouldn't have to specify the table name at all, that would save us having to figure out the table from the explain. Monty answers: It may be that the eq-replacment we have would solve this. ESTIMATED WORK TIME ESTIMATED COMPLETION DATE ----------------------------------------------------------------------- WorkLog (v3.5.9)

1 0

[Maria-developers] New (by Bothorsen): Implement a new modifier for flush table (57)
by worklog-noreply＠askmonty.org 22 Oct '09

22 Oct '09

----------------------------------------------------------------------- WORKLOG TASK -=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=- TASK...........: Implement a new modifier for flush table CREATION DATE..: Thu, 22 Oct 2009, 12:25 SUPERVISOR.....: Bothorsen IMPLEMENTOR....: COPIES TO......: Bothorsen CATEGORY.......: Server-RawIdeaBin TASK ID........: 57 (http://askmonty.org/worklog/?tid=57) VERSION........: WorkLog-3.4 STATUS.........: Un-Assigned PRIORITY.......: 60 WORKED HOURS...: 0 ESTIMATE.......: 0 (hours remain) ORIG. ESTIMATE.: 0 PROGRESS NOTES: DESCRIPTION: Create a patch to implement a modifier for "flush table" that allows you to only flush a limited number of blocks. One of the difficulties we have is that we would like to persist some things to disk, but flushing an entire table worth of delay_key_write'd indexes takes minutes, during which time the server is so I/O bound that it cannot serve read traffic. If it was possible to spread this flush over time, we would be able to actually do it without removing the server from read production beforehand. Monty answered this question in an email: The right way to do this is probably to add a background thread that flushes blocks in a 'slow' manner. Note that in the Maria storage engine we do already have this, but I wouldn't recommend you to use this yet; This is more a note about what to expect in the future.. ESTIMATED WORK TIME ESTIMATED COMPLETION DATE ----------------------------------------------------------------------- WorkLog (v3.5.9)

1 0

[Maria-developers] bzr commit into MariaDB 5.1, with Maria 1.5:maria branch (monty:2790)
by Michael Widenius 21 Oct '09

21 Oct '09

#At lp:maria based on revid:psergey@askmonty.org-20091016154458-g89p8v46silum9v7 2790 Michael Widenius 2009-10-21 Updated make_binary_distribution for MariaDB - Changed tar file base name - Added binaries needed by the Maria storage engine modified: scripts/make_binary_distribution.sh === modified file 'scripts/make_binary_distribution.sh' --- a/scripts/make_binary_distribution.sh 2009-09-07 20:50:10 +0000 +++ b/scripts/make_binary_distribution.sh 2009-10-20 23:32:21 +0000 @@ -171,9 +171,9 @@ case $VERSION_NAME in *-ndb-* ) VERSION_NAME=`echo $VERSION_NAME | sed -e 's/[.0-9]*-ndb-//'` ;; esac if [ x"$SHORT_PRODUCT_TAG" != x"" ] ; then - NEW_NAME=mysql-$SHORT_PRODUCT_TAG-$VERSION_NAME-$PLATFORM$SUFFIX + NEW_NAME=mariadb-$SHORT_PRODUCT_TAG-$VERSION_NAME-$PLATFORM$SUFFIX else - NEW_NAME=mysql@MYSQL_SERVER_SUFFIX@-$VERSION_NAME-$PLATFORM$SUFFIX + NEW_NAME=mariadb@MYSQL_SERVER_SUFFIX@-$VERSION_NAME-$PLATFORM$SUFFIX fi # ---------------------------------------------------------------------- @@ -389,6 +389,9 @@ BIN_FILES="extra/comp_err$BS extra/repla extra/resolve_stack_dump$BS extra/mysql_waitpid$BS \ storage/myisam/myisamchk$BS storage/myisam/myisampack$BS \ storage/myisam/myisamlog$BS storage/myisam/myisam_ftdump$BS \ + storage/maria/maria_chk$BS storage/maria/maria_pack$BS \ + storage/maria/maria_ftdump$BS storage/maria/maria_read_log$BS \ + storage/maria/maria_dump_log$BS \ sql/mysqld$BS sql/mysqld-debug$BS \ sql/mysql_tzinfo_to_sql$BS \ server-tools/instance-manager/mysqlmanager$BS \

1 0

[Maria-developers] Windows installer, part 2
by Sergey Petrunya 20 Oct '09

20 Oct '09

Hi! Thanks everyone for input on the input so far! I believe I've collected it all and put into the WL entry: http://askmonty.org/worklog/Server-RawIdeaBin/?tid=55 * If there is something missing there, please bring that argument up again. * Final call for objections/comments on Step#1. Please voice them now, otherwise what is described in step #1 may end up implemented. BR Sergey -- Sergey Petrunia, Software Developer Monty Program AB, http://askmonty.org Blog: http://s.petrunia.net/blog

4 10

[Maria-developers] bzr commit into Mariadb 5.2, with Maria 2.0:maria/5.2 branch (monty:2726)
by Michael Widenius 19 Oct '09

19 Oct '09

#At lp:maria/5.2 based on revid:monty@askmonty.org-20091014125853-gix3sqkenbazrrqz 2726 Michael Widenius 2009-10-19 This is based on the userstatv2 patch from Percona and OurDelta. The original code comes, as far as I know, from Google (Mark Callaghan's team) with additional work from Percona, Ourdelta and Weldon Whipple. This code provides the same functionallity, but with a lot of changes to make it faster and better fit the MariaDB infrastucture. Added new status variables: - Com_show_client_statistics, Com_show_index_statistics, Com_show_table_statistics, Com_show_user_statistics - Access_denied_errors, Busy_time (clock time), Binlog_bytes_written, Cpu_time, Empty_queries, Rows_sent, Rows_read Added new variable / startup option 'userstat' to control if user statistics should be enabled or not Added my_getcputime(); Returns cpu time used by this thread. New FLUSH commands: - FLUSH SLOW QUERY LOG - FLUSH TABLE_STATISTICS - FLUSH INDEX_STATISTICS - FLUSH USER_STATISTICS - FLUSH CLIENT_STATISTICS New SHOW commands: - SHOW CLIENT_STATISTICS - SHOW USER_STATISTICS - SHOW TABLE_STATISTICS - SHOW INDEX_STATISTICS New Information schemas: - CLIENT_STATISTICS - USER_STATISTICS - INDEX_STATISTICS - TABLE_STATISTICS Added support for all new flush commands to mysqladmin Added handler::ha_... wrappers for all handler read calls to do statistics counting - Changed all code to use new ha_... calls - Count number of read rows, changed rows and rows read trough an index Added counting of number of bytes sent to binary log (status variable Binlog_bytes_written) Added counting of access denied errors (status variable Access_denied_erors) Bugs fixed: - Fixed bug in add_to_status() and add_diff_to_status() where longlong variables where threated as long - CLOCK_GETTIME was not propely working on Linuxm added: mysql-test/r/status_user.result mysql-test/t/status_user-master.opt mysql-test/t/status_user.test modified: client/mysqladmin.cc configure.in include/my_sys.h include/mysql_com.h mysql-test/r/information_schema.result mysql-test/r/information_schema_all_engines.result mysql-test/r/information_schema_db.result mysql-test/r/log_slow.result mysql-test/t/log_slow.test mysys/my_getsystime.c sql/authors.h sql/event_data_objects.cc sql/event_db_repository.cc sql/filesort.cc sql/ha_partition.cc sql/handler.cc sql/handler.h sql/item_subselect.cc sql/lex.h sql/log.cc sql/log.h sql/log_event.cc sql/log_event_old.cc sql/mysql_priv.h sql/mysqld.cc sql/opt_range.cc sql/opt_range.h sql/opt_sum.cc sql/records.cc sql/set_var.cc sql/sp.cc sql/sql_acl.cc sql/sql_base.cc sql/sql_class.cc sql/sql_class.h sql/sql_connect.cc sql/sql_cursor.cc sql/sql_handler.cc sql/sql_help.cc sql/sql_insert.cc sql/sql_lex.h sql/sql_parse.cc sql/sql_plugin.cc sql/sql_prepare.cc sql/sql_select.cc sql/sql_servers.cc sql/sql_show.cc sql/sql_table.cc sql/sql_udf.cc sql/sql_update.cc sql/sql_yacc.yy sql/structs.h sql/table.cc sql/table.h sql/tztime.cc per-file messages: client/mysqladmin.cc Added support for all new flush commmands and some common combinations: flush-slow-log flush-table-statistics flush-index-statistics flush-user-statistics flush-client-statistics flush-all-status flush-all-statistics configure.in Added checking if clock_gettime needs the librt. (Fixes Bug #37639 clock_gettime is never used/enabled in Linux/Unix) include/my_sys.h Added my_getcputime() include/mysql_com.h Added LIST_PROCESS_HOST_LEN & new REFRESH target defines mysql-test/r/information_schema.result New information schema tables added mysql-test/r/information_schema_all_engines.result New information schema tables added mysql-test/r/information_schema_db.result New information schema tables added mysql-test/r/log_slow.result Added testing that flosh slow query logs is accepted mysql-test/r/status_user.result Basic testing of user, client, table and index statistics mysql-test/t/log_slow.test Added testing that flosh slow query logs is accepted mysql-test/t/status_user-master.opt Ensure that we get a fresh restart before running status_user.test mysql-test/t/status_user.test Basic testing of user, client, table and index statistics mysys/my_getsystime.c Added my_getcputime() Returns cpu time used by this thread. sql/authors.h Updated authors to have core and original MySQL developers first. sql/event_data_objects.cc Updated call to mysql_reset_thd_for_next_command() sql/event_db_repository.cc Changed to use new ha_... calls sql/filesort.cc Changed to use new ha_... calls sql/ha_partition.cc Changed to use new ha_... calls Fixed comment syntax sql/handler.cc Changed to use new ha_... calls Reset table statistics Added code to update global table and index status Added counting of rows changed sql/handler.h Added table and index statistics variables Added function reset_statistics() Added handler::ha_... wrappers for all handler read calls to do statistics counting Protected all normal read calls to ensure we use the new calls in the server. Made ha_partition a friend class so that partition code can call the old read functions sql/item_subselect.cc Changed to use new ha_... calls sql/lex.h Added keywords for new information schema tables and flush commands sql/log.cc Added flush_slow_log() Added counting of number of bytes sent to binary log Removed not needed test of thd (It's used before, so it's safe to use) Added THD object to MYSQL_BIN_LOG::write_cache() to simplify statistics counting sql/log.h Added new parameter to write_cache() Added flush_slow_log() functions. sql/log_event.cc Updated call to mysql_reset_thd_for_next_command() Changed to use new ha_... calls sql/log_event_old.cc Updated call to mysql_reset_thd_for_next_command() Changed to use new ha_... calls sql/mysql_priv.h Updated call to mysql_reset_thd_for_next_command() Added new statistics functions and variables needed by these. sql/mysqld.cc Added new statistics variables and structures to handle these Added new status variables: - Com_show_client_statistics, Com_show_index_statistics, Com_show_table_statistics, Com_show_user_statistics - Access_denied_errors, Busy_time (clock time), Binlog_bytes_written, Cpu_time, Empty_queries, Rows_set, Rows_read Added new option 'userstat' to control if user statistics should be enabled or not sql/opt_range.cc Changed to use new ha_... calls sql/opt_range.h Changed to use new ha_... calls sql/opt_sum.cc Changed to use new ha_... calls sql/records.cc Changed to use new ha_... calls sql/set_var.cc Added variable 'userstat' sql/sp.cc Changed to use new ha_... calls sql/sql_acl.cc Changed to use new ha_... calls Added counting of access_denied_errors sql/sql_base.cc Added call to statistics functions sql/sql_class.cc Added usage of org_status_var, to store status variables at start of command Added functions THD::update_stats(), THD::update_all_stats() Fixed bug in add_to_status() and add_diff_to_status() where longlong variables where threated as long sql/sql_class.h Added new status variables to status_var Moved variables that was not ulong in status_var last. Added variables to THD for storing temporary values during statistics counting sql/sql_connect.cc Variables and functions to calculate user and client statistics Added counting of access_denied_errors and lost_connections sql/sql_cursor.cc Changed to use new ha_... calls sql/sql_handler.cc Changed to use new ha_... calls sql/sql_help.cc Changed to use new ha_... calls sql/sql_insert.cc Changed to use new ha_... calls sql/sql_lex.h Added SQLCOM_SHOW_USER_STATS, SQLCOM_SHOW_TABLE_STATS, SQLCOM_SHOW_INDEX_STATS, SQLCOM_SHOW_CLIENT_STATS sql/sql_parse.cc Added handling of: - SHOW CLIENT_STATISTICS - SHOW USER_STATISTICS - SHOW TABLE_STATISTICS - SHOW INDEX_STATISTICS Added handling of new FLUSH commands: - FLUSH SLOW QUERY LOGS - FLUSH TABLE_STATISTICS - FLUSH INDEX_STATISTICS - FLUSH USER_STATISTICS - FLUSH CLIENT_STATISTICS Added THD parameter to mysql_reset_thd_for_next_command() Added initialization and calls to user statistics functions Added increment of statistics variables empty_queries, rows_sent and access_denied_errors. Added counting of cpu time per query sql/sql_plugin.cc Changed to use new ha_... calls sql/sql_prepare.cc Updated call to mysql_reset_thd_for_next_command() sql/sql_select.cc Changed to use new ha_... calls Indentation changes sql/sql_servers.cc Changed to use new ha_... calls sql/sql_show.cc Added counting of access denied errors Added function for new information schema tables: - CLIENT_STATISTICS - USER_STATISTICS - INDEX_STATISTICS - TABLE_STATISTICS Changed to use new ha_... calls sql/sql_table.cc Changed to use new ha_... calls sql/sql_udf.cc Changed to use new ha_... calls sql/sql_update.cc Changed to use new ha_... calls sql/sql_yacc.yy Add new show and flush commands sql/structs.h Add name_length to KEY to avoid some strlen Added cache_name to KEY for fast storage of keyvalue in cache Added structs USER_STATS, TABLE_STATS, INDEX_STATS Added function prototypes for statistics functions sql/table.cc Store db+table+index name into keyinfo->cache_name sql/table.h Added new information schema tables sql/tztime.cc Changed to use new ha_... calls === modified file 'client/mysqladmin.cc' --- a/client/mysqladmin.cc 2009-09-07 20:50:10 +0000 +++ b/client/mysqladmin.cc 2009-10-19 17:14:48 +0000 @@ -23,7 +23,7 @@ #include <sys/stat.h> #include <mysql.h> -#define ADMIN_VERSION "8.42" +#define ADMIN_VERSION "9.0" #define MAX_MYSQL_VAR 512 #define SHUTDOWN_DEF_TIMEOUT 3600 /* Wait for shutdown */ #define MAX_TRUNC_LENGTH 3 @@ -96,7 +96,10 @@ enum commands { ADMIN_FLUSH_HOSTS, ADMIN_FLUSH_TABLES, ADMIN_PASSWORD, ADMIN_PING, ADMIN_EXTENDED_STATUS, ADMIN_FLUSH_STATUS, ADMIN_FLUSH_PRIVILEGES, ADMIN_START_SLAVE, ADMIN_STOP_SLAVE, - ADMIN_FLUSH_THREADS, ADMIN_OLD_PASSWORD + ADMIN_FLUSH_THREADS, ADMIN_OLD_PASSWORD, ADMIN_FLUSH_SLOW_LOG, + ADMIN_FLUSH_TABLE_STATISTICS, ADMIN_FLUSH_INDEX_STATISTICS, + ADMIN_FLUSH_USER_STATISTICS, ADMIN_FLUSH_CLIENT_STATISTICS, + ADMIN_FLUSH_ALL_STATUS, ADMIN_FLUSH_ALL_STATISTICS }; static const char *command_names[]= { "create", "drop", "shutdown", @@ -106,7 +109,10 @@ static const char *command_names[]= { "flush-hosts", "flush-tables", "password", "ping", "extended-status", "flush-status", "flush-privileges", "start-slave", "stop-slave", - "flush-threads","old-password", + "flush-threads", "old-password", "flush-slow-log", + "flush-table-statistics", "flush-index-statistics", + "flush-user-statistics", "flush-client-statistics", + "flush-all-status", "flush-all-statistics", NullS }; @@ -518,7 +524,8 @@ static int execute_commands(MYSQL *mysql for (; argc > 0 ; argv++,argc--) { - switch (find_type(argv[0],&command_typelib,2)) { + int command; + switch ((command= find_type(argv[0],&command_typelib,2))) { case ADMIN_CREATE: { char buff[FN_REFLEN+20]; @@ -596,7 +603,11 @@ static int execute_commands(MYSQL *mysql if (mysql_refresh(mysql, (uint) ~(REFRESH_GRANT | REFRESH_STATUS | REFRESH_READ_LOCK | REFRESH_SLAVE | - REFRESH_MASTER))) + REFRESH_MASTER | REFRESH_TABLE_STATS | + REFRESH_INDEX_STATS | + REFRESH_USER_STATS | + REFRESH_SLOW_QUERY_LOG | + REFRESH_CLIENT_STATS))) { my_printf_error(0, "refresh failed; error: '%s'", error_flags, mysql_error(mysql)); @@ -614,7 +625,8 @@ static int execute_commands(MYSQL *mysql case ADMIN_VER: new_line=1; print_version(); - puts("Copyright 2000-2008 MySQL AB, 2008 Sun Microsystems, Inc."); + puts("Copyright 2000-2008 MySQL AB, 2008 Sun Microsystems, Inc,\n" + "2009 Monty Program Ab"); puts("This software comes with ABSOLUTELY NO WARRANTY. This is free software,\nand you are welcome to modify and redistribute it under the GPL license\n"); printf("Server version\t\t%s\n", mysql_get_server_info(mysql)); printf("Protocol version\t%d\n", mysql_get_proto_info(mysql)); @@ -790,9 +802,19 @@ static int execute_commands(MYSQL *mysql } case ADMIN_FLUSH_LOGS: { - if (mysql_refresh(mysql,REFRESH_LOG)) + if (mysql_query(mysql,"flush logs")) { - my_printf_error(0, "refresh failed; error: '%s'", error_flags, + my_printf_error(0, "flush failed; error: '%s'", error_flags, + mysql_error(mysql)); + return -1; + } + break; + } + case ADMIN_FLUSH_SLOW_LOG: + { + if (mysql_query(mysql,"flush slow query logs")) + { + my_printf_error(0, "flush failed; error: '%s'", error_flags, mysql_error(mysql)); return -1; } @@ -802,7 +824,7 @@ static int execute_commands(MYSQL *mysql { if (mysql_query(mysql,"flush hosts")) { - my_printf_error(0, "refresh failed; error: '%s'", error_flags, + my_printf_error(0, "flush failed; error: '%s'", error_flags, mysql_error(mysql)); return -1; } @@ -812,7 +834,7 @@ static int execute_commands(MYSQL *mysql { if (mysql_query(mysql,"flush tables")) { - my_printf_error(0, "refresh failed; error: '%s'", error_flags, + my_printf_error(0, "flush failed; error: '%s'", error_flags, mysql_error(mysql)); return -1; } @@ -822,7 +844,71 @@ static int execute_commands(MYSQL *mysql { if (mysql_query(mysql,"flush status")) { - my_printf_error(0, "refresh failed; error: '%s'", error_flags, + my_printf_error(0, "flush failed; error: '%s'", error_flags, + mysql_error(mysql)); + return -1; + } + break; + } + case ADMIN_FLUSH_TABLE_STATISTICS: + { + if (mysql_query(mysql,"flush table_statistics")) + { + my_printf_error(0, "flush failed; error: '%s'", error_flags, + mysql_error(mysql)); + return -1; + } + break; + } + case ADMIN_FLUSH_INDEX_STATISTICS: + { + if (mysql_query(mysql,"flush index_statistics")) + { + my_printf_error(0, "flush failed; error: '%s'", error_flags, + mysql_error(mysql)); + return -1; + } + break; + } + case ADMIN_FLUSH_USER_STATISTICS: + { + if (mysql_query(mysql,"flush user_statistics")) + { + my_printf_error(0, "flush failed; error: '%s'", error_flags, + mysql_error(mysql)); + return -1; + } + break; + } + case ADMIN_FLUSH_CLIENT_STATISTICS: + { + if (mysql_query(mysql,"flush client_statistics")) + { + my_printf_error(0, "flush failed; error: '%s'", error_flags, + mysql_error(mysql)); + return -1; + } + break; + } + case ADMIN_FLUSH_ALL_STATISTICS: + { + if (mysql_query(mysql, + "flush table_statistics,index_statistics," + "user_statistics,client_statistics")) + { + my_printf_error(0, "flush failed; error: '%s'", error_flags, + mysql_error(mysql)); + return -1; + } + break; + } + case ADMIN_FLUSH_ALL_STATUS: + { + if (mysql_query(mysql, + "flush status,table_statistics,index_statistics," + "user_statistics,client_statistics")) + { + my_printf_error(0, "flush failed; error: '%s'", error_flags, mysql_error(mysql)); return -1; } @@ -994,7 +1080,8 @@ static void print_version(void) static void usage(void) { print_version(); - puts("Copyright 2000-2008 MySQL AB, 2008 Sun Microsystems, Inc."); + puts("Copyright 2000-2008 MySQL AB, 2008 Sun Microsystems, Inc,\n" + "2009 Monty Program Ab"); puts("This software comes with ABSOLUTELY NO WARRANTY. This is free software,\nand you are welcome to modify and redistribute it under the GPL license\n"); puts("Administration program for the mysqld daemon."); printf("Usage: %s [OPTIONS] command command....\n", my_progname); @@ -1002,16 +1089,23 @@ static void usage(void) my_print_variables(my_long_options); print_defaults("my",load_default_groups); puts("\nWhere command is a one or more of: (Commands may be shortened)\n\ - create databasename Create a new database\n\ - debug Instruct server to write debug information to log\n\ - drop databasename Delete a database and all its tables\n\ - extended-status Gives an extended status message from the server\n\ - flush-hosts Flush all cached hosts\n\ - flush-logs Flush all logs\n\ - flush-status Clear status variables\n\ - flush-tables Flush all tables\n\ - flush-threads Flush the thread cache\n\ - flush-privileges Reload grant tables (same as reload)\n\ + create databasename Create a new database\n\ + debug Instruct server to write debug information to log\n\ + drop databasename Delete a database and all its tables\n\ + extended-status Gives an extended status message from the server\n\ + flush-all-statistics Flush all statistics tables\n\ + flush-all-status Flush status and statistics\n\ + flush-client-statistics Flush client statistics\n\ + flush-hosts Flush all cached hosts\n\ + flush-index-statistics Flush index statistics\n\ + flush-logs Flush all logs\n\ + flush-privileges Reload grant tables (same as reload)\n\ + flush-slow-log Flush slow query log\n\ + flush-status Clear status variables\n\ + flush-table-statistics Clear table statistics\n\ + flush-tables Flush all tables\n\ + flush-threads Flush the thread cache\n\ + flush-user-statistics Flush user statistics\n\ kill id,id,... Kill mysql threads"); #if MYSQL_VERSION_ID >= 32200 puts("\ === modified file 'configure.in' --- a/configure.in 2009-10-08 09:43:31 +0000 +++ b/configure.in 2009-10-19 17:14:48 +0000 @@ -829,7 +829,7 @@ AC_CHECK_HEADERS(fcntl.h fenv.h float.h sys/timeb.h sys/types.h sys/un.h sys/vadvise.h sys/wait.h term.h \ unistd.h utime.h sys/utime.h termio.h termios.h sched.h crypt.h alloca.h \ sys/ioctl.h malloc.h sys/malloc.h sys/ipc.h sys/shm.h linux/config.h \ - sys/prctl.h sys/resource.h sys/param.h port.h ieeefp.h \ + sys/prctl.h sys/resource.h sys/param.h port.h ieeefp.h linux/unistd.h \ execinfo.h) AC_CHECK_HEADERS([xfs/xfs.h]) @@ -2096,7 +2096,18 @@ case "$target" in # We also disable for SCO for the time being, the headers for the # thread library we use conflicts with other headers. ;; - *) AC_CHECK_FUNCS(clock_gettime) +*) + # most systems require the program be linked with librt library to use + # the function clock_gettime + my_save_LIBS="$LIBS" + LIBS="" + AC_CHECK_LIB(rt,clock_gettime) + LIBRT=$LIBS + LIBS="$my_save_LIBS" + AC_SUBST(LIBRT) + + LIBS="$LIBS $LIBRT" + AC_CHECK_FUNCS(clock_gettime) ;; esac @@ -2786,7 +2797,7 @@ then fi sql_client_dirs="$sql_client_dirs client" -CLIENT_LIBS="$NON_THREADED_LIBS $openssl_libs $ZLIB_LIBS $STATIC_NSS_FLAGS" +CLIENT_LIBS="$NON_THREADED_LIBS $openssl_libs $ZLIB_LIBS $STATIC_NSS_FLAGS $LIBRT" AC_SUBST(CLIENT_LIBS) AC_SUBST(CLIENT_THREAD_LIBS) === modified file 'include/my_sys.h' --- a/include/my_sys.h 2009-09-07 20:50:10 +0000 +++ b/include/my_sys.h 2009-10-19 17:14:48 +0000 @@ -904,6 +904,7 @@ void my_free_open_file_info(void); extern time_t my_time(myf flags); extern ulonglong my_getsystime(void); +extern ulonglong my_getcputime(void); extern ulonglong my_micro_time(); extern ulonglong my_micro_time_and_time(time_t *time_arg); time_t my_time_possible_from_micro(ulonglong microtime); === modified file 'include/mysql_com.h' --- a/include/mysql_com.h 2008-10-10 15:28:41 +0000 +++ b/include/mysql_com.h 2009-10-19 17:14:48 +0000 @@ -29,6 +29,7 @@ #define SERVER_VERSION_LENGTH 60 #define SQLSTATE_LENGTH 5 +#define LIST_PROCESS_HOST_LEN 64 /* USER_HOST_BUFF_SIZE -- length of string buffer, that is enough to contain @@ -115,6 +116,11 @@ enum enum_server_command thread */ #define REFRESH_MASTER 128 /* Remove all bin logs in the index and truncate the index */ +#define REFRESH_TABLE_STATS 256 /* Refresh table stats hash table */ +#define REFRESH_INDEX_STATS 512 /* Refresh index stats hash table */ +#define REFRESH_USER_STATS 1024 /* Refresh user stats hash table */ +#define REFRESH_SLOW_QUERY_LOG 4096 /* Flush slow query log and rotate*/ +#define REFRESH_CLIENT_STATS 8192 /* Refresh client stats hash table */ /* The following can't be set with mysql_refresh() */ #define REFRESH_READ_LOCK 16384 /* Lock tables for read */ === modified file 'mysql-test/r/information_schema.result' --- a/mysql-test/r/information_schema.result 2009-09-29 20:19:43 +0000 +++ b/mysql-test/r/information_schema.result 2009-10-19 17:14:48 +0000 @@ -45,6 +45,7 @@ NOT (table_schema = 'INFORMATION_SCHEMA' select * from v1 ORDER BY c COLLATE utf8_bin; c CHARACTER_SETS +CLIENT_STATISTICS COLLATIONS COLLATION_CHARACTER_SET_APPLICABILITY COLUMNS @@ -54,6 +55,7 @@ EVENTS FILES GLOBAL_STATUS GLOBAL_VARIABLES +INDEX_STATISTICS INNODB_BUFFER_POOL_PAGES INNODB_BUFFER_POOL_PAGES_BLOB INNODB_BUFFER_POOL_PAGES_INDEX @@ -82,8 +84,10 @@ STATISTICS TABLES TABLE_CONSTRAINTS TABLE_PRIVILEGES +TABLE_STATISTICS TRIGGERS USER_PRIVILEGES +USER_STATISTICS VIEWS XTRADB_ENHANCEMENTS columns_priv @@ -121,6 +125,7 @@ c table_name TABLES TABLES TABLE_CONSTRAINTS TABLE_CONSTRAINTS TABLE_PRIVILEGES TABLE_PRIVILEGES +TABLE_STATISTICS TABLE_STATISTICS TRIGGERS TRIGGERS tables_priv tables_priv time_zone time_zone @@ -140,6 +145,7 @@ c table_name TABLES TABLES TABLE_CONSTRAINTS TABLE_CONSTRAINTS TABLE_PRIVILEGES TABLE_PRIVILEGES +TABLE_STATISTICS TABLE_STATISTICS TRIGGERS TRIGGERS tables_priv tables_priv time_zone time_zone @@ -159,6 +165,7 @@ c table_name TABLES TABLES TABLE_CONSTRAINTS TABLE_CONSTRAINTS TABLE_PRIVILEGES TABLE_PRIVILEGES +TABLE_STATISTICS TABLE_STATISTICS TRIGGERS TRIGGERS tables_priv tables_priv time_zone time_zone @@ -640,12 +647,13 @@ from information_schema.tables where table_schema='information_schema' limit 2; TABLE_NAME TABLE_TYPE ENGINE CHARACTER_SETS SYSTEM VIEW MEMORY -COLLATIONS SYSTEM VIEW MEMORY +CLIENT_STATISTICS SYSTEM VIEW MEMORY show tables from information_schema like "T%"; Tables_in_information_schema (T%) TABLES TABLE_CONSTRAINTS TABLE_PRIVILEGES +TABLE_STATISTICS TRIGGERS create database information_schema; ERROR 42000: Access denied for user 'root'@'localhost' to database 'information_schema' @@ -655,6 +663,7 @@ Tables_in_information_schema (T%) Table_ TABLES SYSTEM VIEW TABLE_CONSTRAINTS SYSTEM VIEW TABLE_PRIVILEGES SYSTEM VIEW +TABLE_STATISTICS SYSTEM VIEW TRIGGERS SYSTEM VIEW create table t1(a int); ERROR 42S02: Unknown table 't1' in information_schema @@ -667,6 +676,7 @@ Tables_in_information_schema (T%) TABLES TABLE_CONSTRAINTS TABLE_PRIVILEGES +TABLE_STATISTICS TRIGGERS select table_name from tables where table_name='user'; table_name @@ -856,6 +866,7 @@ TABLE_NAME COLUMN_NAME PRIVILEGES COLUMNS TABLE_NAME select COLUMN_PRIVILEGES TABLE_NAME select FILES TABLE_NAME select +INDEX_STATISTICS TABLE_NAME select KEY_COLUMN_USAGE TABLE_NAME select PARTITIONS TABLE_NAME select REFERENTIAL_CONSTRAINTS TABLE_NAME select @@ -863,6 +874,7 @@ STATISTICS TABLE_NAME select TABLES TABLE_NAME select TABLE_CONSTRAINTS TABLE_NAME select TABLE_PRIVILEGES TABLE_NAME select +TABLE_STATISTICS TABLE_NAME select VIEWS TABLE_NAME select INNODB_BUFFER_POOL_PAGES_INDEX table_name select INNODB_INDEX_STATS table_name select === modified file 'mysql-test/r/information_schema_all_engines.result' --- a/mysql-test/r/information_schema_all_engines.result 2009-08-03 20:09:53 +0000 +++ b/mysql-test/r/information_schema_all_engines.result 2009-10-19 17:14:48 +0000 @@ -2,6 +2,7 @@ use INFORMATION_SCHEMA; show tables; Tables_in_information_schema CHARACTER_SETS +CLIENT_STATISTICS COLLATIONS COLLATION_CHARACTER_SET_APPLICABILITY COLUMNS @@ -11,6 +12,7 @@ EVENTS FILES GLOBAL_STATUS GLOBAL_VARIABLES +INDEX_STATISTICS KEY_COLUMN_USAGE PARTITIONS PLUGINS @@ -26,8 +28,10 @@ STATISTICS TABLES TABLE_CONSTRAINTS TABLE_PRIVILEGES +TABLE_STATISTICS TRIGGERS USER_PRIVILEGES +USER_STATISTICS VIEWS INNODB_BUFFER_POOL_PAGES PBXT_STATISTICS @@ -60,6 +64,7 @@ c2.column_name LIKE '%SCHEMA%' ); table_name column_name CHARACTER_SETS CHARACTER_SET_NAME +CLIENT_STATISTICS CLIENT COLLATIONS COLLATION_NAME COLLATION_CHARACTER_SET_APPLICABILITY COLLATION_NAME COLUMNS TABLE_SCHEMA @@ -69,6 +74,7 @@ EVENTS EVENT_SCHEMA FILES TABLE_SCHEMA GLOBAL_STATUS VARIABLE_NAME GLOBAL_VARIABLES VARIABLE_NAME +INDEX_STATISTICS TABLE_SCHEMA KEY_COLUMN_USAGE CONSTRAINT_SCHEMA PARTITIONS TABLE_SCHEMA PLUGINS PLUGIN_NAME @@ -84,8 +90,10 @@ STATISTICS TABLE_SCHEMA TABLES TABLE_SCHEMA TABLE_CONSTRAINTS CONSTRAINT_SCHEMA TABLE_PRIVILEGES TABLE_SCHEMA +TABLE_STATISTICS TABLE_SCHEMA TRIGGERS TRIGGER_SCHEMA USER_PRIVILEGES GRANTEE +USER_STATISTICS USER VIEWS TABLE_SCHEMA INNODB_BUFFER_POOL_PAGES page_type PBXT_STATISTICS ID @@ -118,6 +126,7 @@ c2.column_name LIKE '%SCHEMA%' ); table_name column_name CHARACTER_SETS CHARACTER_SET_NAME +CLIENT_STATISTICS CLIENT COLLATIONS COLLATION_NAME COLLATION_CHARACTER_SET_APPLICABILITY COLLATION_NAME COLUMNS TABLE_SCHEMA @@ -127,6 +136,7 @@ EVENTS EVENT_SCHEMA FILES TABLE_SCHEMA GLOBAL_STATUS VARIABLE_NAME GLOBAL_VARIABLES VARIABLE_NAME +INDEX_STATISTICS TABLE_SCHEMA KEY_COLUMN_USAGE CONSTRAINT_SCHEMA PARTITIONS TABLE_SCHEMA PLUGINS PLUGIN_NAME @@ -142,8 +152,10 @@ STATISTICS TABLE_SCHEMA TABLES TABLE_SCHEMA TABLE_CONSTRAINTS CONSTRAINT_SCHEMA TABLE_PRIVILEGES TABLE_SCHEMA +TABLE_STATISTICS TABLE_SCHEMA TRIGGERS TRIGGER_SCHEMA USER_PRIVILEGES GRANTEE +USER_STATISTICS USER VIEWS TABLE_SCHEMA INNODB_BUFFER_POOL_PAGES page_type PBXT_STATISTICS ID @@ -182,6 +194,7 @@ group by c2.column_type order by num lim group by t.table_name order by num1, t.table_name; table_name group_concat(t.table_schema, '.', t.table_name) num1 CHARACTER_SETS information_schema.CHARACTER_SETS 1 +CLIENT_STATISTICS information_schema.CLIENT_STATISTICS 1 COLLATIONS information_schema.COLLATIONS 1 COLLATION_CHARACTER_SET_APPLICABILITY information_schema.COLLATION_CHARACTER_SET_APPLICABILITY 1 COLUMNS information_schema.COLUMNS 1 @@ -191,6 +204,7 @@ EVENTS information_schema.EVENTS 1 FILES information_schema.FILES 1 GLOBAL_STATUS information_schema.GLOBAL_STATUS 1 GLOBAL_VARIABLES information_schema.GLOBAL_VARIABLES 1 +INDEX_STATISTICS information_schema.INDEX_STATISTICS 1 INNODB_BUFFER_POOL_PAGES information_schema.INNODB_BUFFER_POOL_PAGES 1 INNODB_BUFFER_POOL_PAGES_BLOB information_schema.INNODB_BUFFER_POOL_PAGES_BLOB 1 INNODB_BUFFER_POOL_PAGES_INDEX information_schema.INNODB_BUFFER_POOL_PAGES_INDEX 1 @@ -220,8 +234,10 @@ STATISTICS information_schema.STATISTICS TABLES information_schema.TABLES 1 TABLE_CONSTRAINTS information_schema.TABLE_CONSTRAINTS 1 TABLE_PRIVILEGES information_schema.TABLE_PRIVILEGES 1 +TABLE_STATISTICS information_schema.TABLE_STATISTICS 1 TRIGGERS information_schema.TRIGGERS 1 USER_PRIVILEGES information_schema.USER_PRIVILEGES 1 +USER_STATISTICS information_schema.USER_STATISTICS 1 VIEWS information_schema.VIEWS 1 XTRADB_ENHANCEMENTS information_schema.XTRADB_ENHANCEMENTS 1 Database: information_schema @@ -229,6 +245,7 @@ Database: information_schema | Tables | +---------------------------------------+ | CHARACTER_SETS | +| CLIENT_STATISTICS | | COLLATIONS | | COLLATION_CHARACTER_SET_APPLICABILITY | | COLUMNS | @@ -238,6 +255,7 @@ Database: information_schema | FILES | | GLOBAL_STATUS | | GLOBAL_VARIABLES | +| INDEX_STATISTICS | | KEY_COLUMN_USAGE | | PARTITIONS | | PLUGINS | @@ -253,8 +271,10 @@ Database: information_schema | TABLES | | TABLE_CONSTRAINTS | | TABLE_PRIVILEGES | +| TABLE_STATISTICS | | TRIGGERS | | USER_PRIVILEGES | +| USER_STATISTICS | | VIEWS | | INNODB_BUFFER_POOL_PAGES | | PBXT_STATISTICS | @@ -277,6 +297,7 @@ Database: INFORMATION_SCHEMA | Tables | +---------------------------------------+ | CHARACTER_SETS | +| CLIENT_STATISTICS | | COLLATIONS | | COLLATION_CHARACTER_SET_APPLICABILITY | | COLUMNS | @@ -286,6 +307,7 @@ Database: INFORMATION_SCHEMA | FILES | | GLOBAL_STATUS | | GLOBAL_VARIABLES | +| INDEX_STATISTICS | | KEY_COLUMN_USAGE | | PARTITIONS | | PLUGINS | @@ -301,8 +323,10 @@ Database: INFORMATION_SCHEMA | TABLES | | TABLE_CONSTRAINTS | | TABLE_PRIVILEGES | +| TABLE_STATISTICS | | TRIGGERS | | USER_PRIVILEGES | +| USER_STATISTICS | | VIEWS | | INNODB_BUFFER_POOL_PAGES | | PBXT_STATISTICS | @@ -328,5 +352,5 @@ Wildcard: inf_rmation_schema +--------------------+ SELECT table_schema, count(*) FROM information_schema.TABLES WHERE table_schema IN ('mysql', 'INFORMATION_SCHEMA', 'test', 'mysqltest') AND table_name<>'ndb_binlog_index' AND table_name<>'ndb_apply_status' GROUP BY TABLE_SCHEMA; table_schema count(*) -information_schema 43 +information_schema 47 mysql 22 === modified file 'mysql-test/r/information_schema_db.result' --- a/mysql-test/r/information_schema_db.result 2009-09-07 20:50:10 +0000 +++ b/mysql-test/r/information_schema_db.result 2009-10-19 17:14:48 +0000 @@ -7,6 +7,7 @@ Tables_in_information_schema (T%) TABLES TABLE_CONSTRAINTS TABLE_PRIVILEGES +TABLE_STATISTICS TRIGGERS create database `inf%`; create database mbase; === modified file 'mysql-test/r/log_slow.result' --- a/mysql-test/r/log_slow.result 2009-09-03 14:05:38 +0000 +++ b/mysql-test/r/log_slow.result 2009-10-19 17:14:48 +0000 @@ -56,5 +56,6 @@ last_insert_id int(11) NO NULL insert_id int(11) NO NULL server_id int(10) unsigned NO NULL sql_text mediumtext NO NULL +flush slow query logs; set @@log_slow_filter=default; set @@log_slow_verbosity=default; === added file 'mysql-test/r/status_user.result' --- a/mysql-test/r/status_user.result 1970-01-01 00:00:00 +0000 +++ b/mysql-test/r/status_user.result 2009-10-19 17:14:48 +0000 @@ -0,0 +1,166 @@ +DROP TABLE IF EXISTS t1; +select variable_value from information_schema.global_status where variable_name="handler_read_key" into @global_read_key; +show columns from information_schema.client_statistics; +Field Type Null Key Default Extra +CLIENT varchar(64) NO +TOTAL_CONNECTIONS int(21) NO 0 +CONCURRENT_CONNECTIONS int(21) NO 0 +CONNECTED_TIME int(21) NO 0 +BUSY_TIME double NO 0 +CPU_TIME double NO 0 +BYTES_RECEIVED int(21) NO 0 +BYTES_SENT int(21) NO 0 +BINLOG_BYTES_WRITTEN int(21) NO 0 +ROWS_READ int(21) NO 0 +ROWS_SENT int(21) NO 0 +ROWS_DELETED int(21) NO 0 +ROWS_INSERTED int(21) NO 0 +ROWS_UPDATED int(21) NO 0 +SELECT_COMMANDS int(21) NO 0 +UPDATE_COMMANDS int(21) NO 0 +OTHER_COMMANDS int(21) NO 0 +COMMIT_TRANSACTIONS int(21) NO 0 +ROLLBACK_TRANSACTIONS int(21) NO 0 +DENIED_CONNECTIONS int(21) NO 0 +LOST_CONNECTIONS int(21) NO 0 +ACCESS_DENIED int(21) NO 0 +EMPTY_QUERIES int(21) NO 0 +show columns from information_schema.user_statistics; +Field Type Null Key Default Extra +USER varchar(48) NO +TOTAL_CONNECTIONS int(21) NO 0 +CONCURRENT_CONNECTIONS int(21) NO 0 +CONNECTED_TIME int(21) NO 0 +BUSY_TIME double NO 0 +CPU_TIME double NO 0 +BYTES_RECEIVED int(21) NO 0 +BYTES_SENT int(21) NO 0 +BINLOG_BYTES_WRITTEN int(21) NO 0 +ROWS_READ int(21) NO 0 +ROWS_SENT int(21) NO 0 +ROWS_DELETED int(21) NO 0 +ROWS_INSERTED int(21) NO 0 +ROWS_UPDATED int(21) NO 0 +SELECT_COMMANDS int(21) NO 0 +UPDATE_COMMANDS int(21) NO 0 +OTHER_COMMANDS int(21) NO 0 +COMMIT_TRANSACTIONS int(21) NO 0 +ROLLBACK_TRANSACTIONS int(21) NO 0 +DENIED_CONNECTIONS int(21) NO 0 +LOST_CONNECTIONS int(21) NO 0 +ACCESS_DENIED int(21) NO 0 +EMPTY_QUERIES int(21) NO 0 +show columns from information_schema.index_statistics; +Field Type Null Key Default Extra +TABLE_SCHEMA varchar(192) NO +TABLE_NAME varchar(192) NO +INDEX_NAME varchar(192) NO +ROWS_READ int(21) NO 0 +show columns from information_schema.table_statistics; +Field Type Null Key Default Extra +TABLE_SCHEMA varchar(192) NO +TABLE_NAME varchar(192) NO +ROWS_READ int(21) NO 0 +ROWS_CHANGED int(21) NO 0 +ROWS_CHANGED_X_INDEXES int(21) NO 0 +set @save_general_log=@@global.general_log; +set @@global.general_log=0; +set @@global.userstat=1; +flush status; +create table t1 (a int, primary key (a), b int default 0) engine=myisam; +insert into t1 (a) values (1),(2),(3),(4); +update t1 set b=1; +update t1 set b=5 where a=2; +delete from t1 where a=3; +/* Empty query */ +select * from t1 where a=999; +a b +drop table t1; +create table t1 (a int, primary key (a), b int default 0) engine=innodb; +begin; +insert into t1 values(1,1); +commit; +begin; +insert into t1 values(2,2); +commit; +begin; +insert into t1 values(3,3); +rollback; +drop table t1; +select sleep(1); +sleep(1) +0 +show status like "rows%"; +Variable_name Value +Rows_read 6 +Rows_sent 1 +show status like "ha%"; +Variable_name Value +Handler_commit 10 +Handler_delete 1 +Handler_discover 0 +Handler_prepare 10 +Handler_read_first 0 +Handler_read_key 3 +Handler_read_next 0 +Handler_read_prev 0 +Handler_read_rnd 0 +Handler_read_rnd_next 5 +Handler_rollback 2 +Handler_savepoint 0 +Handler_savepoint_rollback 0 +Handler_update 5 +Handler_write 7 +select variable_value - @global_read_key as "handler_read_key" from information_schema.global_status where variable_name="handler_read_key"; +handler_read_key +3 +set @@global.userstat=0; +select * from information_schema.index_statistics; +TABLE_SCHEMA TABLE_NAME INDEX_NAME ROWS_READ +test t1 PRIMARY 2 +select * from information_schema.table_statistics; +TABLE_SCHEMA TABLE_NAME ROWS_READ ROWS_CHANGED ROWS_CHANGED_X_INDEXES +test t1 6 13 13 +show table_statistics; +Table_schema Table_name Rows_read Rows_changed Rows_changed_x_#indexes +test t1 6 13 13 +show index_statistics; +Table_schema Table_name Index_name Rows_read +test t1 PRIMARY 2 +select TOTAL_CONNECTIONS, CONCURRENT_CONNECTIONS, ROWS_READ, ROWS_SENT, +ROWS_DELETED, ROWS_INSERTED, ROWS_UPDATED, SELECT_COMMANDS, +UPDATE_COMMANDS, OTHER_COMMANDS, COMMIT_TRANSACTIONS, +ROLLBACK_TRANSACTIONS, DENIED_CONNECTIONS, LOST_CONNECTIONS, +ACCESS_DENIED, EMPTY_QUERIES from information_schema.client_statistics; +TOTAL_CONNECTIONS CONCURRENT_CONNECTIONS ROWS_READ ROWS_SENT ROWS_DELETED ROWS_INSERTED ROWS_UPDATED SELECT_COMMANDS UPDATE_COMMANDS OTHER_COMMANDS COMMIT_TRANSACTIONS ROLLBACK_TRANSACTIONS DENIED_CONNECTIONS LOST_CONNECTIONS ACCESS_DENIED EMPTY_QUERIES +1 0 6 2 1 8 5 3 11 9 10 2 0 0 0 1 +select TOTAL_CONNECTIONS, CONCURRENT_CONNECTIONS, ROWS_READ, ROWS_SENT, +ROWS_DELETED, ROWS_INSERTED, ROWS_UPDATED, SELECT_COMMANDS, +UPDATE_COMMANDS, OTHER_COMMANDS, COMMIT_TRANSACTIONS, +ROLLBACK_TRANSACTIONS, DENIED_CONNECTIONS, LOST_CONNECTIONS, +ACCESS_DENIED, EMPTY_QUERIES from information_schema.user_statistics; +TOTAL_CONNECTIONS CONCURRENT_CONNECTIONS ROWS_READ ROWS_SENT ROWS_DELETED ROWS_INSERTED ROWS_UPDATED SELECT_COMMANDS UPDATE_COMMANDS OTHER_COMMANDS COMMIT_TRANSACTIONS ROLLBACK_TRANSACTIONS DENIED_CONNECTIONS LOST_CONNECTIONS ACCESS_DENIED EMPTY_QUERIES +1 0 6 2 1 8 5 3 11 9 10 2 0 0 0 1 +flush table_statistics; +flush index_statistics; +select * from information_schema.index_statistics; +TABLE_SCHEMA TABLE_NAME INDEX_NAME ROWS_READ +select * from information_schema.table_statistics; +TABLE_SCHEMA TABLE_NAME ROWS_READ ROWS_CHANGED ROWS_CHANGED_X_INDEXES +show status like "%statistics%"; +Variable_name Value +Com_show_client_statistics 0 +Com_show_index_statistics 1 +Com_show_table_statistics 1 +Com_show_user_statistics 0 +select connected_time <> 0, busy_time <> 0, bytes_received <> 0, +bytes_sent <> 0, binlog_bytes_written <> 0 +from information_schema.user_statistics; +connected_time <> 0 busy_time <> 0 bytes_received <> 0 bytes_sent <> 0 binlog_bytes_written <> 0 +1 1 1 1 1 +select connected_time <> 0, busy_time <> 0, bytes_received <> 0, +bytes_sent <> 0, binlog_bytes_written <> 0 +from information_schema.client_statistics; +connected_time <> 0 busy_time <> 0 bytes_received <> 0 bytes_sent <> 0 binlog_bytes_written <> 0 +1 1 1 1 1 +set @@global.general_log=@save_general_log; === modified file 'mysql-test/t/log_slow.test' --- a/mysql-test/t/log_slow.test 2009-09-03 14:05:38 +0000 +++ b/mysql-test/t/log_slow.test 2009-10-19 17:14:48 +0000 @@ -36,6 +36,12 @@ select @@log_slow_verbosity; show fields from mysql.slow_log; +# +# Check flush command +# + +flush slow query logs; + # Reset used variables set @@log_slow_filter=default; === added file 'mysql-test/t/status_user-master.opt' --- a/mysql-test/t/status_user-master.opt 1970-01-01 00:00:00 +0000 +++ b/mysql-test/t/status_user-master.opt 2009-10-19 17:14:48 +0000 @@ -0,0 +1 @@ +--force-restart === added file 'mysql-test/t/status_user.test' --- a/mysql-test/t/status_user.test 1970-01-01 00:00:00 +0000 +++ b/mysql-test/t/status_user.test 2009-10-19 17:14:48 +0000 @@ -0,0 +1,97 @@ +# +# Testing of user status (the userstat variable). +# Note that this test requires a fresh restart to not problems with +# old status + +-- source include/have_innodb.inc +-- source include/have_log_bin.inc + +--disable_warnings +DROP TABLE IF EXISTS t1; +--enable_warnings + +select variable_value from information_schema.global_status where variable_name="handler_read_key" into @global_read_key; +show columns from information_schema.client_statistics; +show columns from information_schema.user_statistics; +show columns from information_schema.index_statistics; +show columns from information_schema.table_statistics; + +# Disable logging to get right number of writes into the tables. +set @save_general_log=@@global.general_log; +set @@global.general_log=0; +set @@global.userstat=1; +flush status; + +create table t1 (a int, primary key (a), b int default 0) engine=myisam; +insert into t1 (a) values (1),(2),(3),(4); +update t1 set b=1; +update t1 set b=5 where a=2; +delete from t1 where a=3; + +/* Empty query */ +select * from t1 where a=999; + +drop table t1; + +# +# Test the commit and rollback are counted +# + +create table t1 (a int, primary key (a), b int default 0) engine=innodb; +begin; +insert into t1 values(1,1); +commit; +begin; +insert into t1 values(2,2); +commit; +begin; +insert into t1 values(3,3); +rollback; +drop table t1; + +select sleep(1); + +show status like "rows%"; +show status like "ha%"; +select variable_value - @global_read_key as "handler_read_key" from information_schema.global_status where variable_name="handler_read_key"; + +# Ensure that the following commands doesn't change statistics + +set @@global.userstat=0; + +# +# Check that we got right statistics +# +select * from information_schema.index_statistics; +select * from information_schema.table_statistics; +show table_statistics; +show index_statistics; +select TOTAL_CONNECTIONS, CONCURRENT_CONNECTIONS, ROWS_READ, ROWS_SENT, + ROWS_DELETED, ROWS_INSERTED, ROWS_UPDATED, SELECT_COMMANDS, + UPDATE_COMMANDS, OTHER_COMMANDS, COMMIT_TRANSACTIONS, + ROLLBACK_TRANSACTIONS, DENIED_CONNECTIONS, LOST_CONNECTIONS, + ACCESS_DENIED, EMPTY_QUERIES from information_schema.client_statistics; +select TOTAL_CONNECTIONS, CONCURRENT_CONNECTIONS, ROWS_READ, ROWS_SENT, + ROWS_DELETED, ROWS_INSERTED, ROWS_UPDATED, SELECT_COMMANDS, + UPDATE_COMMANDS, OTHER_COMMANDS, COMMIT_TRANSACTIONS, + ROLLBACK_TRANSACTIONS, DENIED_CONNECTIONS, LOST_CONNECTIONS, + ACCESS_DENIED, EMPTY_QUERIES from information_schema.user_statistics; +flush table_statistics; +flush index_statistics; +select * from information_schema.index_statistics; +select * from information_schema.table_statistics; +show status like "%statistics%"; + +# +# Test that some variables are not 0 +# + +select connected_time <> 0, busy_time <> 0, bytes_received <> 0, + bytes_sent <> 0, binlog_bytes_written <> 0 + from information_schema.user_statistics; +select connected_time <> 0, busy_time <> 0, bytes_received <> 0, + bytes_sent <> 0, binlog_bytes_written <> 0 + from information_schema.client_statistics; + +# Cleanup +set @@global.general_log=@save_general_log; === modified file 'mysys/my_getsystime.c' --- a/mysys/my_getsystime.c 2008-04-28 16:24:05 +0000 +++ b/mysys/my_getsystime.c 2009-10-19 17:14:48 +0000 @@ -28,6 +28,10 @@ #ifdef __NETWARE__ #include <nks/time.h> #endif +#ifdef HAVE_LINUX_UNISTD_H +#include <linux/unistd.h> +#endif + ulonglong my_getsystime() { @@ -222,3 +226,25 @@ time_t my_time_possible_from_micro(ulong return (time_t) (microtime / 1000000); #endif /* defined(__WIN__) */ } + + +/* + Return cpu time in milliseconds * 10 +*/ + +ulonglong my_getcputime() +{ +#ifdef HAVE_CLOCK_GETTIME + struct timespec tp; + if (clock_gettime(CLOCK_THREAD_CPUTIME_ID, &tp)) + return 0; + return (ulonglong)tp.tv_sec*10000000+(ulonglong)tp.tv_nsec/100; +#elif defined(__NR_clock_gettime) + struct timespec tp; + if (syscall(__NR_clock_gettime, CLOCK_THREAD_CPUTIME_ID, &tp)) + return 0; + return (ulonglong)tp.tv_sec*10000000+(ulonglong)tp.tv_nsec/100; +#else + return 0; +#endif /* HAVE_CLOCK_GETTIME */ +} === modified file 'sql/authors.h' --- a/sql/authors.h 2007-03-16 06:39:07 +0000 +++ b/sql/authors.h 2009-10-19 17:14:48 +0000 @@ -34,23 +34,35 @@ struct show_table_authors_st { */ struct show_table_authors_st show_table_authors[]= { + { "Michael (Monty) Widenius", "Tusby, Finland", + "Lead developer and main author" }, + { "David Axmark", "London, England", + "MySQL founder; Small stuff long time ago, Monty ripped it out!" }, + { "Sergei Golubchik", "Kerpen, Germany", + "Full-text search, precision math" }, + { "Igor Babaev", "Bellevue, USA", "Optimizer, keycache, core work"}, + { "Sergey Petrunia", "St. Petersburg, Russia", "Optimizer"}, + { "Oleksandr Byelkin", "Lugansk, Ukraine", + "Query Cache (4.0), Subqueries (4.1), Views (5.0)" }, { "Brian (Krow) Aker", "Seattle, WA, USA", "Architecture, archive, federated, bunch of little stuff :)" }, - { "Venu Anuganti", "", "Client/server protocol (4.1)" }, - { "David Axmark", "Uppsala, Sweden", - "Small stuff long time ago, Monty ripped it out!" }, + { "Kristian Nielsen", "Copenhagen, Denmark", + "General build stuff," }, { "Alexander (Bar) Barkov", "Izhevsk, Russia", "Unicode and character sets (4.1)" }, + { "Guilhem Bichot", "Bordeaux, France", "Replication (since 4.0)" }, + { "Venu Anuganti", "", "Client/server protocol (4.1)" }, + { "Konstantin Osipov", "Moscow, Russia", + "Prepared statements (4.1), Cursors (5.0)" }, + { "Dmitri Lenev", "Moscow, Russia", + "Time zones support (4.1), Triggers (5.0)" }, { "Omer BarNir", "Sunnyvale, CA, USA", "Testing (sometimes) and general QA stuff" }, - { "Guilhem Bichot", "Bordeaux, France", "Replication (since 4.0)" }, { "John Birrell", "", "Emulation of pthread_mutex() for OS/2" }, { "Andreas F. Bobak", "", "AGGREGATE extension to user-defined functions" }, { "Alexey Botchkov (Holyfoot)", "Izhevsk, Russia", "GIS extensions (4.1), embedded server (4.1), precision math (5.0)"}, { "Reggie Burnett", "Nashville, TN, USA", "Windows development, Connectors" }, - { "Oleksandr Byelkin", "Lugansk, Ukraine", - "Query Cache (4.0), Subqueries (4.1), Views (5.0)" }, { "Kent Boortz", "Orebro, Sweden", "Test platform, and general build stuff" }, { "Tim Bunce", "", "mysqlhotcopy" }, { "Yves Carlier", "", "mysqlaccess" }, @@ -67,8 +79,6 @@ struct show_table_authors_st show_table_ { "Yuri Dario", "", "OS/2 port" }, { "Andrei Elkin", "Espoo, Finland", "Replication" }, { "Patrick Galbraith", "Sharon, NH", "Federated Engine, mysqlslap" }, - { "Sergei Golubchik", "Kerpen, Germany", - "Full-text search, precision math" }, { "Lenz Grimmer", "Hamburg, Germany", "Production (build and release) engineering" }, { "Nikolay Grishakin", "Austin, TX, USA", "Testing - Server" }, @@ -83,8 +93,6 @@ struct show_table_authors_st show_table_ { "Hakan K��ky��lmaz", "Walldorf, Germany", "Testing - Server" }, { "Greg (Groggy) Lehey", "Uchunga, SA, Australia", "Backup" }, { "Matthias Leich", "Berlin, Germany", "Testing - Server" }, - { "Dmitri Lenev", "Moscow, Russia", - "Time zones support (4.1), Triggers (5.0)" }, { "Arjen Lentz", "Brisbane, Australia", "Documentation (2001-2004), Dutch error messages, LOG2()" }, { "Marc Liyanage", "", "Created Mac OS X packages" }, @@ -96,8 +104,6 @@ struct show_table_authors_st show_table_ { "Jonathan (Jeb) Miller", "Kyle, TX, USA", "Testing - Cluster, Replication" }, { "Elliot Murphy", "Cocoa, FL, USA", "Replication and backup" }, - { "Kristian Nielsen", "Copenhagen, Denmark", - "General build stuff" }, { "Pekka Nouisiainen", "Stockholm, Sweden", "NDB Cluster: BLOB support, character set support, ordered indexes" }, { "Alexander Nozdrin", "Moscow, Russia", @@ -105,8 +111,6 @@ struct show_table_authors_st show_table_ { "Per Eric Olsson", "", "Testing of dynamic record format" }, { "Jonas Oreland", "Stockholm, Sweden", "NDB Cluster, Online Backup, lots of other things" }, - { "Konstantin Osipov", "Moscow, Russia", - "Prepared statements (4.1), Cursors (5.0)" }, { "Alexander (Sasha) Pachev", "Provo, UT, USA", "Statement-based replication, SHOW CREATE TABLE, mysql-bench" }, { "Irena Pancirov", "", "Port to Windows with Borland compiler" }, @@ -144,9 +148,9 @@ struct show_table_authors_st show_table_ { "Sergey Vojtovich", "Izhevsk, Russia", "Plugins infrastructure (5.1)" }, { "Matt Wagner", "Northfield, MN, USA", "Bug fixing" }, { "Jim Winstead Jr.", "Los Angeles, CA, USA", "Bug fixing" }, - { "Michael (Monty) Widenius", "Tusby, Finland", - "Lead developer and main author" }, { "Peter Zaitsev", "Tacoma, WA, USA", "SHA1(), AES_ENCRYPT(), AES_DECRYPT(), bug fixing" }, + {"Mark Mark Callaghan", "Texas, USA", "Statistics patches"}, + {"Percona", "CA, USA", "Microslow patches"}, {NULL, NULL, NULL} }; === modified file 'sql/event_data_objects.cc' --- a/sql/event_data_objects.cc 2009-09-15 10:46:35 +0000 +++ b/sql/event_data_objects.cc 2009-10-19 17:14:48 +0000 @@ -1366,7 +1366,7 @@ Event_job_data::execute(THD *thd, bool d DBUG_ENTER("Event_job_data::execute"); - mysql_reset_thd_for_next_command(thd); + mysql_reset_thd_for_next_command(thd, 0); /* MySQL parser currently assumes that current database is either === modified file 'sql/event_db_repository.cc' --- a/sql/event_db_repository.cc 2009-02-15 10:58:34 +0000 +++ b/sql/event_db_repository.cc 2009-10-19 17:14:48 +0000 @@ -404,17 +404,18 @@ Event_db_repository::index_read_for_db_f } key_copy(key_buf, event_table->record[0], key_info, key_len); - if (!(ret= event_table->file->index_read_map(event_table->record[0], key_buf, - (key_part_map)1, - HA_READ_PREFIX))) + if (!(ret= event_table->file->ha_index_read_map(event_table->record[0], + key_buf, + (key_part_map)1, + HA_READ_PREFIX))) { DBUG_PRINT("info",("Found rows. Let's retrieve them. ret=%d", ret)); do { ret= copy_event_to_schema_table(thd, schema_table, event_table); if (ret == 0) - ret= event_table->file->index_next_same(event_table->record[0], - key_buf, key_len); + ret= event_table->file->ha_index_next_same(event_table->record[0], + key_buf, key_len); } while (ret == 0); } DBUG_PRINT("info", ("Scan finished. ret=%d", ret)); @@ -883,8 +884,9 @@ Event_db_repository::find_named_event(LE key_copy(key, table->record[0], table->key_info, table->key_info->key_length); - if (table->file->index_read_idx_map(table->record[0], 0, key, HA_WHOLE_KEY, - HA_READ_KEY_EXACT)) + if (table->file->ha_index_read_idx_map(table->record[0], 0, key, + HA_WHOLE_KEY, + HA_READ_KEY_EXACT)) { DBUG_PRINT("info", ("Row not found")); DBUG_RETURN(TRUE); === modified file 'sql/filesort.cc' --- a/sql/filesort.cc 2009-09-03 14:05:38 +0000 +++ b/sql/filesort.cc 2009-10-19 17:14:48 +0000 @@ -577,11 +577,11 @@ static ha_rows find_all_keys(SORTPARAM * error= my_errno ? my_errno : -1; /* Abort */ break; } - error=file->rnd_pos(sort_form->record[0],next_pos); + error=file->ha_rnd_pos(sort_form->record[0],next_pos); } else { - error=file->rnd_next(sort_form->record[0]); + error=file->ha_rnd_next(sort_form->record[0]); if (!flag) { my_store_ptr(ref_pos,ref_length,record); // Position to row === modified file 'sql/ha_partition.cc' --- a/sql/ha_partition.cc 2009-09-07 20:50:10 +0000 +++ b/sql/ha_partition.cc 2009-10-19 17:14:48 +0000 @@ -1636,7 +1636,7 @@ int ha_partition::copy_partitions(ulongl goto error; while (TRUE) { - if ((result= file->rnd_next(m_rec0))) + if ((result= file->ha_rnd_next(m_rec0))) { if (result == HA_ERR_RECORD_DELETED) continue; //Probably MyISAM @@ -3495,7 +3495,7 @@ int ha_partition::rnd_next(uchar *buf) while (TRUE) { - result= file->rnd_next(buf); + result= file->ha_rnd_next(buf); if (!result) { m_last_part= part_id; @@ -4345,8 +4345,8 @@ int ha_partition::handle_unordered_next( } else if (is_next_same) { - if (!(error= file->index_next_same(buf, m_start_key.key, - m_start_key.length))) + if (!(error= file->ha_index_next_same(buf, m_start_key.key, + m_start_key.length))) { m_last_part= m_part_spec.start_part; DBUG_RETURN(0); @@ -4354,7 +4354,7 @@ int ha_partition::handle_unordered_next( } else { - if (!(error= file->index_next(buf))) + if (!(error= file->ha_index_next(buf))) { m_last_part= m_part_spec.start_part; DBUG_RETURN(0); // Row was in range @@ -4409,24 +4409,26 @@ int ha_partition::handle_unordered_scan_ break; case partition_index_read: DBUG_PRINT("info", ("index_read on partition %d", i)); - error= file->index_read_map(buf, m_start_key.key, - m_start_key.keypart_map, - m_start_key.flag); + error= file->ha_index_read_map(buf, m_start_key.key, + m_start_key.keypart_map, + m_start_key.flag); break; case partition_index_first: DBUG_PRINT("info", ("index_first on partition %d", i)); - /* MyISAM engine can fail if we call index_first() when indexes disabled */ - /* that happens if the table is empty. */ - /* Here we use file->stats.records instead of file->records() because */ - /* file->records() is supposed to return an EXACT count, and it can be */ - /* possibly slow. We don't need an exact number, an approximate one- from*/ - /* the last ::info() call - is sufficient. */ + /* + MyISAM engine can fail if we call index_first() when indexes disabled + that happens if the table is empty. + Here we use file->stats.records instead of file->records() because + file->records() is supposed to return an EXACT count, and it can be + possibly slow. We don't need an exact number, an approximate one- from + the last ::info() call - is sufficient. + */ if (file->stats.records == 0) { error= HA_ERR_END_OF_FILE; break; } - error= file->index_first(buf); + error= file->ha_index_first(buf); break; case partition_index_first_unordered: /* @@ -4507,45 +4509,49 @@ int ha_partition::handle_ordered_index_s switch (m_index_scan_type) { case partition_index_read: - error= file->index_read_map(rec_buf_ptr, - m_start_key.key, - m_start_key.keypart_map, - m_start_key.flag); + error= file->ha_index_read_map(rec_buf_ptr, + m_start_key.key, + m_start_key.keypart_map, + m_start_key.flag); break; case partition_index_first: - /* MyISAM engine can fail if we call index_first() when indexes disabled */ - /* that happens if the table is empty. */ - /* Here we use file->stats.records instead of file->records() because */ - /* file->records() is supposed to return an EXACT count, and it can be */ - /* possibly slow. We don't need an exact number, an approximate one- from*/ - /* the last ::info() call - is sufficient. */ + /* + MyISAM engine can fail if we call index_first() when indexes disabled + that happens if the table is empty. + Here we use file->stats.records instead of file->records() because + file->records() is supposed to return an EXACT count, and it can be + possibly slow. We don't need an exact number, an approximate one- from + the last ::info() call - is sufficient. + */ if (file->stats.records == 0) { error= HA_ERR_END_OF_FILE; break; } - error= file->index_first(rec_buf_ptr); + error= file->ha_index_first(rec_buf_ptr); reverse_order= FALSE; break; case partition_index_last: - /* MyISAM engine can fail if we call index_last() when indexes disabled */ - /* that happens if the table is empty. */ - /* Here we use file->stats.records instead of file->records() because */ - /* file->records() is supposed to return an EXACT count, and it can be */ - /* possibly slow. We don't need an exact number, an approximate one- from*/ - /* the last ::info() call - is sufficient. */ + /* + MyISAM engine can fail if we call index_last() when indexes disabled + that happens if the table is empty. + Here we use file->stats.records instead of file->records() because + file->records() is supposed to return an EXACT count, and it can be + possibly slow. We don't need an exact number, an approximate one- from + the last ::info() call - is sufficient. + */ if (file->stats.records == 0) { error= HA_ERR_END_OF_FILE; break; } - error= file->index_last(rec_buf_ptr); + error= file->ha_index_last(rec_buf_ptr); reverse_order= TRUE; break; case partition_index_read_last: - error= file->index_read_last_map(rec_buf_ptr, - m_start_key.key, - m_start_key.keypart_map); + error= file->ha_index_read_last_map(rec_buf_ptr, + m_start_key.key, + m_start_key.keypart_map); reverse_order= TRUE; break; case partition_read_range: @@ -4647,10 +4653,10 @@ int ha_partition::handle_ordered_next(uc memcpy(rec_buf(part_id), table->record[0], m_rec_length); } else if (!is_next_same) - error= file->index_next(rec_buf(part_id)); + error= file->ha_index_next(rec_buf(part_id)); else - error= file->index_next_same(rec_buf(part_id), m_start_key.key, - m_start_key.length); + error= file->ha_index_next_same(rec_buf(part_id), m_start_key.key, + m_start_key.length); if (error) { if (error == HA_ERR_END_OF_FILE) @@ -4695,7 +4701,7 @@ int ha_partition::handle_ordered_prev(uc handler *file= m_file[part_id]; DBUG_ENTER("ha_partition::handle_ordered_prev"); - if ((error= file->index_prev(rec_buf(part_id)))) + if ((error= file->ha_index_prev(rec_buf(part_id)))) { if (error == HA_ERR_END_OF_FILE) { === modified file 'sql/handler.cc' --- a/sql/handler.cc 2009-09-09 21:06:57 +0000 +++ b/sql/handler.cc 2009-10-19 17:14:48 +0000 @@ -1236,6 +1236,7 @@ int ha_commit_one_phase(THD *thd, bool a my_error(ER_ERROR_DURING_COMMIT, MYF(0), err); error=1; } + /* Should this be done only if is_real_trans is set ? */ status_var_increment(thd->status_var.ha_commit_count); ha_info_next= ha_info->next(); ha_info->reset(); /* keep it conveniently zero-filled */ @@ -2092,6 +2093,8 @@ int handler::ha_open(TABLE *table_arg, c dup_ref=ref+ALIGN_SIZE(ref_length); cached_table_flags= table_flags(); } + rows_read= rows_changed= 0; + memset(index_rows_read, 0, sizeof(index_rows_read)); DBUG_RETURN(error); } @@ -2513,9 +2516,10 @@ void handler::get_auto_increment(ulonglo key_copy(key, table->record[0], table->key_info + table->s->next_number_index, table->s->next_number_key_offset); - error= index_read_map(table->record[1], key, - make_prev_keypart_map(table->s->next_number_keypart), - HA_READ_PREFIX_LAST); + error= ha_index_read_map(table->record[1], key, + make_prev_keypart_map(table->s-> + next_number_keypart), + HA_READ_PREFIX_LAST); /* MySQL needs to call us for next row: assume we are inserting ("a",null) here, we return 3, and next this statement will want to insert @@ -3549,6 +3553,122 @@ void handler::get_dynamic_partition_info } +/* + Updates the global table stats with the TABLE this handler represents +*/ + +void handler::update_global_table_stats() +{ + TABLE_STATS * table_stats; + + status_var_add(table->in_use->status_var.rows_read, rows_read); + + if (!table->in_use->userstat_running) + { + rows_read= rows_changed= 0; + return; + } + + if (rows_read + rows_changed == 0) + return; // Nothing to update. + + DBUG_ASSERT(table->s && table->s->table_cache_key.str); + + pthread_mutex_lock(&LOCK_global_table_stats); + /* Gets the global table stats, creating one if necessary. */ + if (!(table_stats= (TABLE_STATS*) + hash_search(&global_table_stats, + (uchar*) table->s->table_cache_key.str, + table->s->table_cache_key.length))) + { + if (!(table_stats = ((TABLE_STATS*) + my_malloc(sizeof(TABLE_STATS), + MYF(MY_WME | MY_ZEROFILL))))) + { + /* Out of memory error already given */ + goto end; + } + memcpy(table_stats->table, table->s->table_cache_key.str, + table->s->table_cache_key.length); + table_stats->table_name_length= table->s->table_cache_key.length; + table_stats->engine_type= ht->db_type; + /* No need to set variables to 0, as we use MY_ZEROFILL above */ + + if (my_hash_insert(&global_table_stats, (uchar*) table_stats)) + { + /* Out of memory error is already given */ + my_free(table_stats, 0); + goto end; + } + } + // Updates the global table stats. + table_stats->rows_read+= rows_read; + table_stats->rows_changed+= rows_changed; + table_stats->rows_changed_x_indexes+= (rows_changed * + (table->s->keys ? table->s->keys : + 1)); + rows_read= rows_changed= 0; +end: + pthread_mutex_unlock(&LOCK_global_table_stats); +} + + +/* + Updates the global index stats with this handler's accumulated index reads. +*/ + +void handler::update_global_index_stats() +{ + DBUG_ASSERT(table->s); + + if (!table->in_use->userstat_running) + { + /* Reset all index read values */ + bzero(index_rows_read, sizeof(index_rows_read[0]) * table->s->keys); + return; + } + + for (uint index = 0; index < table->s->keys; index++) + { + if (index_rows_read[index]) + { + INDEX_STATS* index_stats; + uint key_length; + KEY *key_info = &table->key_info[index]; // Rows were read using this + + DBUG_ASSERT(key_info->cache_name); + if (!key_info->cache_name) + continue; + key_length= table->s->table_cache_key.length + key_info->name_length + 1; + pthread_mutex_lock(&LOCK_global_index_stats); + // Gets the global index stats, creating one if necessary. + if (!(index_stats= (INDEX_STATS*) hash_search(&global_index_stats, + key_info->cache_name, + key_length))) + { + if (!(index_stats = ((INDEX_STATS*) + my_malloc(sizeof(INDEX_STATS), + MYF(MY_WME | MY_ZEROFILL))))) + goto end; // Error is already given + + memcpy(index_stats->index, key_info->cache_name, key_length); + index_stats->index_name_length= key_length; + if (my_hash_insert(&global_index_stats, (uchar*) index_stats)) + { + my_free(index_stats, 0); + goto end; + } + } + /* Updates the global index stats. */ + index_stats->rows_read+= index_rows_read[index]; + index_rows_read[index]= 0; +end: + pthread_mutex_unlock(&LOCK_global_index_stats); + } + } +} + + /**************************************************************************** ** Some general functions that isn't in the handler class ****************************************************************************/ @@ -4207,17 +4327,16 @@ int handler::read_range_first(const key_ range_key_part= table->key_info[active_index].key_part; if (!start_key) // Read first record - result= index_first(table->record[0]); + result= ha_index_first(table->record[0]); else - result= index_read_map(table->record[0], - start_key->key, - start_key->keypart_map, - start_key->flag); + result= ha_index_read_map(table->record[0], + start_key->key, + start_key->keypart_map, + start_key->flag); if (result) DBUG_RETURN((result == HA_ERR_KEY_NOT_FOUND) ? HA_ERR_END_OF_FILE : result); - DBUG_RETURN (compare_key(end_range) <= 0 ? 0 : HA_ERR_END_OF_FILE); } @@ -4243,11 +4362,11 @@ int handler::read_range_next() if (eq_range) { /* We trust that index_next_same always gives a row in range */ - DBUG_RETURN(index_next_same(table->record[0], - end_range->key, - end_range->length)); + DBUG_RETURN(ha_index_next_same(table->record[0], + end_range->key, + end_range->length)); } - result= index_next(table->record[0]); + result= ha_index_next(table->record[0]); if (result) DBUG_RETURN(result); DBUG_RETURN(compare_key(end_range) <= 0 ? 0 : HA_ERR_END_OF_FILE); @@ -4629,6 +4748,7 @@ int handler::ha_write_row(uchar *buf) if (unlikely(error= write_row(buf))) DBUG_RETURN(error); + rows_changed++; if (unlikely(error= binlog_log_row(table, 0, buf, log_func))) DBUG_RETURN(error); /* purecov: inspected */ DBUG_RETURN(0); @@ -4650,6 +4770,7 @@ int handler::ha_update_row(const uchar * if (unlikely(error= update_row(old_data, new_data))) return error; + rows_changed++; if (unlikely(error= binlog_log_row(table, old_data, new_data, log_func))) return error; return 0; @@ -4664,6 +4785,7 @@ int handler::ha_delete_row(const uchar * if (unlikely(error= delete_row(buf))) return error; + rows_changed++; if (unlikely(error= binlog_log_row(table, buf, 0, log_func))) return error; return 0; === modified file 'sql/handler.h' --- a/sql/handler.h 2009-09-07 20:50:10 +0000 +++ b/sql/handler.h 2009-10-19 17:14:48 +0000 @@ -30,6 +30,10 @@ #define USING_TRANSACTIONS +#if MAX_KEY > 128 +#error MAX_KEY is too large. Values up to 128 are supported. +#endif + // the following is for checking tables #define HA_ADMIN_ALREADY_DONE 1 @@ -601,8 +605,9 @@ struct handlerton SHOW_COMP_OPTION state; /* - Historical number used for frm file to determine the correct storage engine. - This is going away and new engines will just use "name" for this. + Historical number used for frm file to determine the correct + storage engine. This is going away and new engines will just use + "name" for this. */ enum legacy_db_type db_type; /* @@ -1138,6 +1143,12 @@ public: Interval returned by get_auto_increment() and being consumed by the inserter. */ + /* Statistics variables */ + ulonglong rows_read; + ulonglong rows_changed; + /* One bigger than needed to avoid to test if key == MAX_KEY */ + ulonglong index_rows_read[MAX_KEY+1]; + Discrete_interval auto_inc_interval_for_cur_row; /** Number of reserved auto-increment intervals. Serves as a heuristic @@ -1156,7 +1167,10 @@ public: locked(FALSE), implicit_emptied(0), pushed_cond(0), next_insert_id(0), insert_id_for_cur_row(0), auto_inc_intervals_count(0) - {} + { + reset_statistics(); + } + virtual ~handler(void) { DBUG_ASSERT(locked == FALSE); @@ -1278,10 +1292,16 @@ public: virtual void print_error(int error, myf errflag); virtual bool get_error_message(int error, String *buf); uint get_dup_key(int error); + void reset_statistics() + { + rows_read= rows_changed= 0; + bzero(index_rows_read, sizeof(index_rows_read)); + } virtual void change_table_ptr(TABLE *table_arg, TABLE_SHARE *share) { table= table_arg; table_share= share; + reset_statistics(); } virtual double scan_time() { return ulonglong2double(stats.data_file_length) / IO_SIZE + 2; } @@ -1390,22 +1410,23 @@ public: } /** @brief - Positions an index cursor to the index specified in the handle. Fetches the - row if available. If the key value is null, begin at the first key of the - index. + Positions an index cursor to the index specified in the + handle. Fetches the row if available. If the key value is null, + begin at the first key of the index. */ +protected: virtual int index_read_map(uchar * buf, const uchar * key, key_part_map keypart_map, enum ha_rkey_function find_flag) { uint key_len= calculate_key_len(table, active_index, key, keypart_map); - return index_read(buf, key, key_len, find_flag); + return index_read(buf, key, key_len, find_flag); } /** @brief - Positions an index cursor to the index specified in the handle. Fetches the - row if available. If the key value is null, begin at the first key of the - index. + Positions an index cursor to the index specified in the + handle. Fetches the row if available. If the key value is null, + begin at the first key of the index. */ virtual int index_read_idx_map(uchar * buf, uint index, const uchar * key, key_part_map keypart_map, @@ -1430,6 +1451,79 @@ public: uint key_len= calculate_key_len(table, active_index, key, keypart_map); return index_read_last(buf, key, key_len); } + inline void update_index_statistics() + { + index_rows_read[active_index]++; + rows_read++; + } +public: + + /* Similar functions like the above, but does statistics counting */ + inline int ha_index_read_map(uchar * buf, const uchar * key, + key_part_map keypart_map, + enum ha_rkey_function find_flag) + { + int error= index_read_map(buf, key, keypart_map, find_flag); + if (!error) + update_index_statistics(); + return error; + } + inline int ha_index_read_idx_map(uchar * buf, uint index, const uchar * key, + key_part_map keypart_map, + enum ha_rkey_function find_flag) + { + int error= index_read_idx_map(buf, index, key, keypart_map, find_flag); + if (!error) + { + rows_read++; + index_rows_read[index]++; + } + return error; + } + inline int ha_index_next(uchar * buf) + { + int error= index_next(buf); + if (!error) + update_index_statistics(); + return error; + } + inline int ha_index_prev(uchar * buf) + { + int error= index_prev(buf); + if (!error) + update_index_statistics(); + return error; + } + inline int ha_index_first(uchar * buf) + { + int error= index_first(buf); + if (!error) + update_index_statistics(); + return error; + } + inline int ha_index_last(uchar * buf) + { + int error= index_last(buf); + if (!error) + update_index_statistics(); + return error; + } + inline int ha_index_next_same(uchar *buf, const uchar *key, uint keylen) + { + int error= index_next_same(buf, key, keylen); + if (!error) + update_index_statistics(); + return error; + } + inline int ha_index_read_last_map(uchar * buf, const uchar * key, + key_part_map keypart_map) + { + int error= index_read_last_map(buf, key, keypart_map); + if (!error) + update_index_statistics(); + return error; + } + virtual int read_multi_range_first(KEY_MULTI_RANGE **found_range_p, KEY_MULTI_RANGE *ranges, uint range_count, bool sorted, HANDLER_BUFFER *buffer); @@ -1443,6 +1537,7 @@ public: void ft_end() { ft_handler=NULL; } virtual FT_INFO *ft_init_ext(uint flags, uint inx,String *key) { return NULL; } +private: virtual int ft_read(uchar *buf) { return HA_ERR_WRONG_COMMAND; } virtual int rnd_next(uchar *buf)=0; virtual int rnd_pos(uchar * buf, uchar *pos)=0; @@ -1453,11 +1548,50 @@ public: handlers for random position. */ virtual int rnd_pos_by_record(uchar *record) - { - position(record); - return rnd_pos(record, ref); - } + { + position(record); + return rnd_pos(record, ref); + } virtual int read_first_row(uchar *buf, uint primary_key); +public: + + /* Same as above, but with statistics */ + inline int ha_ft_read(uchar *buf) + { + int error= ft_read(buf); + if (!error) + rows_read++; + return error; + } + inline int ha_rnd_next(uchar *buf) + { + int error= rnd_next(buf); + if (!error) + rows_read++; + return error; + } + inline int ha_rnd_pos(uchar *buf, uchar *pos) + { + int error= rnd_pos(buf, pos); + if (!error) + rows_read++; + return error; + } + inline int ha_rnd_pos_by_record(uchar *buf) + { + int error= rnd_pos_by_record(buf); + if (!error) + rows_read++; + return error; + } + inline int ha_read_first_row(uchar *buf, uint primary_key) + { + int error= read_first_row(buf, primary_key); + if (!error) + rows_read++; + return error; + } + /** The following 3 function is only needed for tables that may be internal temporary tables during joins. @@ -1626,6 +1760,9 @@ public: virtual bool is_crashed() const { return 0; } virtual bool auto_repair() const { return 0; } + void update_global_table_stats(); + void update_global_index_stats(); + #define CHF_CREATE_FLAG 0 #define CHF_DELETE_FLAG 1 #define CHF_RENAME_FLAG 2 @@ -1944,6 +2081,7 @@ private: { return HA_ERR_WRONG_COMMAND; } virtual int rename_partitions(const char *path) { return HA_ERR_WRONG_COMMAND; } + friend class ha_partition; }; === modified file 'sql/item_subselect.cc' --- a/sql/item_subselect.cc 2009-09-15 10:46:35 +0000 +++ b/sql/item_subselect.cc 2009-10-19 17:14:48 +0000 @@ -2048,7 +2048,7 @@ int subselect_uniquesubquery_engine::sca table->null_row= 0; for (;;) { - error=table->file->rnd_next(table->record[0]); + error=table->file->ha_rnd_next(table->record[0]); if (error && error != HA_ERR_END_OF_FILE) { error= report_error(table, error); @@ -2222,10 +2222,11 @@ int subselect_uniquesubquery_engine::exe if (!table->file->inited) table->file->ha_index_init(tab->ref.key, 0); - error= table->file->index_read_map(table->record[0], - tab->ref.key_buff, - make_prev_keypart_map(tab->ref.key_parts), - HA_READ_KEY_EXACT); + error= table->file->ha_index_read_map(table->record[0], + tab->ref.key_buff, + make_prev_keypart_map(tab-> + ref.key_parts), + HA_READ_KEY_EXACT); if (error && error != HA_ERR_KEY_NOT_FOUND && error != HA_ERR_END_OF_FILE) error= report_error(table, error); @@ -2343,10 +2344,11 @@ int subselect_indexsubquery_engine::exec if (!table->file->inited) table->file->ha_index_init(tab->ref.key, 1); - error= table->file->index_read_map(table->record[0], - tab->ref.key_buff, - make_prev_keypart_map(tab->ref.key_parts), - HA_READ_KEY_EXACT); + error= table->file->ha_index_read_map(table->record[0], + tab->ref.key_buff, + make_prev_keypart_map(tab-> + ref.key_parts), + HA_READ_KEY_EXACT); if (error && error != HA_ERR_KEY_NOT_FOUND && error != HA_ERR_END_OF_FILE) error= report_error(table, error); @@ -2367,9 +2369,9 @@ int subselect_indexsubquery_engine::exec ((Item_in_subselect *) item)->value= 1; break; } - error= table->file->index_next_same(table->record[0], - tab->ref.key_buff, - tab->ref.key_length); + error= table->file->ha_index_next_same(table->record[0], + tab->ref.key_buff, + tab->ref.key_length); if (error && error != HA_ERR_END_OF_FILE) { error= report_error(table, error); === modified file 'sql/lex.h' --- a/sql/lex.h 2009-09-07 20:50:10 +0000 +++ b/sql/lex.h 2009-10-19 17:14:48 +0000 @@ -106,6 +106,7 @@ static SYMBOL symbols[] = { { "CHECKSUM", SYM(CHECKSUM_SYM)}, { "CIPHER", SYM(CIPHER_SYM)}, { "CLIENT", SYM(CLIENT_SYM)}, + { "CLIENT_STATISTICS", SYM(CLIENT_STATS_SYM)}, { "CLOSE", SYM(CLOSE_SYM)}, { "COALESCE", SYM(COALESCE)}, { "CODE", SYM(CODE_SYM)}, @@ -245,6 +246,7 @@ static SYMBOL symbols[] = { { "IN", SYM(IN_SYM)}, { "INDEX", SYM(INDEX_SYM)}, { "INDEXES", SYM(INDEXES)}, + { "INDEX_STATISTICS", SYM(INDEX_STATS_SYM)}, { "INFILE", SYM(INFILE)}, { "INITIAL_SIZE", SYM(INITIAL_SIZE_SYM)}, { "INNER", SYM(INNER_SYM)}, @@ -478,6 +480,7 @@ static SYMBOL symbols[] = { { "SIGNED", SYM(SIGNED_SYM)}, { "SIMPLE", SYM(SIMPLE_SYM)}, { "SLAVE", SYM(SLAVE)}, + { "SLOW", SYM(SLOW_SYM)}, { "SNAPSHOT", SYM(SNAPSHOT_SYM)}, { "SMALLINT", SYM(SMALLINT)}, { "SOCKET", SYM(SOCKET_SYM)}, @@ -526,6 +529,7 @@ static SYMBOL symbols[] = { { "TABLE", SYM(TABLE_SYM)}, { "TABLES", SYM(TABLES)}, { "TABLESPACE", SYM(TABLESPACE)}, + { "TABLE_STATISTICS", SYM(TABLE_STATS_SYM)}, { "TABLE_CHECKSUM", SYM(TABLE_CHECKSUM_SYM)}, { "TEMPORARY", SYM(TEMPORARY)}, { "TEMPTABLE", SYM(TEMPTABLE_SYM)}, @@ -569,6 +573,7 @@ static SYMBOL symbols[] = { { "USE", SYM(USE_SYM)}, { "USER", SYM(USER)}, { "USER_RESOURCES", SYM(RESOURCES)}, + { "USER_STATISTICS", SYM(USER_STATS_SYM)}, { "USE_FRM", SYM(USE_FRM)}, { "USING", SYM(USING)}, { "UTC_DATE", SYM(UTC_DATE_SYM)}, === modified file 'sql/log.cc' --- a/sql/log.cc 2009-09-15 10:46:35 +0000 +++ b/sql/log.cc 2009-10-19 17:14:48 +0000 @@ -821,6 +821,13 @@ void Log_to_file_event_handler::flush() mysql_slow_log.reopen_file(); } +void Log_to_file_event_handler::flush_slow_log() +{ + /* reopen slow log file */ + if (opt_slow_log) + mysql_slow_log.reopen_file(); +} + /* Log error with all enabled log event handlers @@ -916,8 +923,6 @@ void LOGGER::init_log_tables() bool LOGGER::flush_logs(THD *thd) { - int rc= 0; - /* Now we lock logger, as nobody should be able to use logging routines while log tables are closed @@ -929,7 +934,24 @@ bool LOGGER::flush_logs(THD *thd) /* end of log flush */ logger.unlock(); - return rc; + return 0; +} + + +bool LOGGER::flush_slow_log(THD *thd) +{ + /* + Now we lock logger, as nobody should be able to use logging routines while + log tables are closed + */ + logger.lock_exclusive(); + + /* reopen log files */ + file_log_handler->flush_slow_log(); + + /* end of log flush */ + logger.unlock(); + return 0; } @@ -4070,6 +4092,7 @@ bool MYSQL_BIN_LOG::write(Log_event *eve if (likely(is_open())) { IO_CACHE *file= &log_file; + my_off_t my_org_b_tell; #ifdef HAVE_REPLICATION /* In the future we need to add to the following if tests like @@ -4077,7 +4100,7 @@ bool MYSQL_BIN_LOG::write(Log_event *eve binlog_[wild_]{do|ignore}_table?" (WL#1049)" */ const char *local_db= event_info->get_db(); - if ((thd && !(thd->options & OPTION_BIN_LOG)) || + if ((!(thd->options & OPTION_BIN_LOG)) || (!binlog_filter->db_ok(local_db))) { VOID(pthread_mutex_unlock(&LOCK_log)); @@ -4085,6 +4108,8 @@ bool MYSQL_BIN_LOG::write(Log_event *eve } #endif /* HAVE_REPLICATION */ + my_org_b_tell= my_b_tell(file); + #if defined(USING_TRANSACTIONS) /* Should we write to the binlog cache or to the binlog on disk? @@ -4095,7 +4120,7 @@ bool MYSQL_BIN_LOG::write(Log_event *eve trans/non-trans table types the best possible in binlogging) - or if the event asks for it (cache_stmt == TRUE). */ - if (opt_using_transactions && thd) + if (opt_using_transactions) { if (thd->binlog_setup_trx_data()) goto err; @@ -4136,7 +4161,6 @@ bool MYSQL_BIN_LOG::write(Log_event *eve If row-based binlogging, Insert_id, Rand and other kind of "setting context" events are not needed. */ - if (thd) { if (!thd->current_stmt_binlog_row_based) { @@ -4183,16 +4207,16 @@ bool MYSQL_BIN_LOG::write(Log_event *eve } } - /* - Write the SQL command - */ - + /* Write the SQL command */ if (event_info->write(file) || DBUG_EVALUATE_IF("injecting_fault_writing", 1, 0)) goto err; if (file == &log_file) // we are writing to the real log (disk) { + ulonglong data_written= (my_b_tell(file) - my_org_b_tell); + status_var_add(thd->status_var.binlog_bytes_written, data_written); + if (flush_and_sync()) goto err; signal_update(); @@ -4318,6 +4342,7 @@ uint MYSQL_BIN_LOG::next_file_id() SYNOPSIS write_cache() + thd Current_thread cache Cache to write to the binary log lock_log True if the LOCK_log mutex should be aquired, false otherwise sync_log True if the log should be flushed and sync:ed @@ -4327,7 +4352,8 @@ uint MYSQL_BIN_LOG::next_file_id() be reset as a READ_CACHE to be able to read the contents from it. */ -int MYSQL_BIN_LOG::write_cache(IO_CACHE *cache, bool lock_log, bool sync_log) +int MYSQL_BIN_LOG::write_cache(THD *thd, IO_CACHE *cache, bool lock_log, + bool sync_log) { Mutex_sentry sentry(lock_log ? &LOCK_log : NULL); @@ -4375,6 +4401,7 @@ int MYSQL_BIN_LOG::write_cache(IO_CACHE /* write the first half of the split header */ if (my_b_write(&log_file, header, carry)) return ER_ERROR_ON_WRITE; + status_var_add(thd->status_var.binlog_bytes_written, carry); /* copy fixed second half of header to cache so the correct @@ -4443,6 +4470,8 @@ int MYSQL_BIN_LOG::write_cache(IO_CACHE /* Write data to the binary log file */ if (my_b_write(&log_file, cache->read_pos, length)) return ER_ERROR_ON_WRITE; + status_var_add(thd->status_var.binlog_bytes_written, length); + cache->read_pos=cache->read_end; // Mark buffer used up } while ((length= my_b_fill(cache))); @@ -4494,6 +4523,8 @@ bool MYSQL_BIN_LOG::write_incident(THD * if (lock) pthread_mutex_lock(&LOCK_log); ev.write(&log_file); + status_var_add(thd->status_var.binlog_bytes_written, ev.data_written); + if (lock) { if (!error && !(error= flush_and_sync())) @@ -4565,21 +4596,28 @@ bool MYSQL_BIN_LOG::write(THD *thd, IO_C */ if (qinfo.write(&log_file)) goto err; + status_var_add(thd->status_var.binlog_bytes_written, qinfo.data_written); DBUG_EXECUTE_IF("crash_before_writing_xid", { - if ((write_error= write_cache(cache, false, true))) + if ((write_error= write_cache(thd, cache, FALSE, + TRUE))) DBUG_PRINT("info", ("error writing binlog cache: %d", write_error)); DBUG_PRINT("info", ("crashing before writing xid")); abort(); }); - if ((write_error= write_cache(cache, false, false))) + if ((write_error= write_cache(thd, cache, FALSE, FALSE))) goto err; - if (commit_event && commit_event->write(&log_file)) - goto err; + if (commit_event) + { + if (commit_event->write(&log_file)) + goto err; + status_var_add(thd->status_var.binlog_bytes_written, + commit_event->data_written); + } if (incident && write_incident(thd, FALSE)) goto err; === modified file 'sql/log.h' --- a/sql/log.h 2009-06-18 13:52:46 +0000 +++ b/sql/log.h 2009-10-19 17:14:48 +0000 @@ -359,7 +359,8 @@ public: bool write(THD *thd, IO_CACHE *cache, Log_event *commit_event, bool incident); bool write_incident(THD *thd, bool lock); - int write_cache(IO_CACHE *cache, bool lock_log, bool flush_and_sync); + int write_cache(THD *thd, IO_CACHE *cache, bool lock_log, + bool flush_and_sync); void set_write_error(THD *thd); bool check_write_error(THD *thd); @@ -487,6 +488,7 @@ public: const char *sql_text, uint sql_text_len, CHARSET_INFO *client_cs); void flush(); + void flush_slow_log(); void init_pthread_objects(); MYSQL_QUERY_LOG *get_mysql_slow_log() { return &mysql_slow_log; } MYSQL_QUERY_LOG *get_mysql_log() { return &mysql_log; } @@ -531,6 +533,7 @@ public: void init_base(); void init_log_tables(); bool flush_logs(THD *thd); + bool flush_slow_log(THD *thd); /* Perform basic logger cleanup. this will leave e.g. error log open. */ void cleanup_base(); /* Free memory. Nothing could be logged after this function is called */ === modified file 'sql/log_event.cc' --- a/sql/log_event.cc 2009-09-07 20:50:10 +0000 +++ b/sql/log_event.cc 2009-10-19 17:14:48 +0000 @@ -4465,7 +4465,7 @@ int Load_log_event::do_apply_event(NET* as the present method does not call mysql_parse(). */ lex_start(thd); - mysql_reset_thd_for_next_command(thd); + mysql_reset_thd_for_next_command(thd, 0); if (!use_rli_only_for_errors) { @@ -6262,7 +6262,7 @@ int Append_block_log_event::do_apply_eve as the present method does not call mysql_parse(). */ lex_start(thd); - mysql_reset_thd_for_next_command(thd); + mysql_reset_thd_for_next_command(thd, 0); my_delete(fname, MYF(0)); // old copy may exist already if ((fd= my_create(fname, CREATE_MODE, O_WRONLY | O_BINARY | O_EXCL | O_NOFOLLOW, @@ -7202,7 +7202,7 @@ int Rows_log_event::do_apply_event(Relay we need to do any changes to that value after this function. */ lex_start(thd); - mysql_reset_thd_for_next_command(thd); + mysql_reset_thd_for_next_command(thd, 0); /* The current statement is just about to begin and has not yet modified anything. Note, all.modified is reset @@ -8465,7 +8465,7 @@ Rows_log_event::write_row(const Relay_lo if (table->file->ha_table_flags() & HA_DUPLICATE_POS) { DBUG_PRINT("info",("Locating offending record using rnd_pos()")); - error= table->file->rnd_pos(table->record[1], table->file->dup_ref); + error= table->file->ha_rnd_pos(table->record[1], table->file->dup_ref); if (error) { DBUG_PRINT("info",("rnd_pos() returns error %d",error)); @@ -8497,10 +8497,10 @@ Rows_log_event::write_row(const Relay_lo key_copy((uchar*)key.get(), table->record[0], table->key_info + keynum, 0); - error= table->file->index_read_idx_map(table->record[1], keynum, - (const uchar*)key.get(), - HA_WHOLE_KEY, - HA_READ_KEY_EXACT); + error= table->file->ha_index_read_idx_map(table->record[1], keynum, + (const uchar*)key.get(), + HA_WHOLE_KEY, + HA_READ_KEY_EXACT); if (error) { DBUG_PRINT("info",("index_read_idx() returns %s", HA_ERR(error))); @@ -8768,13 +8768,14 @@ int Rows_log_event::find_row(const Relay length. Something along these lines should work: ADD>>> store_record(table,record[1]); - int error= table->file->rnd_pos(table->record[0], table->file->ref); + int error= table->file->ha_rnd_pos(table->record[0], + table->file->ref); ADD>>> DBUG_ASSERT(memcmp(table->record[1], table->record[0], table->s->reclength) == 0); */ DBUG_PRINT("info",("locating record using primary key (position)")); - int error= table->file->rnd_pos_by_record(table->record[0]); + int error= table->file->ha_rnd_pos_by_record(table->record[0]); if (error) { DBUG_PRINT("info",("rnd_pos returns error %d",error)); @@ -8834,9 +8835,9 @@ int Rows_log_event::find_row(const Relay table->record[0][table->s->null_bytes - 1]|= 256U - (1U << table->s->last_null_bit_pos); - if ((error= table->file->index_read_map(table->record[0], m_key, - HA_WHOLE_KEY, - HA_READ_KEY_EXACT))) + if ((error= table->file->ha_index_read_map(table->record[0], m_key, + HA_WHOLE_KEY, + HA_READ_KEY_EXACT))) { DBUG_PRINT("info",("no record matching the key found in the table")); if (error == HA_ERR_RECORD_DELETED) @@ -8898,7 +8899,7 @@ int Rows_log_event::find_row(const Relay 256U - (1U << table->s->last_null_bit_pos); } - while ((error= table->file->index_next(table->record[0]))) + while ((error= table->file->ha_index_next(table->record[0]))) { /* We just skip records that has already been deleted */ if (error == HA_ERR_RECORD_DELETED) @@ -8934,7 +8935,7 @@ int Rows_log_event::find_row(const Relay do { restart_rnd_next: - error= table->file->rnd_next(table->record[0]); + error= table->file->ha_rnd_next(table->record[0]); DBUG_PRINT("info", ("error: %s", HA_ERR(error))); switch (error) { === modified file 'sql/log_event_old.cc' --- a/sql/log_event_old.cc 2009-05-19 09:28:05 +0000 +++ b/sql/log_event_old.cc 2009-10-19 17:14:48 +0000 @@ -63,7 +63,7 @@ Old_rows_log_event::do_apply_event(Old_r we need to do any changes to that value after this function. */ lex_start(thd); - mysql_reset_thd_for_next_command(thd); + mysql_reset_thd_for_next_command(thd, 0); /* Check if the slave is set to use SBR. If so, it should switch @@ -553,7 +553,7 @@ replace_record(THD *thd, TABLE *table, */ if (table->file->ha_table_flags() & HA_DUPLICATE_POS) { - error= table->file->rnd_pos(table->record[1], table->file->dup_ref); + error= table->file->ha_rnd_pos(table->record[1], table->file->dup_ref); if (error) { DBUG_PRINT("info",("rnd_pos() returns error %d",error)); @@ -579,10 +579,10 @@ replace_record(THD *thd, TABLE *table, key_copy((uchar*)key.get(), table->record[0], table->key_info + keynum, 0); - error= table->file->index_read_idx_map(table->record[1], keynum, - (const uchar*)key.get(), - HA_WHOLE_KEY, - HA_READ_KEY_EXACT); + error= table->file->ha_index_read_idx_map(table->record[1], keynum, + (const uchar*)key.get(), + HA_WHOLE_KEY, + HA_READ_KEY_EXACT); if (error) { DBUG_PRINT("info", ("index_read_idx() returns error %d", error)); @@ -694,13 +694,13 @@ static int find_and_fetch_row(TABLE *tab length. Something along these lines should work: ADD>>> store_record(table,record[1]); - int error= table->file->rnd_pos(table->record[0], table->file->ref); + int error= table->file->ha_rnd_pos(table->record[0], table->file->ref); ADD>>> DBUG_ASSERT(memcmp(table->record[1], table->record[0], table->s->reclength) == 0); */ table->file->position(table->record[0]); - int error= table->file->rnd_pos(table->record[0], table->file->ref); + int error= table->file->ha_rnd_pos(table->record[0], table->file->ref); /* rnd_pos() returns the record in table->record[0], so we have to move it to table->record[1]. @@ -738,8 +738,9 @@ static int find_and_fetch_row(TABLE *tab my_ptrdiff_t const pos= table->s->null_bytes > 0 ? table->s->null_bytes - 1 : 0; table->record[1][pos]= 0xFF; - if ((error= table->file->index_read_map(table->record[1], key, HA_WHOLE_KEY, - HA_READ_KEY_EXACT))) + if ((error= table->file->ha_index_read_map(table->record[1], key, + HA_WHOLE_KEY, + HA_READ_KEY_EXACT))) { table->file->print_error(error, MYF(0)); table->file->ha_index_end(); @@ -793,7 +794,7 @@ static int find_and_fetch_row(TABLE *tab 256U - (1U << table->s->last_null_bit_pos); } - while ((error= table->file->index_next(table->record[1]))) + while ((error= table->file->ha_index_next(table->record[1]))) { /* We just skip records that has already been deleted */ if (error == HA_ERR_RECORD_DELETED) @@ -822,7 +823,7 @@ static int find_and_fetch_row(TABLE *tab do { restart_rnd_next: - error= table->file->rnd_next(table->record[1]); + error= table->file->ha_rnd_next(table->record[1]); DBUG_DUMP("record[0]", table->record[0], table->s->reclength); DBUG_DUMP("record[1]", table->record[1], table->s->reclength); @@ -2115,7 +2116,7 @@ Old_rows_log_event::write_row(const Rela if (table->file->ha_table_flags() & HA_DUPLICATE_POS) { DBUG_PRINT("info",("Locating offending record using rnd_pos()")); - error= table->file->rnd_pos(table->record[1], table->file->dup_ref); + error= table->file->ha_rnd_pos(table->record[1], table->file->dup_ref); if (error) { DBUG_PRINT("info",("rnd_pos() returns error %d",error)); @@ -2147,10 +2148,10 @@ Old_rows_log_event::write_row(const Rela key_copy((uchar*)key.get(), table->record[0], table->key_info + keynum, 0); - error= table->file->index_read_idx_map(table->record[1], keynum, - (const uchar*)key.get(), - HA_WHOLE_KEY, - HA_READ_KEY_EXACT); + error= table->file->ha_index_read_idx_map(table->record[1], keynum, + (const uchar*)key.get(), + HA_WHOLE_KEY, + HA_READ_KEY_EXACT); if (error) { DBUG_PRINT("info",("index_read_idx() returns error %d", error)); @@ -2301,13 +2302,13 @@ int Old_rows_log_event::find_row(const R length. Something along these lines should work: ADD>>> store_record(table,record[1]); - int error= table->file->rnd_pos(table->record[0], table->file->ref); + int error= table->file->ha_rnd_pos(table->record[0], table->file->ref); ADD>>> DBUG_ASSERT(memcmp(table->record[1], table->record[0], table->s->reclength) == 0); */ DBUG_PRINT("info",("locating record using primary key (position)")); - int error= table->file->rnd_pos_by_record(table->record[0]); + int error= table->file->ha_rnd_pos_by_record(table->record[0]); if (error) { DBUG_PRINT("info",("rnd_pos returns error %d",error)); @@ -2367,9 +2368,9 @@ int Old_rows_log_event::find_row(const R table->s->null_bytes > 0 ? table->s->null_bytes - 1 : 0; table->record[0][pos]= 0xFF; - if ((error= table->file->index_read_map(table->record[0], m_key, - HA_WHOLE_KEY, - HA_READ_KEY_EXACT))) + if ((error= table->file->ha_index_read_map(table->record[0], m_key, + HA_WHOLE_KEY, + HA_READ_KEY_EXACT))) { DBUG_PRINT("info",("no record matching the key found in the table")); if (error == HA_ERR_RECORD_DELETED) @@ -2431,7 +2432,7 @@ int Old_rows_log_event::find_row(const R 256U - (1U << table->s->last_null_bit_pos); } - while ((error= table->file->index_next(table->record[0]))) + while ((error= table->file->ha_index_next(table->record[0]))) { /* We just skip records that has already been deleted */ if (error == HA_ERR_RECORD_DELETED) @@ -2467,7 +2468,7 @@ int Old_rows_log_event::find_row(const R do { restart_rnd_next: - error= table->file->rnd_next(table->record[0]); + error= table->file->ha_rnd_next(table->record[0]); switch (error) { === modified file 'sql/mysql_priv.h' --- a/sql/mysql_priv.h 2009-10-06 14:53:46 +0000 +++ b/sql/mysql_priv.h 2009-10-19 17:14:48 +0000 @@ -1063,6 +1063,7 @@ bool setup_connection_thread_globals(THD bool login_connection(THD *thd); void end_connection(THD *thd); void prepare_new_connection_state(THD* thd); +void update_global_user_stats(THD* thd, bool create_user, time_t now); int mysql_create_db(THD *thd, char *db, HA_CREATE_INFO *create, bool silent); bool mysql_alter_db(THD *thd, const char *db, HA_CREATE_INFO *create); @@ -1099,14 +1100,22 @@ bool is_update_query(enum enum_sql_comma bool is_log_table_write_query(enum enum_sql_command command); bool alloc_query(THD *thd, const char *packet, uint packet_length); void mysql_init_select(LEX *lex); -void mysql_reset_thd_for_next_command(THD *thd); +void mysql_reset_thd_for_next_command(THD *thd, my_bool calculate_userstat); bool mysql_new_select(LEX *lex, bool move_down); void create_select_for_variable(const char *var_name); void mysql_init_multi_delete(LEX *lex); bool multi_delete_set_locks_and_link_aux_tables(LEX *lex); void init_max_user_conn(void); void init_update_queries(void); +void init_global_user_stats(void); +void init_global_table_stats(void); +void init_global_index_stats(void); +void init_global_client_stats(void); void free_max_user_conn(void); +void free_global_user_stats(void); +void free_global_table_stats(void); +void free_global_index_stats(void); +void free_global_client_stats(void); pthread_handler_t handle_bootstrap(void *arg); int mysql_execute_command(THD *thd); bool do_command(THD *thd); @@ -1967,6 +1976,7 @@ extern ulong max_connect_errors, connect extern ulong extra_max_connections; extern ulong slave_net_timeout, slave_trans_retries; extern uint max_user_connections; +extern ulonglong denied_connections; extern ulong what_to_log,flush_time; extern ulong query_buff_size; extern ulong max_prepared_stmt_count, prepared_stmt_count; @@ -2020,6 +2030,7 @@ extern my_bool opt_safe_show_db, opt_loc extern my_bool opt_slave_compressed_protocol, use_temp_pool; extern ulong slave_exec_mode_options; extern my_bool opt_readonly, lower_case_file_system; +extern my_bool opt_userstat_running; extern my_bool opt_enable_named_pipe, opt_sync_frm, opt_allow_suspicious_udfs; extern my_bool opt_secure_auth; extern char* opt_secure_file_priv; @@ -2060,6 +2071,11 @@ extern pthread_mutex_t LOCK_des_key_file #endif extern pthread_mutex_t LOCK_server_started; extern pthread_cond_t COND_server_started; +extern pthread_mutex_t LOCK_global_user_client_stats; +extern pthread_mutex_t LOCK_global_table_stats; +extern pthread_mutex_t LOCK_global_index_stats; +extern pthread_mutex_t LOCK_stats; + extern int mysqld_server_started; extern rw_lock_t LOCK_grant, LOCK_sys_init_connect, LOCK_sys_init_slave; extern rw_lock_t LOCK_system_variables_hash; @@ -2086,6 +2102,11 @@ extern KNOWN_DATE_TIME_FORMAT known_date extern String null_string; extern HASH open_cache, lock_db_cache; +extern HASH global_user_stats; +extern HASH global_client_stats; +extern HASH global_table_stats; +extern HASH global_index_stats; + extern TABLE *unused_tables; extern const char* any_db; extern struct my_option my_long_options[]; === modified file 'sql/mysqld.cc' --- a/sql/mysqld.cc 2009-10-07 13:07:10 +0000 +++ b/sql/mysqld.cc 2009-10-19 17:14:48 +0000 @@ -416,6 +416,7 @@ static pthread_cond_t COND_thread_cache, bool opt_update_log, opt_bin_log, opt_ignore_builtin_innodb= 0; my_bool opt_log, opt_slow_log; +my_bool opt_userstat_running; ulong log_output_options; my_bool opt_log_queries_not_using_indexes= 0; bool opt_error_log= IF_WIN(1,0); @@ -548,6 +549,7 @@ ulong binlog_cache_use= 0, binlog_cache_ ulong max_connections, max_connect_errors; ulong extra_max_connections; uint max_user_connections= 0; +ulonglong denied_connections; /** Limit of the total number of prepared statements in the server. Is necessary to protect the server against out-of-memory attacks. @@ -649,6 +651,9 @@ pthread_mutex_t LOCK_mysql_create_db, LO LOCK_global_system_variables, LOCK_user_conn, LOCK_slave_list, LOCK_active_mi, LOCK_connection_count, LOCK_uuid_generator; +pthread_mutex_t LOCK_stats, LOCK_global_user_client_stats; +pthread_mutex_t LOCK_global_table_stats, LOCK_global_index_stats; + /** The below lock protects access to two global server variables: max_prepared_stmt_count and prepared_stmt_count. These variables @@ -1342,6 +1347,10 @@ void clean_up(bool print_message) x_free(opt_secure_file_priv); bitmap_free(&temp_pool); free_max_user_conn(); + free_global_user_stats(); + free_global_client_stats(); + free_global_table_stats(); + free_global_index_stats(); #ifdef HAVE_REPLICATION end_slave_list(); #endif @@ -1428,6 +1437,11 @@ static void clean_up_mutexes() (void) pthread_mutex_destroy(&LOCK_bytes_received); (void) pthread_mutex_destroy(&LOCK_user_conn); (void) pthread_mutex_destroy(&LOCK_connection_count); + (void) pthread_mutex_destroy(&LOCK_stats); + (void) pthread_mutex_destroy(&LOCK_global_user_client_stats); + (void) pthread_mutex_destroy(&LOCK_global_table_stats); + (void) pthread_mutex_destroy(&LOCK_global_index_stats); + Events::destroy_mutexes(); #ifdef HAVE_OPENSSL (void) pthread_mutex_destroy(&LOCK_des_key_file); @@ -3203,6 +3217,7 @@ SHOW_VAR com_status_vars[]= { {"show_binlog_events", (char*) offsetof(STATUS_VAR, com_stat[(uint) SQLCOM_SHOW_BINLOG_EVENTS]), SHOW_LONG_STATUS}, {"show_binlogs", (char*) offsetof(STATUS_VAR, com_stat[(uint) SQLCOM_SHOW_BINLOGS]), SHOW_LONG_STATUS}, {"show_charsets", (char*) offsetof(STATUS_VAR, com_stat[(uint) SQLCOM_SHOW_CHARSETS]), SHOW_LONG_STATUS}, + {"show_client_statistics", (char*) offsetof(STATUS_VAR, com_stat[(uint) SQLCOM_SHOW_CLIENT_STATS]), SHOW_LONG_STATUS}, {"show_collations", (char*) offsetof(STATUS_VAR, com_stat[(uint) SQLCOM_SHOW_COLLATIONS]), SHOW_LONG_STATUS}, {"show_column_types", (char*) offsetof(STATUS_VAR, com_stat[(uint) SQLCOM_SHOW_COLUMN_TYPES]), SHOW_LONG_STATUS}, {"show_contributors", (char*) offsetof(STATUS_VAR, com_stat[(uint) SQLCOM_SHOW_CONTRIBUTORS]), SHOW_LONG_STATUS}, @@ -3225,6 +3240,7 @@ SHOW_VAR com_status_vars[]= { {"show_function_status", (char*) offsetof(STATUS_VAR, com_stat[(uint) SQLCOM_SHOW_STATUS_FUNC]), SHOW_LONG_STATUS}, {"show_grants", (char*) offsetof(STATUS_VAR, com_stat[(uint) SQLCOM_SHOW_GRANTS]), SHOW_LONG_STATUS}, {"show_keys", (char*) offsetof(STATUS_VAR, com_stat[(uint) SQLCOM_SHOW_KEYS]), SHOW_LONG_STATUS}, + {"show_index_statistics", (char*) offsetof(STATUS_VAR, com_stat[(uint) SQLCOM_SHOW_INDEX_STATS]), SHOW_LONG_STATUS}, {"show_master_status", (char*) offsetof(STATUS_VAR, com_stat[(uint) SQLCOM_SHOW_MASTER_STAT]), SHOW_LONG_STATUS}, {"show_new_master", (char*) offsetof(STATUS_VAR, com_stat[(uint) SQLCOM_SHOW_NEW_MASTER]), SHOW_LONG_STATUS}, {"show_open_tables", (char*) offsetof(STATUS_VAR, com_stat[(uint) SQLCOM_SHOW_OPEN_TABLES]), SHOW_LONG_STATUS}, @@ -3241,9 +3257,11 @@ SHOW_VAR com_status_vars[]= { {"show_slave_status", (char*) offsetof(STATUS_VAR, com_stat[(uint) SQLCOM_SHOW_SLAVE_STAT]), SHOW_LONG_STATUS}, {"show_status", (char*) offsetof(STATUS_VAR, com_stat[(uint) SQLCOM_SHOW_STATUS]), SHOW_LONG_STATUS}, {"show_storage_engines", (char*) offsetof(STATUS_VAR, com_stat[(uint) SQLCOM_SHOW_STORAGE_ENGINES]), SHOW_LONG_STATUS}, + {"show_table_statistics", (char*) offsetof(STATUS_VAR, com_stat[(uint) SQLCOM_SHOW_TABLE_STATS]), SHOW_LONG_STATUS}, {"show_table_status", (char*) offsetof(STATUS_VAR, com_stat[(uint) SQLCOM_SHOW_TABLE_STATUS]), SHOW_LONG_STATUS}, {"show_tables", (char*) offsetof(STATUS_VAR, com_stat[(uint) SQLCOM_SHOW_TABLES]), SHOW_LONG_STATUS}, {"show_triggers", (char*) offsetof(STATUS_VAR, com_stat[(uint) SQLCOM_SHOW_TRIGGERS]), SHOW_LONG_STATUS}, + {"show_user_statistics", (char*) offsetof(STATUS_VAR, com_stat[(uint) SQLCOM_SHOW_USER_STATS]), SHOW_LONG_STATUS}, {"show_variables", (char*) offsetof(STATUS_VAR, com_stat[(uint) SQLCOM_SHOW_VARIABLES]), SHOW_LONG_STATUS}, {"show_warnings", (char*) offsetof(STATUS_VAR, com_stat[(uint) SQLCOM_SHOW_WARNS]), SHOW_LONG_STATUS}, {"slave_start", (char*) offsetof(STATUS_VAR, com_stat[(uint) SQLCOM_SLAVE_START]), SHOW_LONG_STATUS}, @@ -3642,6 +3660,12 @@ static int init_thread_environment() (void) pthread_mutex_init(&LOCK_prepared_stmt_count, MY_MUTEX_INIT_FAST); (void) pthread_mutex_init(&LOCK_uuid_generator, MY_MUTEX_INIT_FAST); (void) pthread_mutex_init(&LOCK_connection_count, MY_MUTEX_INIT_FAST); + (void) pthread_mutex_init(&LOCK_stats, MY_MUTEX_INIT_FAST); + (void) pthread_mutex_init(&LOCK_global_user_client_stats, + MY_MUTEX_INIT_FAST); + (void) pthread_mutex_init(&LOCK_global_table_stats, MY_MUTEX_INIT_FAST); + (void) pthread_mutex_init(&LOCK_global_index_stats, MY_MUTEX_INIT_FAST); + #ifdef HAVE_OPENSSL (void) pthread_mutex_init(&LOCK_des_key_file,MY_MUTEX_INIT_FAST); #ifndef HAVE_YASSL @@ -4005,6 +4029,9 @@ server."); /* call ha_init_key_cache() on all key caches to init them */ process_key_caches(&ha_init_key_cache); + init_global_table_stats(); + init_global_index_stats(); + /* Allow storage engine to give real error messages */ if (ha_init_errors()) DBUG_RETURN(1); @@ -4210,6 +4237,8 @@ server."); init_max_user_conn(); init_update_queries(); + init_global_user_stats(); + init_global_client_stats(); DBUG_RETURN(0); } @@ -5019,6 +5048,7 @@ static void create_new_thread(THD *thd) DBUG_PRINT("error",("Too many connections")); close_connection(thd, ER_CON_COUNT_ERROR, 1); + statistic_increment(denied_connections, &LOCK_status); delete thd; DBUG_VOID_RETURN; } @@ -5810,6 +5840,7 @@ enum options_mysqld OPT_LOG_SLOW_RATE_LIMIT, OPT_LOG_SLOW_VERBOSITY, OPT_LOG_SLOW_FILTER, + OPT_USERSTAT, OPT_GENERAL_LOG_FILE, OPT_SLOW_QUERY_LOG_FILE, OPT_IGNORE_BUILTIN_INNODB @@ -7209,6 +7240,10 @@ The minimum value for this variable is 4 (uchar**) &max_system_variables.net_wait_timeout, 0, GET_ULONG, REQUIRED_ARG, NET_WAIT_TIMEOUT, 1, IF_WIN(INT_MAX32/1000, LONG_TIMEOUT), 0, 1, 0}, + {"userstat", OPT_USERSTAT, + "Control USER_STATISTICS, CLIENT_STATISTICS, INDEX_STATISTICS and TABLE_STATISTICS running", + (uchar**) &opt_userstat_running, (uchar**) &opt_userstat_running, + 0, GET_BOOL, NO_ARG, 0, 0, 1, 0, 1, 0}, {0, 0, 0, 0, 0, 0, GET_NO_ARG, NO_ARG, 0, 0, 0, 0, 0, 0} }; @@ -7579,19 +7614,24 @@ static int show_ssl_get_cipher_list(THD SHOW_VAR status_vars[]= { {"Aborted_clients", (char*) &aborted_threads, SHOW_LONG}, {"Aborted_connects", (char*) &aborted_connects, SHOW_LONG}, + {"Access_denied_errors", (char*) offsetof(STATUS_VAR, access_denied_errors), SHOW_LONG_STATUS}, {"Binlog_cache_disk_use", (char*) &binlog_cache_disk_use, SHOW_LONG}, {"Binlog_cache_use", (char*) &binlog_cache_use, SHOW_LONG}, + {"Busy_time", (char*) offsetof(STATUS_VAR, busy_time), SHOW_DOUBLE_STATUS}, {"Bytes_received", (char*) offsetof(STATUS_VAR, bytes_received), SHOW_LONGLONG_STATUS}, {"Bytes_sent", (char*) offsetof(STATUS_VAR, bytes_sent), SHOW_LONGLONG_STATUS}, + {"Binlog_bytes_written", (char*) offsetof(STATUS_VAR, binlog_bytes_written), SHOW_LONGLONG_STATUS}, {"Com", (char*) com_status_vars, SHOW_ARRAY}, {"Compression", (char*) &show_net_compression, SHOW_FUNC}, {"Connections", (char*) &thread_id, SHOW_LONG_NOFLUSH}, + {"Cpu_time", (char*) offsetof(STATUS_VAR, cpu_time), SHOW_DOUBLE_STATUS}, {"Created_tmp_disk_tables", (char*) offsetof(STATUS_VAR, created_tmp_disk_tables), SHOW_LONG_STATUS}, {"Created_tmp_files", (char*) &my_tmp_file_created, SHOW_LONG}, {"Created_tmp_tables", (char*) offsetof(STATUS_VAR, created_tmp_tables), SHOW_LONG_STATUS}, {"Delayed_errors", (char*) &delayed_insert_errors, SHOW_LONG}, {"Delayed_insert_threads", (char*) &delayed_insert_threads, SHOW_LONG_NOFLUSH}, {"Delayed_writes", (char*) &delayed_insert_writes, SHOW_LONG}, + {"Empty_queries", (char*) offsetof(STATUS_VAR, empty_queries), SHOW_LONG_STATUS}, {"Flush_commands", (char*) &refresh_version, SHOW_LONG_NOFLUSH}, {"Handler_commit", (char*) offsetof(STATUS_VAR, ha_commit_count), SHOW_LONG_STATUS}, {"Handler_delete", (char*) offsetof(STATUS_VAR, ha_delete_count), SHOW_LONG_STATUS}, @@ -7626,6 +7666,8 @@ SHOW_VAR status_vars[]= { {"Opened_tables", (char*) offsetof(STATUS_VAR, opened_tables), SHOW_LONG_STATUS}, {"Opened_table_definitions", (char*) offsetof(STATUS_VAR, opened_shares), SHOW_LONG_STATUS}, {"Prepared_stmt_count", (char*) &show_prepared_stmt_count, SHOW_FUNC}, + {"Rows_sent", (char*) offsetof(STATUS_VAR, rows_sent), SHOW_LONG_STATUS}, + {"Rows_read", (char*) offsetof(STATUS_VAR, rows_read), SHOW_LONG_STATUS}, #ifdef HAVE_QUERY_CACHE {"Qcache_free_blocks", (char*) &query_cache.free_memory_blocks, SHOW_LONG_NOFLUSH}, {"Qcache_free_memory", (char*) &query_cache.free_memory, SHOW_LONG_NOFLUSH}, @@ -9110,6 +9152,8 @@ void refresh_status(THD *thd) /* Reset thread's status variables */ bzero((uchar*) &thd->status_var, sizeof(thd->status_var)); + bzero((uchar*) &thd->org_status_var, sizeof(thd->org_status_var)); + thd->start_bytes_received= 0; /* Reset some global variables */ reset_status_vars(); === modified file 'sql/opt_range.cc' --- a/sql/opt_range.cc 2009-09-09 21:59:28 +0000 +++ b/sql/opt_range.cc 2009-10-19 17:14:48 +0000 @@ -8230,7 +8230,7 @@ int QUICK_ROR_INTERSECT_SELECT::get_next /* We get here if we got the same row ref in all scans. */ if (need_to_fetch_row) - error= head->file->rnd_pos(head->record[0], last_rowid); + error= head->file->ha_rnd_pos(head->record[0], last_rowid); } while (error == HA_ERR_RECORD_DELETED); DBUG_RETURN(error); } @@ -8296,7 +8296,7 @@ int QUICK_ROR_UNION_SELECT::get_next() cur_rowid= prev_rowid; prev_rowid= tmp; - error= head->file->rnd_pos(quick->record, prev_rowid); + error= head->file->ha_rnd_pos(quick->record, prev_rowid); } while (error == HA_ERR_RECORD_DELETED); DBUG_RETURN(error); } @@ -8521,10 +8521,12 @@ int QUICK_RANGE_SELECT::get_next_prefix( key_range start_key, end_key; if (last_range) { - /* Read the next record in the same range with prefix after cur_prefix. */ + /* + Read the next record in the same range with prefix after cur_prefix. + */ DBUG_ASSERT(cur_prefix != 0); - result= file->index_read_map(record, cur_prefix, keypart_map, - HA_READ_AFTER_KEY); + result= file->ha_index_read_map(record, cur_prefix, keypart_map, + HA_READ_AFTER_KEY); if (result || (file->compare_key(file->end_range) <= 0)) DBUG_RETURN(result); } @@ -8580,8 +8582,8 @@ int QUICK_RANGE_SELECT_GEOM::get_next() if (last_range) { // Already read through key - result= file->index_next_same(record, last_range->min_key, - last_range->min_length); + result= file->ha_index_next_same(record, last_range->min_key, + last_range->min_length); if (result != HA_ERR_END_OF_FILE) DBUG_RETURN(result); } @@ -8595,10 +8597,10 @@ int QUICK_RANGE_SELECT_GEOM::get_next() } last_range= *(cur_range++); - result= file->index_read_map(record, last_range->min_key, - last_range->min_keypart_map, - (ha_rkey_function)(last_range->flag ^ - GEOM_FLAG)); + result= file->ha_index_read_map(record, last_range->min_key, + last_range->min_keypart_map, + (ha_rkey_function)(last_range->flag ^ + GEOM_FLAG)); if (result != HA_ERR_KEY_NOT_FOUND && result != HA_ERR_END_OF_FILE) DBUG_RETURN(result); last_range= 0; // Not found, to next range @@ -8710,9 +8712,9 @@ int QUICK_SELECT_DESC::get_next() { // Already read through key result = ((last_range->flag & EQ_RANGE && used_key_parts <= head->key_info[index].key_parts) ? - file->index_next_same(record, last_range->min_key, + file->ha_index_next_same(record, last_range->min_key, last_range->min_length) : - file->index_prev(record)); + file->ha_index_prev(record)); if (!result) { if (cmp_prev(*rev_it.ref()) == 0) @@ -8728,7 +8730,7 @@ int QUICK_SELECT_DESC::get_next() if (last_range->flag & NO_MAX_RANGE) // Read last record { int local_error; - if ((local_error=file->index_last(record))) + if ((local_error= file->ha_index_last(record))) DBUG_RETURN(local_error); // Empty table if (cmp_prev(last_range) == 0) DBUG_RETURN(0); @@ -8740,9 +8742,9 @@ int QUICK_SELECT_DESC::get_next() used_key_parts <= head->key_info[index].key_parts) { - result = file->index_read_map(record, last_range->max_key, - last_range->max_keypart_map, - HA_READ_KEY_EXACT); + result= file->ha_index_read_map(record, last_range->max_key, + last_range->max_keypart_map, + HA_READ_KEY_EXACT); } else { @@ -8750,11 +8752,11 @@ int QUICK_SELECT_DESC::get_next() (last_range->flag & EQ_RANGE && used_key_parts > head->key_info[index].key_parts) || range_reads_after_key(last_range)); - result=file->index_read_map(record, last_range->max_key, - last_range->max_keypart_map, - ((last_range->flag & NEAR_MAX) ? - HA_READ_BEFORE_KEY : - HA_READ_PREFIX_LAST_OR_PREV)); + result= file->ha_index_read_map(record, last_range->max_key, + last_range->max_keypart_map, + ((last_range->flag & NEAR_MAX) ? + HA_READ_BEFORE_KEY : + HA_READ_PREFIX_LAST_OR_PREV)); } if (result) { @@ -10467,7 +10469,7 @@ int QUICK_GROUP_MIN_MAX_SELECT::reset(vo DBUG_RETURN(result); if (quick_prefix_select && quick_prefix_select->reset()) DBUG_RETURN(1); - result= file->index_last(record); + result= file->ha_index_last(record); if (result == HA_ERR_END_OF_FILE) DBUG_RETURN(0); /* Save the prefix of the last group. */ @@ -10569,9 +10571,9 @@ int QUICK_GROUP_MIN_MAX_SELECT::get_next first sub-group with the extended prefix. */ if (!have_min && !have_max && key_infix_len > 0) - result= file->index_read_map(record, group_prefix, - make_prev_keypart_map(real_key_parts), - HA_READ_KEY_EXACT); + result= file->ha_index_read_map(record, group_prefix, + make_prev_keypart_map(real_key_parts), + HA_READ_KEY_EXACT); result= have_min ? min_res : have_max ? max_res : result; } while ((result == HA_ERR_KEY_NOT_FOUND || result == HA_ERR_END_OF_FILE) && @@ -10633,9 +10635,10 @@ int QUICK_GROUP_MIN_MAX_SELECT::next_min /* Apply the constant equality conditions to the non-group select fields */ if (key_infix_len > 0) { - if ((result= file->index_read_map(record, group_prefix, - make_prev_keypart_map(real_key_parts), - HA_READ_KEY_EXACT))) + if ((result= + file->ha_index_read_map(record, group_prefix, + make_prev_keypart_map(real_key_parts), + HA_READ_KEY_EXACT))) DBUG_RETURN(result); } @@ -10650,9 +10653,9 @@ int QUICK_GROUP_MIN_MAX_SELECT::next_min { /* Find the first subsequent record without NULL in the MIN/MAX field. */ key_copy(tmp_record, record, index_info, 0); - result= file->index_read_map(record, tmp_record, - make_keypart_map(real_key_parts), - HA_READ_AFTER_KEY); + result= file->ha_index_read_map(record, tmp_record, + make_keypart_map(real_key_parts), + HA_READ_AFTER_KEY); /* Check if the new record belongs to the current group by comparing its prefix with the group's prefix. If it is from the next group, then the @@ -10707,9 +10710,9 @@ int QUICK_GROUP_MIN_MAX_SELECT::next_max if (min_max_ranges.elements > 0) result= next_max_in_range(); else - result= file->index_read_map(record, group_prefix, - make_prev_keypart_map(real_key_parts), - HA_READ_PREFIX_LAST); + result= file->ha_index_read_map(record, group_prefix, + make_prev_keypart_map(real_key_parts), + HA_READ_PREFIX_LAST); DBUG_RETURN(result); } @@ -10752,7 +10755,7 @@ int QUICK_GROUP_MIN_MAX_SELECT::next_pre { if (!seen_first_key) { - result= file->index_first(record); + result= file->ha_index_first(record); if (result) DBUG_RETURN(result); seen_first_key= TRUE; @@ -10760,9 +10763,9 @@ int QUICK_GROUP_MIN_MAX_SELECT::next_pre else { /* Load the first key in this group into record. */ - result= file->index_read_map(record, group_prefix, - make_prev_keypart_map(group_key_parts), - HA_READ_AFTER_KEY); + result= file->ha_index_read_map(record, group_prefix, + make_prev_keypart_map(group_key_parts), + HA_READ_AFTER_KEY); if (result) DBUG_RETURN(result); } @@ -10839,7 +10842,8 @@ int QUICK_GROUP_MIN_MAX_SELECT::next_min HA_READ_AFTER_KEY : HA_READ_KEY_OR_NEXT; } - result= file->index_read_map(record, group_prefix, keypart_map, find_flag); + result= file->ha_index_read_map(record, group_prefix, keypart_map, + find_flag); if (result) { if ((result == HA_ERR_KEY_NOT_FOUND || result == HA_ERR_END_OF_FILE) && @@ -10978,7 +10982,8 @@ int QUICK_GROUP_MIN_MAX_SELECT::next_max HA_READ_BEFORE_KEY : HA_READ_PREFIX_LAST_OR_PREV; } - result= file->index_read_map(record, group_prefix, keypart_map, find_flag); + result= file->ha_index_read_map(record, group_prefix, keypart_map, + find_flag); if (result) { === modified file 'sql/opt_range.h' --- a/sql/opt_range.h 2009-09-02 08:40:18 +0000 +++ b/sql/opt_range.h 2009-10-19 17:14:48 +0000 @@ -727,7 +727,7 @@ public: ~FT_SELECT() { file->ft_end(); } int init() { return error=file->ft_init(); } int reset() { return 0; } - int get_next() { return error=file->ft_read(record); } + int get_next() { return error= file->ha_ft_read(record); } int get_type() { return QS_TYPE_FULLTEXT; } }; === modified file 'sql/opt_sum.cc' --- a/sql/opt_sum.cc 2009-09-07 20:50:10 +0000 +++ b/sql/opt_sum.cc 2009-10-19 17:14:48 +0000 @@ -254,7 +254,7 @@ int opt_sum_query(TABLE_LIST *tables, Li error= table->file->ha_index_init((uint) ref.key, 1); if (!ref.key_length) - error= table->file->index_first(table->record[0]); + error= table->file->ha_index_first(table->record[0]); else { /* @@ -276,10 +276,10 @@ int opt_sum_query(TABLE_LIST *tables, Li Closed interval: Either The MIN argument is non-nullable, or we have a >= predicate for the MIN argument. */ - error= table->file->index_read_map(table->record[0], - ref.key_buff, - make_prev_keypart_map(ref.key_parts), - HA_READ_KEY_OR_NEXT); + error= table->file->ha_index_read_map(table->record[0], + ref.key_buff, + make_prev_keypart_map(ref.key_parts), + HA_READ_KEY_OR_NEXT); else { /* @@ -288,10 +288,10 @@ int opt_sum_query(TABLE_LIST *tables, Li 2) there is a > predicate on it, nullability is irrelevant. We need to scan the next bigger record first. */ - error= table->file->index_read_map(table->record[0], - ref.key_buff, - make_prev_keypart_map(ref.key_parts), - HA_READ_AFTER_KEY); + error= table->file->ha_index_read_map(table->record[0], + ref.key_buff, + make_prev_keypart_map(ref.key_parts), + HA_READ_AFTER_KEY); /* If the found record is outside the group formed by the search prefix, or there is no such record at all, check if all @@ -314,10 +314,10 @@ int opt_sum_query(TABLE_LIST *tables, Li key_cmp_if_same(table, ref.key_buff, ref.key, prefix_len))) { DBUG_ASSERT(item_field->field->real_maybe_null()); - error= table->file->index_read_map(table->record[0], - ref.key_buff, - make_prev_keypart_map(ref.key_parts), - HA_READ_KEY_EXACT); + error= table->file->ha_index_read_map(table->record[0], + ref.key_buff, + make_prev_keypart_map(ref.key_parts), + HA_READ_KEY_EXACT); } } } @@ -402,13 +402,13 @@ int opt_sum_query(TABLE_LIST *tables, Li error= table->file->ha_index_init((uint) ref.key, 1); if (!ref.key_length) - error= table->file->index_last(table->record[0]); + error= table->file->ha_index_last(table->record[0]); else - error= table->file->index_read_map(table->record[0], key_buff, - make_prev_keypart_map(ref.key_parts), - range_fl & NEAR_MAX ? - HA_READ_BEFORE_KEY : - HA_READ_PREFIX_LAST_OR_PREV); + error= table->file->ha_index_read_map(table->record[0], key_buff, + make_prev_keypart_map(ref.key_parts), + range_fl & NEAR_MAX ? + HA_READ_BEFORE_KEY : + HA_READ_PREFIX_LAST_OR_PREV); if (!error && reckey_in_range(1, &ref, item_field->field, conds, range_fl, prefix_len)) error= HA_ERR_KEY_NOT_FOUND; === modified file 'sql/records.cc' --- a/sql/records.cc 2009-05-06 12:03:24 +0000 +++ b/sql/records.cc 2009-10-19 17:14:48 +0000 @@ -342,7 +342,7 @@ static int rr_quick(READ_RECORD *info) static int rr_index_first(READ_RECORD *info) { - int tmp= info->file->index_first(info->record); + int tmp= info->file->ha_index_first(info->record); info->read_record= rr_index; if (tmp) tmp= rr_handle_error(info, tmp); @@ -368,7 +368,7 @@ static int rr_index_first(READ_RECORD *i static int rr_index(READ_RECORD *info) { - int tmp= info->file->index_next(info->record); + int tmp= info->file->ha_index_next(info->record); if (tmp) tmp= rr_handle_error(info, tmp); return tmp; @@ -378,7 +378,7 @@ static int rr_index(READ_RECORD *info) int rr_sequential(READ_RECORD *info) { int tmp; - while ((tmp=info->file->rnd_next(info->record))) + while ((tmp= info->file->ha_rnd_next(info->record))) { if (info->thd->killed) { @@ -406,7 +406,7 @@ static int rr_from_tempfile(READ_RECORD { if (my_b_read(info->io_cache,info->ref_pos,info->ref_length)) return -1; /* End of file */ - if (!(tmp=info->file->rnd_pos(info->record,info->ref_pos))) + if (!(tmp= info->file->ha_rnd_pos(info->record,info->ref_pos))) break; /* The following is extremely unlikely to happen */ if (tmp == HA_ERR_RECORD_DELETED || @@ -457,7 +457,7 @@ static int rr_from_pointers(READ_RECORD cache_pos= info->cache_pos; info->cache_pos+= info->ref_length; - if (!(tmp=info->file->rnd_pos(info->record,cache_pos))) + if (!(tmp= info->file->ha_rnd_pos(info->record,cache_pos))) break; /* The following is extremely unlikely to happen */ @@ -590,7 +590,7 @@ static int rr_from_cache(READ_RECORD *in record=uint3korr(position); position+=3; record_pos=info->cache+record*info->reclength; - if ((error=(int16) info->file->rnd_pos(record_pos,info->ref_pos))) + if ((error=(int16) info->file->ha_rnd_pos(record_pos,info->ref_pos))) { record_pos[info->error_offset]=1; shortstore(record_pos,error); === modified file 'sql/set_var.cc' --- a/sql/set_var.cc 2009-09-15 10:46:35 +0000 +++ b/sql/set_var.cc 2009-10-19 17:14:48 +0000 @@ -511,6 +511,9 @@ static sys_var_const sys_prot static sys_var_thd_ulong sys_read_buff_size(&vars, "read_buffer_size", &SV::read_buff_size); static sys_var_opt_readonly sys_readonly(&vars, "read_only", &opt_readonly); +static sys_var_bool_ptr sys_userstat(&vars, "userstat", + &opt_userstat_running); + static sys_var_thd_ulong sys_read_rnd_buff_size(&vars, "read_rnd_buffer_size", &SV::read_rnd_buff_size); static sys_var_thd_ulong sys_div_precincrement(&vars, "div_precision_increment", === modified file 'sql/sp.cc' --- a/sql/sp.cc 2009-07-28 22:39:58 +0000 +++ b/sql/sp.cc 2009-10-19 17:14:48 +0000 @@ -344,8 +344,9 @@ db_find_routine_aux(THD *thd, int type, key_copy(key, table->record[0], table->key_info, table->key_info->key_length); - if (table->file->index_read_idx_map(table->record[0], 0, key, HA_WHOLE_KEY, - HA_READ_KEY_EXACT)) + if (table->file->ha_index_read_idx_map(table->record[0], 0, key, + HA_WHOLE_KEY, + HA_READ_KEY_EXACT)) DBUG_RETURN(SP_KEY_NOT_FOUND); DBUG_RETURN(SP_OK); @@ -1101,9 +1102,9 @@ sp_drop_db_routines(THD *thd, char *db) ret= SP_OK; table->file->ha_index_init(0, 1); - if (! table->file->index_read_map(table->record[0], - (uchar *)table->field[MYSQL_PROC_FIELD_DB]->ptr, - (key_part_map)1, HA_READ_KEY_EXACT)) + if (!table->file->ha_index_read_map(table->record[0], + (uchar *) table->field[MYSQL_PROC_FIELD_DB]->ptr, + (key_part_map)1, HA_READ_KEY_EXACT)) { int nxtres; bool deleted= FALSE; @@ -1118,9 +1119,11 @@ sp_drop_db_routines(THD *thd, char *db) nxtres= 0; break; } - } while (! (nxtres= table->file->index_next_same(table->record[0], - (uchar *)table->field[MYSQL_PROC_FIELD_DB]->ptr, - key_len))); + } while (!(nxtres= table->file-> + ha_index_next_same(table->record[0], + (uchar *)table->field[MYSQL_PROC_FIELD_DB]-> + ptr, + key_len))); if (nxtres != HA_ERR_END_OF_FILE) ret= SP_KEY_NOT_FOUND; if (deleted) === modified file 'sql/sql_acl.cc' --- a/sql/sql_acl.cc 2009-09-07 20:50:10 +0000 +++ b/sql/sql_acl.cc 2009-10-19 17:14:48 +0000 @@ -1834,9 +1834,9 @@ static bool update_user_table(THD *thd, key_copy((uchar *) user_key, table->record[0], table->key_info, table->key_info->key_length); - if (table->file->index_read_idx_map(table->record[0], 0, - (uchar *) user_key, HA_WHOLE_KEY, - HA_READ_KEY_EXACT)) + if (table->file->ha_index_read_idx_map(table->record[0], 0, + (uchar *) user_key, HA_WHOLE_KEY, + HA_READ_KEY_EXACT)) { my_message(ER_PASSWORD_NO_MATCH, ER(ER_PASSWORD_NO_MATCH), MYF(0)); /* purecov: deadcode */ @@ -1927,9 +1927,9 @@ static int replace_user_table(THD *thd, key_copy(user_key, table->record[0], table->key_info, table->key_info->key_length); - if (table->file->index_read_idx_map(table->record[0], 0, user_key, - HA_WHOLE_KEY, - HA_READ_KEY_EXACT)) + if (table->file->ha_index_read_idx_map(table->record[0], 0, user_key, + HA_WHOLE_KEY, + HA_READ_KEY_EXACT)) { /* what == 'N' means revoke */ if (what == 'N') @@ -2151,9 +2151,9 @@ static int replace_db_table(TABLE *table key_copy(user_key, table->record[0], table->key_info, table->key_info->key_length); - if (table->file->index_read_idx_map(table->record[0],0, user_key, - HA_WHOLE_KEY, - HA_READ_KEY_EXACT)) + if (table->file->ha_index_read_idx_map(table->record[0],0, user_key, + HA_WHOLE_KEY, + HA_READ_KEY_EXACT)) { if (what == 'N') { // no row, no revoke @@ -2369,8 +2369,9 @@ GRANT_TABLE::GRANT_TABLE(TABLE *form, TA col_privs->field[4]->store("",0, &my_charset_latin1); col_privs->file->ha_index_init(0, 1); - if (col_privs->file->index_read_map(col_privs->record[0], (uchar*) key, - (key_part_map)15, HA_READ_KEY_EXACT)) + if (col_privs->file->ha_index_read_map(col_privs->record[0], (uchar*) key, + (key_part_map)15, + HA_READ_KEY_EXACT)) { cols = 0; /* purecov: deadcode */ col_privs->file->ha_index_end(); @@ -2391,7 +2392,7 @@ GRANT_TABLE::GRANT_TABLE(TABLE *form, TA return; /* purecov: deadcode */ } my_hash_insert(&hash_columns, (uchar *) mem_check); - } while (!col_privs->file->index_next(col_privs->record[0]) && + } while (!col_privs->file->ha_index_next(col_privs->record[0]) && !key_cmp_if_same(col_privs,key,0,key_prefix_len)); col_privs->file->ha_index_end(); } @@ -2532,8 +2533,8 @@ static int replace_column_table(GRANT_TA key_copy(user_key, table->record[0], table->key_info, table->key_info->key_length); - if (table->file->index_read_map(table->record[0], user_key, HA_WHOLE_KEY, - HA_READ_KEY_EXACT)) + if (table->file->ha_index_read_map(table->record[0], user_key, + HA_WHOLE_KEY, HA_READ_KEY_EXACT)) { if (revoke_grant) { @@ -2610,9 +2611,9 @@ static int replace_column_table(GRANT_TA key_copy(user_key, table->record[0], table->key_info, key_prefix_length); - if (table->file->index_read_map(table->record[0], user_key, - (key_part_map)15, - HA_READ_KEY_EXACT)) + if (table->file->ha_index_read_map(table->record[0], user_key, + (key_part_map)15, + HA_READ_KEY_EXACT)) goto end; /* Scan through all rows with the same host,db,user and table */ @@ -2663,7 +2664,7 @@ static int replace_column_table(GRANT_TA hash_delete(&g_t->hash_columns,(uchar*) grant_column); } } - } while (!table->file->index_next(table->record[0]) && + } while (!table->file->ha_index_next(table->record[0]) && !key_cmp_if_same(table, key, 0, key_prefix_length)); } @@ -2713,9 +2714,9 @@ static int replace_table_table(THD *thd, key_copy(user_key, table->record[0], table->key_info, table->key_info->key_length); - if (table->file->index_read_idx_map(table->record[0], 0, user_key, - HA_WHOLE_KEY, - HA_READ_KEY_EXACT)) + if (table->file->ha_index_read_idx_map(table->record[0], 0, user_key, + HA_WHOLE_KEY, + HA_READ_KEY_EXACT)) { /* The following should never happen as we first check the in memory @@ -2840,10 +2841,10 @@ static int replace_routine_table(THD *th TRUE); store_record(table,record[1]); // store at pos 1 - if (table->file->index_read_idx_map(table->record[0], 0, - (uchar*) table->field[0]->ptr, - HA_WHOLE_KEY, - HA_READ_KEY_EXACT)) + if (table->file->ha_index_read_idx_map(table->record[0], 0, + (uchar*) table->field[0]->ptr, + HA_WHOLE_KEY, + HA_READ_KEY_EXACT)) { /* The following should never happen as we first check the in memory @@ -3548,7 +3549,7 @@ static my_bool grant_load_procs_priv(TAB p_table->file->ha_index_init(0, 1); p_table->use_all_columns(); - if (!p_table->file->index_first(p_table->record[0])) + if (!p_table->file->ha_index_first(p_table->record[0])) { memex_ptr= &memex; my_pthread_setspecific_ptr(THR_MALLOC, &memex_ptr); @@ -3600,7 +3601,7 @@ static my_bool grant_load_procs_priv(TAB goto end_unlock; } } - while (!p_table->file->index_next(p_table->record[0])); + while (!p_table->file->ha_index_next(p_table->record[0])); } /* Return ok */ return_val= 0; @@ -3650,7 +3651,7 @@ static my_bool grant_load(THD *thd, TABL t_table->use_all_columns(); c_table->use_all_columns(); - if (!t_table->file->index_first(t_table->record[0])) + if (!t_table->file->ha_index_first(t_table->record[0])) { memex_ptr= &memex; my_pthread_setspecific_ptr(THR_MALLOC, &memex_ptr); @@ -3685,7 +3686,7 @@ static my_bool grant_load(THD *thd, TABL goto end_unlock; } } - while (!t_table->file->index_next(t_table->record[0])); + while (!t_table->file->ha_index_next(t_table->record[0])); } return_val=0; // Return ok @@ -3957,6 +3958,8 @@ err: { char command[128]; get_privilege_desc(command, sizeof(command), want_access); + status_var_increment(thd->status_var.access_denied_errors); + my_error(ER_TABLEACCESS_DENIED_ERROR, MYF(0), command, sctx->priv_user, @@ -5203,9 +5206,9 @@ static int handle_grant_table(TABLE_LIST table->key_info->key_part[1].store_length); key_copy(user_key, table->record[0], table->key_info, key_prefix_length); - if ((error= table->file->index_read_idx_map(table->record[0], 0, - user_key, (key_part_map)3, - HA_READ_KEY_EXACT))) + if ((error= table->file->ha_index_read_idx_map(table->record[0], 0, + user_key, (key_part_map)3, + HA_READ_KEY_EXACT))) { if (error != HA_ERR_KEY_NOT_FOUND && error != HA_ERR_END_OF_FILE) { @@ -5240,7 +5243,7 @@ static int handle_grant_table(TABLE_LIST DBUG_PRINT("info",("scan table: '%s' search: '%s'@'%s'", table->s->table_name.str, user_str, host_str)); #endif - while ((error= table->file->rnd_next(table->record[0])) != + while ((error= table->file->ha_rnd_next(table->record[0])) != HA_ERR_END_OF_FILE) { if (error) === modified file 'sql/sql_base.cc' --- a/sql/sql_base.cc 2009-09-07 20:50:10 +0000 +++ b/sql/sql_base.cc 2009-10-19 17:14:48 +0000 @@ -1373,6 +1373,12 @@ bool close_thread_table(THD *thd, TABLE DBUG_PRINT("tcache", ("table: '%s'.'%s' 0x%lx", table->s->db.str, table->s->table_name.str, (long) table)); + if (table->file) + { + table->file->update_global_table_stats(); + table->file->update_global_index_stats(); + } + *table_ptr=table->next; /* When closing a MERGE parent or child table, detach the children first. @@ -1902,6 +1908,13 @@ void close_temporary(TABLE *table, bool DBUG_PRINT("tmptable", ("closing table: '%s'.'%s'", table->s->db.str, table->s->table_name.str)); + /* in_use is not set for replication temporary tables during shutdown */ + if (table->in_use) + { + table->file->update_global_table_stats(); + table->file->update_global_index_stats(); + } + free_io_cache(table); closefrm(table, 0); if (delete_table) === modified file 'sql/sql_class.cc' --- a/sql/sql_class.cc 2009-09-15 10:46:35 +0000 +++ b/sql/sql_class.cc 2009-10-19 17:14:48 +0000 @@ -615,6 +615,7 @@ THD::THD() mysys_var=0; binlog_evt_union.do_union= FALSE; enable_slow_log= 0; + #ifndef DBUG_OFF dbug_sentry=THD_SENTRY_MAGIC; #endif @@ -817,7 +818,63 @@ void THD::init(void) update_charset(); reset_current_stmt_binlog_row_based(); bzero((char *) &status_var, sizeof(status_var)); + bzero((char *) &org_status_var, sizeof(org_status_var)); sql_log_bin_toplevel= options & OPTION_BIN_LOG; + select_commands= update_commands= other_commands= 0; + /* Set to handle counting of aborted connections */ + userstat_running= opt_userstat_running; + last_global_update_time= current_connect_time= time(NULL); +} + + +/* Updates some status variables to be used by update_global_user_stats */ + +void THD::update_stats(void) +{ + /* sql_command == SQLCOM_END in case of parse errors or quit */ + if (lex->sql_command != SQLCOM_END) + { + /* The replication thread has the COM_CONNECT command */ + DBUG_ASSERT(command == COM_QUERY || command == COM_CONNECT); + + /* A SQL query. */ + if (lex->sql_command == SQLCOM_SELECT) + select_commands++; + else if (! sql_command_flags[lex->sql_command] & CF_STATUS_COMMAND) + { + /* Ignore 'SHOW ' commands */ + } + else if (is_update_query(lex->sql_command)) + update_commands++; + else + other_commands++; + } +} + + +void THD::update_all_stats() +{ + time_t save_time; + ulonglong end_cpu_time, end_utime; + double busy_time, cpu_time; + + /* This is set at start of query if opt_userstat_running was set */ + if (!userstat_running) + return; + + end_cpu_time= my_getcputime(); + end_utime= my_micro_time_and_time(&save_time); + busy_time= (end_utime - start_utime) / 1000000.0; + cpu_time= (end_cpu_time - start_cpu_time) / 10000000.0; + /* In case there are bad values, 2629743 is the #seconds in a month. */ + if (cpu_time > 2629743.0) + cpu_time= 0; + status_var_add(status_var.cpu_time, cpu_time); + status_var_add(status_var.busy_time, busy_time); + + /* Updates THD stats and the global user stats. */ + update_stats(); + update_global_user_stats(this, TRUE, save_time); } @@ -984,9 +1041,8 @@ THD::~THD() from_var from this array NOTES - This function assumes that all variables are long/ulong. - If this assumption will change, then we have to explictely add - the other variables after the while loop + This function assumes that all variables at start are long/ulong and + other types are handled explicitely */ void add_to_status(STATUS_VAR *to_var, STATUS_VAR *from_var) @@ -998,6 +1054,13 @@ void add_to_status(STATUS_VAR *to_var, S while (to != end) *(to++)+= *(from++); + + /* Handle the not ulong variables. See end of system_status_var */ + to_var->bytes_received= from_var->bytes_received; + to_var->bytes_sent+= from_var->bytes_sent; + to_var->binlog_bytes_written= from_var->binlog_bytes_written; + to_var->cpu_time+= from_var->cpu_time; + to_var->busy_time+= from_var->busy_time; } /* @@ -1010,7 +1073,8 @@ void add_to_status(STATUS_VAR *to_var, S dec_var minus this array NOTE - This function assumes that all variables are long/ulong. + This function assumes that all variables at start are long/ulong and + other types are handled explicitely */ void add_diff_to_status(STATUS_VAR *to_var, STATUS_VAR *from_var, @@ -1023,6 +1087,14 @@ void add_diff_to_status(STATUS_VAR *to_v while (to != end) *(to++)+= *(from++) - *(dec++); + + to_var->bytes_received= (from_var->bytes_received - + dec_var->bytes_received); + to_var->bytes_sent+= from_var->bytes_sent - dec_var->bytes_sent; + to_var->binlog_bytes_written= (from_var->binlog_bytes_written - + dec_var->binlog_bytes_written); + to_var->cpu_time+= from_var->cpu_time - dec_var->cpu_time; + to_var->busy_time+= from_var->busy_time - dec_var->busy_time; } #define SECONDS_TO_WAIT_FOR_KILL 2 @@ -2773,7 +2845,8 @@ void thd_increment_bytes_sent(ulong leng { THD *thd=current_thd; if (likely(thd != 0)) - { /* current_thd==0 when close_connection() calls net_send_error() */ + { + /* current_thd == 0 when close_connection() calls net_send_error() */ thd->status_var.bytes_sent+= length; } } === modified file 'sql/sql_class.h' --- a/sql/sql_class.h 2009-09-15 10:46:35 +0000 +++ b/sql/sql_class.h 2009-10-19 17:14:48 +0000 @@ -415,8 +415,6 @@ struct system_variables typedef struct system_status_var { - ulonglong bytes_received; - ulonglong bytes_sent; ulong com_other; ulong com_stat[(uint) SQLCOM_END]; ulong created_tmp_disk_tables; @@ -455,6 +453,8 @@ typedef struct system_status_var ulong select_range_count; ulong select_range_check_count; ulong select_scan_count; + ulong rows_read; + ulong rows_sent; ulong long_query_count; ulong filesort_merge_passes; ulong filesort_range_count; @@ -472,6 +472,9 @@ typedef struct system_status_var Number of statements sent from the client */ ulong questions; + ulong empty_queries; + ulong access_denied_errors; /* Can only be 0 or 1 */ + ulong lost_connections; /* IMPORTANT! SEE last_system_status_var DEFINITION BELOW. @@ -480,12 +483,16 @@ typedef struct system_status_var Status variables which it does not make sense to add to global status variable counter */ + ulonglong bytes_received; + ulonglong bytes_sent; + ulonglong binlog_bytes_written; double last_query_cost; + double cpu_time, busy_time; } STATUS_VAR; /* This is used for 'SHOW STATUS'. It must be updated to the last ulong - variable in system_status_var which is makes sens to add to the global + variable in system_status_var which is makes sense to add to the global counter */ @@ -1299,6 +1306,7 @@ public: struct my_rnd_struct rand; // used for authentication struct system_variables variables; // Changeable local variables struct system_status_var status_var; // Per thread statistic vars + struct system_status_var org_status_var; // For user statistics struct system_status_var *initial_status_var; /* used by show status */ THR_LOCK_INFO lock_info; // Locking info of this thread THR_LOCK_OWNER main_lock_id; // To use for conventional queries @@ -1399,6 +1407,8 @@ public: uint in_sub_stmt; /* TRUE when the current top has SQL_LOG_BIN ON */ bool sql_log_bin_toplevel; + /* True when opt_userstat_running is set at start of query */ + bool userstat_running; /* container for handler's private per-connection data */ Ha_data ha_data[MAX_HA]; @@ -1842,6 +1852,21 @@ public: */ LOG_INFO* current_linfo; NET* slave_net; // network connection from slave -> m. + + /* + Used to update global user stats. The global user stats are updated + occasionally with the 'diff' variables. After the update, the 'diff' + variables are reset to 0. + */ + /* Time when the current thread connected to MySQL. */ + time_t current_connect_time; + /* Last time when THD stats were updated in global_user_stats. */ + time_t last_global_update_time; + /* Number of commands not reflected in global_user_stats yet. */ + uint select_commands, update_commands, other_commands; + ulonglong start_cpu_time; + ulonglong start_bytes_received; + /* Used by the sys_var class to store temporary values */ union { @@ -1902,6 +1927,8 @@ public: alloc_root. */ void init_for_queries(); + void update_all_stats(); + void update_stats(void); void change_user(void); void cleanup(void); void cleanup_after_query(); @@ -2319,7 +2346,6 @@ private: MEM_ROOT main_mem_root; }; - /** A short cut for thd->main_da.set_ok_status(). */ inline void === modified file 'sql/sql_connect.cc' --- a/sql/sql_connect.cc 2009-09-07 20:50:10 +0000 +++ b/sql/sql_connect.cc 2009-10-19 17:14:48 +0000 @@ -20,6 +20,13 @@ #include "mysql_priv.h" +HASH global_user_stats, global_client_stats, global_table_stats; +HASH global_index_stats; +/* Protects the above global stats */ +extern pthread_mutex_t LOCK_global_user_client_stats; +extern pthread_mutex_t LOCK_global_table_stats; +extern pthread_mutex_t LOCK_global_index_stats; + #ifdef HAVE_OPENSSL /* Without SSL the handshake consists of one packet. This packet @@ -459,6 +466,7 @@ check_user(THD *thd, enum enum_server_co check_for_max_user_connections(thd, thd->user_connect)) { /* The error is set in check_for_max_user_connections(). */ + status_var_increment(denied_connections); DBUG_RETURN(1); } @@ -470,6 +478,7 @@ check_user(THD *thd, enum enum_server_co /* mysql_change_db() has pushed the error message. */ if (thd->user_connect) decrease_user_connections(thd->user_connect); + status_var_increment(thd->status_var.access_denied_errors); DBUG_RETURN(1); } } @@ -493,6 +502,8 @@ check_user(THD *thd, enum enum_server_co thd->main_security_ctx.user, thd->main_security_ctx.host_or_ip, passwd_len ? ER(ER_YES) : ER(ER_NO)); + status_var_increment(thd->status_var.access_denied_errors); + DBUG_RETURN(1); #endif /* NO_EMBEDDED_ACCESS_CHECKS */ } @@ -520,10 +531,14 @@ extern "C" void free_user(struct user_co void init_max_user_conn(void) { #ifndef NO_EMBEDDED_ACCESS_CHECKS - (void) hash_init(&hash_user_connections,system_charset_info,max_connections, - 0,0, - (hash_get_key) get_key_conn, (hash_free_key) free_user, - 0); + if (hash_init(&hash_user_connections,system_charset_info,max_connections, + 0,0, + (hash_get_key) get_key_conn, (hash_free_key) free_user, + 0)) + { + sql_print_error("Initializing hash_user_connections failed."); + exit(1); + } #endif } @@ -576,6 +591,445 @@ void reset_mqh(LEX_USER *lu, bool get_th #endif /* NO_EMBEDDED_ACCESS_CHECKS */ } +/***************************************************************************** + Handle users statistics +*****************************************************************************/ + +/* 'mysql_system_user' is used for when the user is not defined for a THD. */ +static const char mysql_system_user[]= "#mysql_system#"; + +// Returns 'user' if it's not NULL. Returns 'mysql_system_user' otherwise. +static const char * get_valid_user_string(char* user) +{ + return user ? user : mysql_system_user; +} + +/* + Returns string as 'IP' for the client-side of the connection represented by + 'client'. Does not allocate memory. May return "". +*/ + +static const char *get_client_host(THD *client) +{ + return client->security_ctx->host_or_ip[0] ? + client->security_ctx->host_or_ip : + client->security_ctx->host ? client->security_ctx->host : ""; +} + +extern "C" uchar *get_key_user_stats(USER_STATS *user_stats, size_t *length, + my_bool not_used __attribute__((unused))) +{ + *length= user_stats->user_name_length; + return (uchar*) user_stats->user; +} + +void free_user_stats(USER_STATS* user_stats) +{ + my_free(user_stats, MYF(0)); +} + +void init_user_stats(USER_STATS *user_stats, + const char *user, + size_t user_length, + const char *priv_user, + uint total_connections, + uint concurrent_connections, + time_t connected_time, + double busy_time, + double cpu_time, + ulonglong bytes_received, + ulonglong bytes_sent, + ulonglong binlog_bytes_written, + ha_rows rows_sent, + ha_rows rows_read, + ha_rows rows_inserted, + ha_rows rows_deleted, + ha_rows rows_updated, + ulonglong select_commands, + ulonglong update_commands, + ulonglong other_commands, + ulonglong commit_trans, + ulonglong rollback_trans, + ulonglong denied_connections, + ulonglong lost_connections, + ulonglong access_denied_errors, + ulonglong empty_queries) +{ + DBUG_ENTER("init_user_stats"); + DBUG_PRINT("enter", ("user: %s priv_user: %s", user, priv_user)); + + user_length= min(user_length, sizeof(user_stats->user)-1); + memcpy(user_stats->user, user, user_length); + user_stats->user[user_length]= 0; + user_stats->user_name_length= user_length; + strmake(user_stats->priv_user, priv_user, sizeof(user_stats->priv_user)-1); + + user_stats->total_connections= total_connections; + user_stats->concurrent_connections= concurrent_connections; + user_stats->connected_time= connected_time; + user_stats->busy_time= busy_time; + user_stats->cpu_time= cpu_time; + user_stats->bytes_received= bytes_received; + user_stats->bytes_sent= bytes_sent; + user_stats->binlog_bytes_written= binlog_bytes_written; + user_stats->rows_sent= rows_sent; + user_stats->rows_updated= rows_updated; + user_stats->rows_read= rows_read; + user_stats->select_commands= select_commands; + user_stats->update_commands= update_commands; + user_stats->other_commands= other_commands; + user_stats->commit_trans= commit_trans; + user_stats->rollback_trans= rollback_trans; + user_stats->denied_connections= denied_connections; + user_stats->lost_connections= lost_connections; + user_stats->access_denied_errors= access_denied_errors; + user_stats->empty_queries= empty_queries; + DBUG_VOID_RETURN; +} + + +#ifdef COMPLEAT_PATCH_NOT_ADDED_YET + +void add_user_stats(USER_STATS *user_stats, + uint total_connections, + uint concurrent_connections, + time_t connected_time, + double busy_time, + double cpu_time, + ulonglong bytes_received, + ulonglong bytes_sent, + ulonglong binlog_bytes_written, + ha_rows rows_sent, + ha_rows rows_read, + ha_rows rows_inserted, + ha_rows rows_deleted, + ha_rows rows_updated, + ulonglong select_commands, + ulonglong update_commands, + ulonglong other_commands, + ulonglong commit_trans, + ulonglong rollback_trans, + ulonglong denied_connections, + ulonglong lost_connections, + ulonglong access_denied_errors, + ulonglong empty_queries) +{ + user_stats->total_connections+= total_connections; + user_stats->concurrent_connections+= concurrent_connections; + user_stats->connected_time+= connected_time; + user_stats->busy_time+= busy_time; + user_stats->cpu_time+= cpu_time; + user_stats->bytes_received+= bytes_received; + user_stats->bytes_sent+= bytes_sent; + user_stats->binlog_bytes_written+= binlog_bytes_written; + user_stats->rows_sent+= rows_sent; + user_stats->rows_inserted+= rows_inserted; + user_stats->rows_deleted+= rows_deleted; + user_stats->rows_updated+= rows_updated; + user_stats->rows_read+= rows_read; + user_stats->select_commands+= select_commands; + user_stats->update_commands+= update_commands; + user_stats->other_commands+= other_commands; + user_stats->commit_trans+= commit_trans; + user_stats->rollback_trans+= rollback_trans; + user_stats->denied_connections+= denied_connections; + user_stats->lost_connections+= lost_connections; + user_stats->access_denied_errors+= access_denied_errors; + user_stats->empty_queries+= empty_queries; +} +#endif + + +void init_global_user_stats(void) +{ + if (hash_init(&global_user_stats, system_charset_info, max_connections, + 0, 0, (hash_get_key) get_key_user_stats, + (hash_free_key)free_user_stats, 0)) + { + sql_print_error("Initializing global_user_stats failed."); + exit(1); + } +} + +void init_global_client_stats(void) +{ + if (hash_init(&global_client_stats, system_charset_info, max_connections, + 0, 0, (hash_get_key) get_key_user_stats, + (hash_free_key)free_user_stats, 0)) + { + sql_print_error("Initializing global_client_stats failed."); + exit(1); + } +} + +extern "C" uchar *get_key_table_stats(TABLE_STATS *table_stats, size_t *length, + my_bool not_used __attribute__((unused))) +{ + *length= table_stats->table_name_length; + return (uchar*) table_stats->table; +} + +extern "C" void free_table_stats(TABLE_STATS* table_stats) +{ + my_free(table_stats, MYF(0)); +} + +void init_global_table_stats(void) +{ + if (hash_init(&global_table_stats, system_charset_info, max_connections, + 0, 0, (hash_get_key) get_key_table_stats, + (hash_free_key)free_table_stats, 0)) { + sql_print_error("Initializing global_table_stats failed."); + exit(1); + } +} + +extern "C" uchar *get_key_index_stats(INDEX_STATS *index_stats, size_t *length, + my_bool not_used __attribute__((unused))) +{ + *length= index_stats->index_name_length; + return (uchar*) index_stats->index; +} + +extern "C" void free_index_stats(INDEX_STATS* index_stats) +{ + my_free(index_stats, MYF(0)); +} + +void init_global_index_stats(void) +{ + if (hash_init(&global_index_stats, system_charset_info, max_connections, + 0, 0, (hash_get_key) get_key_index_stats, + (hash_free_key)free_index_stats, 0)) + { + sql_print_error("Initializing global_index_stats failed."); + exit(1); + } +} + + +void free_global_user_stats(void) +{ + hash_free(&global_user_stats); +} + +void free_global_table_stats(void) +{ + hash_free(&global_table_stats); +} + +void free_global_index_stats(void) +{ + hash_free(&global_index_stats); +} + +void free_global_client_stats(void) +{ + hash_free(&global_client_stats); +} + +/* + Increments the global stats connection count for an entry from + global_client_stats or global_user_stats. Returns 0 on success + and 1 on error. +*/ + +static bool increment_count_by_name(const char *name, size_t name_length, + const char *role_name, + HASH *users_or_clients, THD *thd) +{ + USER_STATS *user_stats; + + if (!(user_stats= (USER_STATS*) hash_search(users_or_clients, (uchar*) name, + name_length))) + { + /* First connection for this user or client */ + if (!(user_stats= ((USER_STATS*) + my_malloc(sizeof(USER_STATS), + MYF(MY_WME | MY_ZEROFILL))))) + return TRUE; // Out of memory + + init_user_stats(user_stats, name, name_length, role_name, + 0, 0, // connections + 0, 0, 0, // time + 0, 0, 0, // bytes sent, received and written + 0, 0, // Rows sent and read + 0, 0, 0, // rows inserted, deleted and updated + 0, 0, 0, // select, update and other commands + 0, 0, // commit and rollback trans + thd->status_var.access_denied_errors, + 0, // lost connections + 0, // access denied errors + 0); // empty queries + + if (my_hash_insert(users_or_clients, (uchar*)user_stats)) + { + my_free(user_stats, 0); + return TRUE; // Out of memory + } + } + user_stats->total_connections++; + return FALSE; +} + + +/* + Increments the global user and client stats connection count. + + @param use_lock if true, LOCK_global_user_client_stats will be locked + + @retval 0 ok + @retval 1 error. +*/ + +#ifndef EMBEDDED_LIBRARY +static bool increment_connection_count(THD* thd, bool use_lock) +{ + const char *user_string= get_valid_user_string(thd->main_security_ctx.user); + const char *client_string= get_client_host(thd); + bool return_value= FALSE; + + if (!thd->userstat_running) + return FALSE; + + if (use_lock) + pthread_mutex_lock(&LOCK_global_user_client_stats); + + if (increment_count_by_name(user_string, strlen(user_string), user_string, + &global_user_stats, thd)) + { + return_value= TRUE; + goto end; + } + if (increment_count_by_name(client_string, strlen(client_string), + user_string, &global_client_stats, thd)) + { + return_value= TRUE; + goto end; + } + +end: + if (use_lock) + pthread_mutex_unlock(&LOCK_global_user_client_stats); + return return_value; +} +#endif + +/* + Used to update the global user and client stats +*/ + +static void update_global_user_stats_with_user(THD *thd, + USER_STATS *user_stats, + time_t now) +{ + DBUG_ASSERT(thd->userstat_running); + + user_stats->connected_time+= now - thd->last_global_update_time; + user_stats->busy_time+= (thd->status_var.busy_time - + thd->org_status_var.busy_time); + user_stats->cpu_time+= (thd->status_var.cpu_time - + thd->org_status_var.cpu_time); + /* + This is handle specially as bytes_recieved is incremented BEFORE + org_status_var is copied. + */ + user_stats->bytes_received+= (thd->org_status_var.bytes_received- + thd->start_bytes_received); + user_stats->bytes_sent+= (thd->status_var.bytes_sent - + thd->org_status_var.bytes_sent); + user_stats->binlog_bytes_written+= + (thd->status_var.binlog_bytes_written - + thd->org_status_var.binlog_bytes_written); + user_stats->rows_read+= (thd->status_var.rows_read - + thd->org_status_var.rows_read); + user_stats->rows_sent+= (thd->status_var.rows_sent - + thd->org_status_var.rows_sent); + user_stats->rows_inserted+= (thd->status_var.ha_write_count - + thd->org_status_var.ha_write_count); + user_stats->rows_deleted+= (thd->status_var.ha_delete_count - + thd->org_status_var.ha_delete_count); + user_stats->rows_updated+= (thd->status_var.ha_update_count - + thd->org_status_var.ha_update_count); + user_stats->select_commands+= thd->select_commands; + user_stats->update_commands+= thd->update_commands; + user_stats->other_commands+= thd->other_commands; + user_stats->commit_trans+= (thd->status_var.ha_commit_count - + thd->org_status_var.ha_commit_count); + user_stats->rollback_trans+= (thd->status_var.ha_rollback_count + + thd->status_var.ha_savepoint_rollback_count - + thd->org_status_var.ha_rollback_count - + thd->org_status_var. + ha_savepoint_rollback_count); + user_stats->access_denied_errors+= + (thd->status_var.access_denied_errors - + thd->org_status_var.access_denied_errors); + user_stats->empty_queries+= (thd->status_var.empty_queries - + thd->org_status_var.empty_queries); + + /* The following can only contain 0 or 1 and then connection ends */ + user_stats->denied_connections+= thd->status_var.access_denied_errors; + user_stats->lost_connections+= thd->status_var.lost_connections; +} + + +/* Updates the global stats of a user or client */ +void update_global_user_stats(THD *thd, bool create_user, time_t now) +{ + const char *user_string, *client_string; + USER_STATS *user_stats; + size_t user_string_length, client_string_length; + DBUG_ASSERT(thd->userstat_running); + + user_string= get_valid_user_string(thd->main_security_ctx.user); + user_string_length= strlen(user_string); + client_string= get_client_host(thd); + client_string_length= strlen(client_string); + + pthread_mutex_lock(&LOCK_global_user_client_stats); + + // Update by user name + if ((user_stats= (USER_STATS*) hash_search(&global_user_stats, + (uchar*) user_string, + user_string_length))) + { + /* Found user. */ + update_global_user_stats_with_user(thd, user_stats, now); + } + else + { + /* Create the entry */ + if (create_user) + { + increment_count_by_name(user_string, user_string_length, user_string, + &global_user_stats, thd); + } + } + + /* Update by client IP */ + if ((user_stats= (USER_STATS*)hash_search(&global_client_stats, + (uchar*) client_string, + client_string_length))) + { + // Found by client IP + update_global_user_stats_with_user(thd, user_stats, now); + } + else + { + // Create the entry + if (create_user) + { + increment_count_by_name(client_string, client_string_length, + user_string, &global_client_stats, thd); + } + } + /* Reset variables only used for counting */ + thd->select_commands= thd->update_commands= thd->other_commands= 0; + thd->last_global_update_time= now; + + pthread_mutex_unlock(&LOCK_global_user_client_stats); +} + void thd_init_client_charset(THD *thd, uint cs_number) { @@ -970,6 +1424,14 @@ bool login_connection(THD *thd) /* Connect completed, set read/write timeouts back to default */ my_net_set_read_timeout(net, thd->variables.net_read_timeout); my_net_set_write_timeout(net, thd->variables.net_write_timeout); + + /* Updates global user connection stats. */ + if (increment_connection_count(thd, TRUE)) + { + net_send_error(thd, ER_OUTOFMEMORY); // Out of memory + DBUG_RETURN(1); + } + DBUG_RETURN(0); } @@ -991,6 +1453,7 @@ void end_connection(THD *thd) if (thd->killed || (net->error && net->vio != 0)) { statistic_increment(aborted_threads,&LOCK_status); + status_var_increment(thd->status_var.lost_connections); } if (net->error && net->vio != 0) @@ -1117,10 +1580,14 @@ pthread_handler_t handle_one_connection( for (;;) { NET *net= &thd->net; + bool create_user= TRUE; lex_start(thd); if (login_connection(thd)) + { + create_user= FALSE; goto end_thread; + } prepare_new_connection_state(thd); @@ -1134,12 +1601,14 @@ pthread_handler_t handle_one_connection( end_thread: close_connection(thd, 0, 1); + if (thd->userstat_running) + update_global_user_stats(thd, create_user, time(NULL)); + if (thd->scheduler->end_thread(thd,1)) return 0; // Probably no-threads /* - If end_thread() returns, we are either running with - thread-handler=no-threads or this thread has been schedule to + If end_thread() returns, this thread has been schedule to handle the next connection. */ thd= current_thd; === modified file 'sql/sql_cursor.cc' --- a/sql/sql_cursor.cc 2008-12-10 14:16:21 +0000 +++ b/sql/sql_cursor.cc 2009-10-19 17:14:48 +0000 @@ -655,7 +655,7 @@ void Materialized_cursor::fetch(ulong nu result->begin_dataset(); for (fetch_limit+= num_rows; fetch_count < fetch_limit; fetch_count++) { - if ((res= table->file->rnd_next(table->record[0]))) + if ((res= table->file->ha_rnd_next(table->record[0]))) break; /* Send data only if the read was successful. */ result->send_data(item_list); === modified file 'sql/sql_handler.cc' --- a/sql/sql_handler.cc 2009-07-15 23:23:57 +0000 +++ b/sql/sql_handler.cc 2009-10-19 17:14:48 +0000 @@ -565,8 +565,8 @@ retry: if (table->file->inited != handler::NONE) { error=keyname ? - table->file->index_next(table->record[0]) : - table->file->rnd_next(table->record[0]); + table->file->ha_index_next(table->record[0]) : + table->file->ha_rnd_next(table->record[0]); break; } /* else fall through */ @@ -575,13 +575,13 @@ retry: { table->file->ha_index_or_rnd_end(); table->file->ha_index_init(keyno, 1); - error= table->file->index_first(table->record[0]); + error= table->file->ha_index_first(table->record[0]); } else { table->file->ha_index_or_rnd_end(); if (!(error= table->file->ha_rnd_init(1))) - error= table->file->rnd_next(table->record[0]); + error= table->file->ha_rnd_next(table->record[0]); } mode=RNEXT; break; @@ -589,7 +589,7 @@ retry: DBUG_ASSERT(keyname != 0); if (table->file->inited != handler::NONE) { - error=table->file->index_prev(table->record[0]); + error=table->file->ha_index_prev(table->record[0]); break; } /* else fall through */ @@ -597,13 +597,13 @@ retry: DBUG_ASSERT(keyname != 0); table->file->ha_index_or_rnd_end(); table->file->ha_index_init(keyno, 1); - error= table->file->index_last(table->record[0]); + error= table->file->ha_index_last(table->record[0]); mode=RPREV; break; case RNEXT_SAME: /* Continue scan on "(keypart1,keypart2,...)=(c1, c2, ...) */ DBUG_ASSERT(keyname != 0); - error= table->file->index_next_same(table->record[0], key, key_len); + error= table->file->ha_index_next_same(table->record[0], key, key_len); break; case RKEY: { @@ -643,8 +643,8 @@ retry: table->file->ha_index_or_rnd_end(); table->file->ha_index_init(keyno, 1); key_copy(key, table->record[0], table->key_info + keyno, key_len); - error= table->file->index_read_map(table->record[0], - key, keypart_map, ha_rkey_mode); + error= table->file->ha_index_read_map(table->record[0], + key, keypart_map, ha_rkey_mode); mode=rkey_to_rnext[(int)ha_rkey_mode]; break; } === modified file 'sql/sql_help.cc' --- a/sql/sql_help.cc 2009-09-07 20:50:10 +0000 +++ b/sql/sql_help.cc 2009-10-19 17:14:48 +0000 @@ -294,13 +294,13 @@ int get_topics_for_keyword(THD *thd, TAB rkey_id->store((longlong) key_id, TRUE); rkey_id->get_key_image(buff, rkey_id->pack_length(), Field::itRAW); - int key_res= relations->file->index_read_map(relations->record[0], - buff, (key_part_map) 1, - HA_READ_KEY_EXACT); + int key_res= relations->file->ha_index_read_map(relations->record[0], + buff, (key_part_map) 1, + HA_READ_KEY_EXACT); for ( ; !key_res && key_id == (int16) rkey_id->val_int() ; - key_res= relations->file->index_next(relations->record[0])) + key_res= relations->file->ha_index_next(relations->record[0])) { uchar topic_id_buff[8]; longlong topic_id= rtopic_id->val_int(); @@ -308,8 +308,8 @@ int get_topics_for_keyword(THD *thd, TAB field->store((longlong) topic_id, TRUE); field->get_key_image(topic_id_buff, field->pack_length(), Field::itRAW); - if (!topics->file->index_read_map(topics->record[0], topic_id_buff, - (key_part_map)1, HA_READ_KEY_EXACT)) + if (!topics->file->ha_index_read_map(topics->record[0], topic_id_buff, + (key_part_map)1, HA_READ_KEY_EXACT)) { memorize_variant_topic(thd,topics,count,find_fields, names,name,description,example); === modified file 'sql/sql_insert.cc' --- a/sql/sql_insert.cc 2009-09-07 20:50:10 +0000 +++ b/sql/sql_insert.cc 2009-10-19 17:14:48 +0000 @@ -1425,7 +1425,7 @@ int write_record(THD *thd, TABLE *table, goto err; if (table->file->ha_table_flags() & HA_DUPLICATE_POS) { - if (table->file->rnd_pos(table->record[1],table->file->dup_ref)) + if (table->file->ha_rnd_pos(table->record[1],table->file->dup_ref)) goto err; } else @@ -1446,9 +1446,10 @@ int write_record(THD *thd, TABLE *table, } } key_copy((uchar*) key,table->record[0],table->key_info+key_nr,0); - if ((error=(table->file->index_read_idx_map(table->record[1],key_nr, - (uchar*) key, HA_WHOLE_KEY, - HA_READ_KEY_EXACT)))) + if ((error= (table->file->ha_index_read_idx_map(table->record[1], + key_nr, (uchar*) key, + HA_WHOLE_KEY, + HA_READ_KEY_EXACT)))) goto err; } if (info->handle_duplicates == DUP_UPDATE) === modified file 'sql/sql_lex.h' --- a/sql/sql_lex.h 2009-09-15 10:46:35 +0000 +++ b/sql/sql_lex.h 2009-10-19 17:14:48 +0000 @@ -118,6 +118,8 @@ enum enum_sql_command { SQLCOM_SHOW_CREATE_TRIGGER, SQLCOM_ALTER_DB_UPGRADE, SQLCOM_SHOW_PROFILE, SQLCOM_SHOW_PROFILES, + SQLCOM_SHOW_USER_STATS, SQLCOM_SHOW_TABLE_STATS, SQLCOM_SHOW_INDEX_STATS, + SQLCOM_SHOW_CLIENT_STATS, /* When a command is added here, be sure it's also added in mysqld.cc === modified file 'sql/sql_parse.cc' --- a/sql/sql_parse.cc 2009-09-15 10:46:35 +0000 +++ b/sql/sql_parse.cc 2009-10-19 17:14:48 +0000 @@ -331,10 +331,14 @@ void init_update_queries(void) sql_command_flags[SQLCOM_SHOW_CREATE_EVENT]= CF_STATUS_COMMAND; sql_command_flags[SQLCOM_SHOW_PROFILES]= CF_STATUS_COMMAND; sql_command_flags[SQLCOM_SHOW_PROFILE]= CF_STATUS_COMMAND; + sql_command_flags[SQLCOM_SHOW_CLIENT_STATS]= CF_STATUS_COMMAND; + sql_command_flags[SQLCOM_SHOW_USER_STATS]= CF_STATUS_COMMAND; + sql_command_flags[SQLCOM_SHOW_TABLE_STATS]= CF_STATUS_COMMAND; + sql_command_flags[SQLCOM_SHOW_INDEX_STATS]= CF_STATUS_COMMAND; - sql_command_flags[SQLCOM_SHOW_TABLES]= (CF_STATUS_COMMAND | - CF_SHOW_TABLE_COMMAND | - CF_REEXECUTION_FRAGILE); + sql_command_flags[SQLCOM_SHOW_TABLES]= (CF_STATUS_COMMAND | + CF_SHOW_TABLE_COMMAND | + CF_REEXECUTION_FRAGILE); sql_command_flags[SQLCOM_SHOW_TABLE_STATUS]= (CF_STATUS_COMMAND | CF_SHOW_TABLE_COMMAND | CF_REEXECUTION_FRAGILE); @@ -549,7 +553,6 @@ end: DBUG_RETURN(0); } - /** @brief Check access privs for a MERGE table and fix children lock types. @@ -801,6 +804,8 @@ bool do_command(THD *thd) net_new_transaction(net); + /* Save for user statistics */ + thd->start_bytes_received= thd->status_var.bytes_received; packet_length= my_net_read(net); #if defined(ENABLED_PROFILING) && defined(COMMUNITY_SERVER) thd->profiling.start_new_query(); @@ -1324,7 +1329,7 @@ bool dispatch_command(enum enum_server_c table_list.select_lex= &(thd->lex->select_lex); lex_start(thd); - mysql_reset_thd_for_next_command(thd); + mysql_reset_thd_for_next_command(thd, opt_userstat_running); thd->lex-> select_lex.table_list.link_in_list((uchar*) &table_list, @@ -1609,6 +1614,9 @@ bool dispatch_command(enum enum_server_c /* Free tables */ close_thread_tables(thd); + /* Update status; Must be done after close_thread_tables */ + thd->update_all_stats(); + log_slow_statement(thd); thd_proc_info(thd, "cleaning up"); @@ -1777,6 +1785,12 @@ int prepare_schema_table(THD *thd, LEX * thd->profiling.discard_current_query(); #endif break; + case SCH_USER_STATS: + case SCH_CLIENT_STATS: + if (check_global_access(thd, SUPER_ACL | PROCESS_ACL)) + DBUG_RETURN(1); + case SCH_TABLE_STATS: + case SCH_INDEX_STATS: case SCH_OPEN_TABLES: case SCH_VARIABLES: case SCH_STATUS: @@ -2218,6 +2232,10 @@ mysql_execute_command(THD *thd) case SQLCOM_SHOW_COLLATIONS: case SQLCOM_SHOW_STORAGE_ENGINES: case SQLCOM_SHOW_PROFILE: + case SQLCOM_SHOW_CLIENT_STATS: + case SQLCOM_SHOW_USER_STATS: + case SQLCOM_SHOW_TABLE_STATS: + case SQLCOM_SHOW_INDEX_STATS: case SQLCOM_SELECT: thd->status_var.last_query_cost= 0.0; if (all_tables) @@ -5059,6 +5077,10 @@ static bool execute_sqlcom_select(THD *t delete result; } } + /* Count number of empty select queries */ + if (!thd->sent_row_count) + status_var_increment(thd->status_var.empty_queries); + status_var_add(thd->status_var.rows_sent, thd->sent_row_count); return res; } @@ -5220,6 +5242,7 @@ check_access(THD *thd, ulong want_access if (!no_errors) { const char *db_name= db ? db : thd->db; + status_var_increment(thd->status_var.access_denied_errors); my_error(ER_DBACCESS_DENIED_ERROR, MYF(0), sctx->priv_user, sctx->priv_host, db_name); } @@ -5252,12 +5275,15 @@ check_access(THD *thd, ulong want_access { // We can never grant this DBUG_PRINT("error",("No possible access")); if (!no_errors) + { + status_var_increment(thd->status_var.access_denied_errors); my_error(ER_ACCESS_DENIED_ERROR, MYF(0), sctx->priv_user, sctx->priv_host, (thd->password ? ER(ER_YES) : ER(ER_NO))); /* purecov: tested */ + } DBUG_RETURN(TRUE); /* purecov: tested */ } @@ -5283,11 +5309,14 @@ check_access(THD *thd, ulong want_access DBUG_PRINT("error",("Access denied")); if (!no_errors) + { + status_var_increment(thd->status_var.access_denied_errors); my_error(ER_DBACCESS_DENIED_ERROR, MYF(0), sctx->priv_user, sctx->priv_host, (db ? db : (thd->db ? thd->db : "unknown"))); /* purecov: tested */ + } DBUG_RETURN(TRUE); /* purecov: tested */ } @@ -5316,6 +5345,7 @@ static bool check_show_access(THD *thd, if (!thd->col_access && check_grant_db(thd, dst_db_name)) { + status_var_increment(thd->status_var.access_denied_errors); my_error(ER_DBACCESS_DENIED_ERROR, MYF(0), thd->security_ctx->priv_user, thd->security_ctx->priv_host, @@ -5378,14 +5408,14 @@ check_table_access(THD *thd, ulong want_ { TABLE_LIST *org_tables= tables; TABLE_LIST *first_not_own_table= thd->lex->first_not_own_table(); - uint i= 0; Security_context *sctx= thd->security_ctx, *backup_ctx= thd->security_ctx; + uint i; /* The check that first_not_own_table is not reached is for the case when the given table list refers to the list for prelocking (contains tables of other queries). For simple queries first_not_own_table is 0. */ - for (; i < number && tables != first_not_own_table; + for (i=0; i < number && tables != first_not_own_table; tables= tables->next_global, i++) { if (tables->security_ctx) @@ -5397,9 +5427,12 @@ check_table_access(THD *thd, ulong want_ (want_access & ~(SELECT_ACL | EXTRA_ACL | FILE_ACL))) { if (!no_errors) + { + status_var_increment(thd->status_var.access_denied_errors); my_error(ER_DBACCESS_DENIED_ERROR, MYF(0), sctx->priv_user, sctx->priv_host, INFORMATION_SCHEMA_NAME.str); + } return TRUE; } /* @@ -5563,6 +5596,7 @@ bool check_global_access(THD *thd, ulong return 0; get_privilege_desc(command, sizeof(command), want_access); my_error(ER_SPECIFIC_ACCESS_DENIED_ERROR, MYF(0), command); + status_var_increment(thd->status_var.access_denied_errors); return 1; #else return 0; @@ -5666,7 +5700,7 @@ bool my_yyoverflow(short **yyss, YYSTYPE Call it after we use THD for queries, not before. */ -void mysql_reset_thd_for_next_command(THD *thd) +void mysql_reset_thd_for_next_command(THD *thd, my_bool calculate_userstat) { DBUG_ENTER("mysql_reset_thd_for_next_command"); DBUG_ASSERT(!thd->spcont); /* not for substatements of routines */ @@ -5711,6 +5745,15 @@ void mysql_reset_thd_for_next_command(TH thd->total_warn_count=0; // Warnings for this query thd->rand_used= 0; thd->sent_row_count= thd->examined_row_count= 0; + + /* Copy data for user stats */ + if ((thd->userstat_running= calculate_userstat)) + { + thd->start_cpu_time= my_getcputime(); + memcpy(&thd->org_status_var, &thd->status_var, sizeof(thd->status_var)); + thd->select_commands= thd->update_commands= thd->other_commands= 0; + } + thd->query_plan_flags= QPLAN_INIT; thd->query_plan_fsort_passes= 0; @@ -5909,7 +5952,6 @@ void mysql_parse(THD *thd, const char *i const char ** found_semicolon) { DBUG_ENTER("mysql_parse"); - DBUG_EXECUTE_IF("parser_debug", turn_parser_debug_on();); /* @@ -5929,7 +5971,7 @@ void mysql_parse(THD *thd, const char *i FIXME: cleanup the dependencies in the code to simplify this. */ lex_start(thd); - mysql_reset_thd_for_next_command(thd); + mysql_reset_thd_for_next_command(thd, opt_userstat_running); if (query_cache_send_result_to_client(thd, (char*) inBuf, length) <= 0) { @@ -6001,10 +6043,11 @@ void mysql_parse(THD *thd, const char *i } else { + /* Update statistics for getting the query from the cache */ + thd->lex->sql_command= SQLCOM_SELECT; /* There are no multi queries in the cache. */ *found_semicolon= NULL; } - DBUG_VOID_RETURN; } @@ -6028,7 +6071,7 @@ bool mysql_test_parse_for_slave(THD *thd Parser_state parser_state(thd, inBuf, length); lex_start(thd); - mysql_reset_thd_for_next_command(thd); + mysql_reset_thd_for_next_command(thd, 0); if (!parse_sql(thd, & parser_state, NULL) && all_tables_not_ok(thd,(TABLE_LIST*) lex->select_lex.table_list.first)) @@ -6867,6 +6910,13 @@ bool reload_acl_and_cache(THD *thd, ulon if (flush_error_log()) result=1; } + if (((options & (REFRESH_SLOW_QUERY_LOG | REFRESH_LOG)) == + REFRESH_SLOW_QUERY_LOG)) + { + /* We are only flushing slow query log */ + logger.flush_slow_log(thd); + } + #ifdef HAVE_QUERY_CACHE if (options & REFRESH_QUERY_CACHE_FREE) { @@ -6949,26 +6999,55 @@ bool reload_acl_and_cache(THD *thd, ulon } #endif #ifdef OPENSSL - if (options & REFRESH_DES_KEY_FILE) - { - if (des_key_file && load_des_key_file(des_key_file)) - result= 1; - } + if (options & REFRESH_DES_KEY_FILE) + { + if (des_key_file && load_des_key_file(des_key_file)) + result= 1; + } #endif #ifdef HAVE_REPLICATION - if (options & REFRESH_SLAVE) - { - tmp_write_to_binlog= 0; - pthread_mutex_lock(&LOCK_active_mi); - if (reset_slave(thd, active_mi)) - result=1; - pthread_mutex_unlock(&LOCK_active_mi); - } -#endif - if (options & REFRESH_USER_RESOURCES) - reset_mqh((LEX_USER *) NULL, 0); /* purecov: inspected */ - *write_to_binlog= tmp_write_to_binlog; - return result; + if (options & REFRESH_SLAVE) + { + tmp_write_to_binlog= 0; + pthread_mutex_lock(&LOCK_active_mi); + if (reset_slave(thd, active_mi)) + result=1; + pthread_mutex_unlock(&LOCK_active_mi); + } +#endif + if (options & REFRESH_USER_RESOURCES) + reset_mqh((LEX_USER *) NULL, 0); /* purecov: inspected */ + if (options & REFRESH_TABLE_STATS) + { + pthread_mutex_lock(&LOCK_global_table_stats); + free_global_table_stats(); + init_global_table_stats(); + pthread_mutex_unlock(&LOCK_global_table_stats); + } + if (options & REFRESH_INDEX_STATS) + { + pthread_mutex_lock(&LOCK_global_index_stats); + free_global_index_stats(); + init_global_index_stats(); + pthread_mutex_unlock(&LOCK_global_index_stats); + } + if (options & (REFRESH_USER_STATS | REFRESH_CLIENT_STATS)) + { + pthread_mutex_lock(&LOCK_global_user_client_stats); + if (options & REFRESH_USER_STATS) + { + free_global_user_stats(); + init_global_user_stats(); + } + if (options & REFRESH_CLIENT_STATS) + { + free_global_client_stats(); + init_global_client_stats(); + } + pthread_mutex_unlock(&LOCK_global_user_client_stats); + } + *write_to_binlog= tmp_write_to_binlog; + return result; } @@ -7004,7 +7083,6 @@ uint kill_one_thread(THD *thd, ulong id, VOID(pthread_mutex_unlock(&LOCK_thread_count)); if (tmp) { - /* If we're SUPER, we can KILL anything, including system-threads. No further checks. === modified file 'sql/sql_plugin.cc' --- a/sql/sql_plugin.cc 2009-10-01 21:27:39 +0000 +++ b/sql/sql_plugin.cc 2009-10-19 17:14:48 +0000 @@ -1790,10 +1790,10 @@ bool mysql_uninstall_plugin(THD *thd, co table->use_all_columns(); table->field[0]->store(name->str, name->length, system_charset_info); - if (! table->file->index_read_idx_map(table->record[0], 0, - (uchar *)table->field[0]->ptr, - HA_WHOLE_KEY, - HA_READ_KEY_EXACT)) + if (! table->file->ha_index_read_idx_map(table->record[0], 0, + (uchar *)table->field[0]->ptr, + HA_WHOLE_KEY, + HA_READ_KEY_EXACT)) { int error; /* === modified file 'sql/sql_prepare.cc' --- a/sql/sql_prepare.cc 2009-09-07 20:50:10 +0000 +++ b/sql/sql_prepare.cc 2009-10-19 17:14:48 +0000 @@ -2067,14 +2067,13 @@ void mysqld_stmt_prepare(THD *thd, const Prepared_statement *stmt; bool error; DBUG_ENTER("mysqld_stmt_prepare"); - DBUG_PRINT("prep_query", ("%s", packet)); /* First of all clear possible warnings from the previous command */ - mysql_reset_thd_for_next_command(thd); + mysql_reset_thd_for_next_command(thd, opt_userstat_running); if (! (stmt= new Prepared_statement(thd))) - DBUG_VOID_RETURN; /* out of memory: error is set in Sql_alloc */ + goto end; /* out of memory: error is set in Sql_alloc */ if (thd->stmt_map.insert(thd, stmt)) { @@ -2082,7 +2081,7 @@ void mysqld_stmt_prepare(THD *thd, const The error is set in the insert. The statement itself will be also deleted there (this is how the hash works). */ - DBUG_VOID_RETURN; + goto end; } /* Reset warnings from previous command */ @@ -2109,6 +2108,7 @@ void mysqld_stmt_prepare(THD *thd, const thd->protocol= save_protocol; /* check_prepared_statemnt sends the metadata packet in case of success */ +end: DBUG_VOID_RETURN; } @@ -2450,7 +2450,7 @@ void mysqld_stmt_execute(THD *thd, char packet+= 9; /* stmt_id + 5 bytes of flags */ /* First of all clear possible warnings from the previous command */ - mysql_reset_thd_for_next_command(thd); + mysql_reset_thd_for_next_command(thd, opt_userstat_running); if (!(stmt= find_prepared_statement(thd, stmt_id))) { @@ -2549,7 +2549,8 @@ void mysqld_stmt_fetch(THD *thd, char *p DBUG_ENTER("mysqld_stmt_fetch"); /* First of all clear possible warnings from the previous command */ - mysql_reset_thd_for_next_command(thd); + mysql_reset_thd_for_next_command(thd, opt_userstat_running); + status_var_increment(thd->status_var.com_stmt_fetch); if (!(stmt= find_prepared_statement(thd, stmt_id))) { @@ -2615,7 +2616,7 @@ void mysqld_stmt_reset(THD *thd, char *p DBUG_ENTER("mysqld_stmt_reset"); /* First of all clear possible warnings from the previous command */ - mysql_reset_thd_for_next_command(thd); + mysql_reset_thd_for_next_command(thd, opt_userstat_running); status_var_increment(thd->status_var.com_stmt_reset); if (!(stmt= find_prepared_statement(thd, stmt_id))) === modified file 'sql/sql_select.cc' --- a/sql/sql_select.cc 2009-09-15 10:46:35 +0000 +++ b/sql/sql_select.cc 2009-10-19 17:14:48 +0000 @@ -10603,8 +10603,9 @@ error: static bool open_tmp_table(TABLE *table) { int error; - if ((error=table->file->ha_open(table, table->s->table_name.str,O_RDWR, - HA_OPEN_TMP_TABLE | HA_OPEN_INTERNAL_TABLE))) + if ((error= table->file->ha_open(table, table->s->table_name.str, O_RDWR, + HA_OPEN_TMP_TABLE | + HA_OPEN_INTERNAL_TABLE))) { table->file->print_error(error,MYF(0)); /* purecov: inspected */ table->db_stat=0; @@ -10949,7 +10950,7 @@ create_internal_tmp_table_from_heap2(THD is safe as this is a temporary MyISAM table without timestamp/autoincrement or partitioning. */ - while (!table->file->rnd_next(new_table.record[1])) + while (!table->file->ha_rnd_next(new_table.record[1])) { write_err= new_table.file->ha_write_row(new_table.record[1]); DBUG_EXECUTE_IF("raise_error", write_err= HA_ERR_FOUND_DUPP_KEY ;); @@ -11746,10 +11747,10 @@ int safe_index_read(JOIN_TAB *tab) { int error; TABLE *table= tab->table; - if ((error=table->file->index_read_map(table->record[0], - tab->ref.key_buff, - make_prev_keypart_map(tab->ref.key_parts), - HA_READ_KEY_EXACT))) + if ((error= table->file->ha_index_read_map(table->record[0], + tab->ref.key_buff, + make_prev_keypart_map(tab->ref.key_parts), + HA_READ_KEY_EXACT))) return report_error(table, error); return 0; } @@ -11858,8 +11859,8 @@ join_read_system(JOIN_TAB *tab) int error; if (table->status & STATUS_GARBAGE) // If first read { - if ((error=table->file->read_first_row(table->record[0], - table->s->primary_key))) + if ((error= table->file->ha_read_first_row(table->record[0], + table->s->primary_key))) { if (error != HA_ERR_END_OF_FILE) return report_error(table, error); @@ -11901,10 +11902,10 @@ join_read_const(JOIN_TAB *tab) error=HA_ERR_KEY_NOT_FOUND; else { - error=table->file->index_read_idx_map(table->record[0],tab->ref.key, - (uchar*) tab->ref.key_buff, - make_prev_keypart_map(tab->ref.key_parts), - HA_READ_KEY_EXACT); + error= table->file->ha_index_read_idx_map(table->record[0],tab->ref.key, + (uchar*) tab->ref.key_buff, + make_prev_keypart_map(tab->ref.key_parts), + HA_READ_KEY_EXACT); } if (error) { @@ -11949,10 +11950,10 @@ join_read_key(JOIN_TAB *tab) table->status=STATUS_NOT_FOUND; return -1; } - error=table->file->index_read_map(table->record[0], - tab->ref.key_buff, - make_prev_keypart_map(tab->ref.key_parts), - HA_READ_KEY_EXACT); + error= table->file->ha_index_read_map(table->record[0], + tab->ref.key_buff, + make_prev_keypart_map(tab->ref.key_parts), + HA_READ_KEY_EXACT); if (error && error != HA_ERR_KEY_NOT_FOUND && error != HA_ERR_END_OF_FILE) return report_error(table, error); } @@ -12005,10 +12006,10 @@ join_read_always_key(JOIN_TAB *tab) if (cp_buffer_from_ref(tab->join->thd, table, &tab->ref)) return -1; - if ((error=table->file->index_read_map(table->record[0], - tab->ref.key_buff, - make_prev_keypart_map(tab->ref.key_parts), - HA_READ_KEY_EXACT))) + if ((error= table->file->ha_index_read_map(table->record[0], + tab->ref.key_buff, + make_prev_keypart_map(tab->ref.key_parts), + HA_READ_KEY_EXACT))) { if (error != HA_ERR_KEY_NOT_FOUND && error != HA_ERR_END_OF_FILE) return report_error(table, error); @@ -12039,9 +12040,9 @@ join_read_last_key(JOIN_TAB *tab) } if (cp_buffer_from_ref(tab->join->thd, table, &tab->ref)) return -1; - if ((error=table->file->index_read_last_map(table->record[0], - tab->ref.key_buff, - make_prev_keypart_map(tab->ref.key_parts)))) + if ((error= table->file->ha_index_read_last_map(table->record[0], + tab->ref.key_buff, + make_prev_keypart_map(tab->ref.key_parts)))) { if (error != HA_ERR_KEY_NOT_FOUND && error != HA_ERR_END_OF_FILE) return report_error(table, error); @@ -12066,9 +12067,9 @@ join_read_next_same(READ_RECORD *info) TABLE *table= info->table; JOIN_TAB *tab=table->reginfo.join_tab; - if ((error=table->file->index_next_same(table->record[0], - tab->ref.key_buff, - tab->ref.key_length))) + if ((error= table->file->ha_index_next_same(table->record[0], + tab->ref.key_buff, + tab->ref.key_length))) { if (error != HA_ERR_END_OF_FILE) return report_error(table, error); @@ -12086,7 +12087,7 @@ join_read_prev_same(READ_RECORD *info) TABLE *table= info->table; JOIN_TAB *tab=table->reginfo.join_tab; - if ((error=table->file->index_prev(table->record[0]))) + if ((error= table->file->ha_index_prev(table->record[0]))) return report_error(table, error); if (key_cmp_if_same(table, tab->ref.key_buff, tab->ref.key, tab->ref.key_length)) @@ -12158,7 +12159,7 @@ join_read_first(JOIN_TAB *tab) error= table->file->ha_index_init(tab->index, tab->sorted); if (!error) error= table->file->prepare_index_scan(); - if (error || (error=tab->table->file->index_first(tab->table->record[0]))) + if (error || (error=tab->table->file->ha_index_first(tab->table->record[0]))) { if (error != HA_ERR_KEY_NOT_FOUND && error != HA_ERR_END_OF_FILE) report_error(table, error); @@ -12172,7 +12173,7 @@ static int join_read_next(READ_RECORD *info) { int error; - if ((error=info->file->index_next(info->record))) + if ((error= info->file->ha_index_next(info->record))) return report_error(info->table, error); return 0; } @@ -12199,7 +12200,7 @@ join_read_last(JOIN_TAB *tab) error= table->file->ha_index_init(tab->index, 1); if (!error) error= table->file->prepare_index_scan(); - if (error || (error= tab->table->file->index_last(tab->table->record[0]))) + if (error || (error= tab->table->file->ha_index_last(tab->table->record[0]))) return report_error(table, error); return 0; } @@ -12209,7 +12210,7 @@ static int join_read_prev(READ_RECORD *info) { int error; - if ((error= info->file->index_prev(info->record))) + if ((error= info->file->ha_index_prev(info->record))) return report_error(info->table, error); return 0; } @@ -12234,7 +12235,7 @@ join_ft_read_first(JOIN_TAB *tab) #endif table->file->ft_init(); - if ((error= table->file->ft_read(table->record[0]))) + if ((error= table->file->ha_ft_read(table->record[0]))) return report_error(table, error); return 0; } @@ -12243,7 +12244,7 @@ static int join_ft_read_next(READ_RECORD *info) { int error; - if ((error= info->file->ft_read(info->table->record[0]))) + if ((error= info->file->ha_ft_read(info->table->record[0]))) return report_error(info->table, error); return 0; } @@ -12535,7 +12536,7 @@ end_write(JOIN *join, JOIN_TAB *join_tab { int error; join->found_records++; - if ((error=table->file->ha_write_row(table->record[0]))) + if ((error= table->file->ha_write_row(table->record[0]))) { if (!table->file->is_fatal_error(error, HA_CHECK_DUP)) goto end; @@ -12590,15 +12591,15 @@ end_update(JOIN *join, JOIN_TAB *join_ta if (item->maybe_null) group->buff[-1]= (char) group->field->is_null(); } - if (!table->file->index_read_map(table->record[1], - join->tmp_table_param.group_buff, - HA_WHOLE_KEY, - HA_READ_KEY_EXACT)) + if (!table->file->ha_index_read_map(table->record[1], + join->tmp_table_param.group_buff, + HA_WHOLE_KEY, + HA_READ_KEY_EXACT)) { /* Update old record */ restore_record(table,record[1]); update_tmptable_sum_func(join->sum_funcs,table); - if ((error=table->file->ha_update_row(table->record[1], - table->record[0]))) + if ((error= table->file->ha_update_row(table->record[1], + table->record[0]))) { table->file->print_error(error,MYF(0)); /* purecov: inspected */ DBUG_RETURN(NESTED_LOOP_ERROR); /* purecov: inspected */ @@ -12621,7 +12622,7 @@ end_update(JOIN *join, JOIN_TAB *join_ta } init_tmptable_sum_functions(join->sum_funcs); copy_funcs(join->tmp_table_param.items_to_copy); - if ((error=table->file->ha_write_row(table->record[0]))) + if ((error= table->file->ha_write_row(table->record[0]))) { if (create_internal_tmp_table_from_heap(join->thd, table, &join->tmp_table_param, @@ -12662,7 +12663,7 @@ end_unique_update(JOIN *join, JOIN_TAB * copy_fields(&join->tmp_table_param); // Groups are copied twice. copy_funcs(join->tmp_table_param.items_to_copy); - if (!(error=table->file->ha_write_row(table->record[0]))) + if (!(error= table->file->ha_write_row(table->record[0]))) join->send_records++; // New group else { @@ -12671,15 +12672,15 @@ end_unique_update(JOIN *join, JOIN_TAB * table->file->print_error(error,MYF(0)); /* purecov: inspected */ DBUG_RETURN(NESTED_LOOP_ERROR); /* purecov: inspected */ } - if (table->file->rnd_pos(table->record[1],table->file->dup_ref)) + if (table->file->ha_rnd_pos(table->record[1],table->file->dup_ref)) { table->file->print_error(error,MYF(0)); /* purecov: inspected */ DBUG_RETURN(NESTED_LOOP_ERROR); /* purecov: inspected */ } restore_record(table,record[1]); update_tmptable_sum_func(join->sum_funcs,table); - if ((error=table->file->ha_update_row(table->record[1], - table->record[0]))) + if ((error= table->file->ha_update_row(table->record[1], + table->record[0]))) { table->file->print_error(error,MYF(0)); /* purecov: inspected */ DBUG_RETURN(NESTED_LOOP_ERROR); /* purecov: inspected */ @@ -14016,7 +14017,7 @@ static int remove_dup_with_compare(THD * new_record=(char*) table->record[1]+offset; file->ha_rnd_init(1); - error=file->rnd_next(record); + error= file->ha_rnd_next(record); for (;;) { if (thd->killed) @@ -14035,9 +14036,9 @@ static int remove_dup_with_compare(THD * } if (having && !having->val_int()) { - if ((error=file->ha_delete_row(record))) + if ((error= file->ha_delete_row(record))) goto err; - error=file->rnd_next(record); + error= file->ha_rnd_next(record); continue; } if (copy_blobs(first_field)) @@ -14052,7 +14053,7 @@ static int remove_dup_with_compare(THD * bool found=0; for (;;) { - if ((error=file->rnd_next(record))) + if ((error= file->ha_rnd_next(record))) { if (error == HA_ERR_RECORD_DELETED) continue; @@ -14062,7 +14063,7 @@ static int remove_dup_with_compare(THD * } if (compare_record(table, first_field) == 0) { - if ((error=file->ha_delete_row(record))) + if ((error= file->ha_delete_row(record))) goto err; } else if (!found) @@ -14152,7 +14153,7 @@ static int remove_dup_with_hash_index(TH error=0; goto err; } - if ((error=file->rnd_next(record))) + if ((error= file->ha_rnd_next(record))) { if (error == HA_ERR_RECORD_DELETED) continue; @@ -14162,7 +14163,7 @@ static int remove_dup_with_hash_index(TH } if (having && !having->val_int()) { - if ((error=file->ha_delete_row(record))) + if ((error= file->ha_delete_row(record))) goto err; continue; } @@ -14179,7 +14180,7 @@ static int remove_dup_with_hash_index(TH if (hash_search(&hash, org_key_pos, key_length)) { /* Duplicated found ; Remove the row */ - if ((error=file->ha_delete_row(record))) + if ((error= file->ha_delete_row(record))) goto err; } else === modified file 'sql/sql_servers.cc' --- a/sql/sql_servers.cc 2009-03-20 14:27:53 +0000 +++ b/sql/sql_servers.cc 2009-10-19 17:14:48 +0000 @@ -520,10 +520,10 @@ int insert_server_record(TABLE *table, F system_charset_info); /* read index until record is that specified in server_name */ - if ((error= table->file->index_read_idx_map(table->record[0], 0, - (uchar *)table->field[0]->ptr, - HA_WHOLE_KEY, - HA_READ_KEY_EXACT))) + if ((error= table->file->ha_index_read_idx_map(table->record[0], 0, + (uchar *)table->field[0]->ptr, + HA_WHOLE_KEY, + HA_READ_KEY_EXACT))) { /* if not found, err */ if (error != HA_ERR_KEY_NOT_FOUND && error != HA_ERR_END_OF_FILE) @@ -863,10 +863,10 @@ update_server_record(TABLE *table, FOREI server->server_name_length, system_charset_info); - if ((error= table->file->index_read_idx_map(table->record[0], 0, - (uchar *)table->field[0]->ptr, - ~(longlong)0, - HA_READ_KEY_EXACT))) + if ((error= table->file->ha_index_read_idx_map(table->record[0], 0, + (uchar *)table->field[0]->ptr, + ~(longlong)0, + HA_READ_KEY_EXACT))) { if (error != HA_ERR_KEY_NOT_FOUND && error != HA_ERR_END_OF_FILE) table->file->print_error(error, MYF(0)); @@ -920,10 +920,10 @@ delete_server_record(TABLE *table, /* set the field that's the PK to the value we're looking for */ table->field[0]->store(server_name, server_name_length, system_charset_info); - if ((error= table->file->index_read_idx_map(table->record[0], 0, - (uchar *)table->field[0]->ptr, - HA_WHOLE_KEY, - HA_READ_KEY_EXACT))) + if ((error= table->file->ha_index_read_idx_map(table->record[0], 0, + (uchar *)table->field[0]->ptr, + HA_WHOLE_KEY, + HA_READ_KEY_EXACT))) { if (error != HA_ERR_KEY_NOT_FOUND && error != HA_ERR_END_OF_FILE) table->file->print_error(error, MYF(0)); === modified file 'sql/sql_show.cc' --- a/sql/sql_show.cc 2009-09-23 11:03:47 +0000 +++ b/sql/sql_show.cc 2009-10-19 17:14:48 +0000 @@ -714,6 +714,7 @@ bool mysqld_show_create_db(THD *thd, cha sctx->master_access); if (!(db_access & DB_ACLS) && check_grant_db(thd,dbname)) { + status_var_increment(thd->status_var.access_denied_errors); my_error(ER_DBACCESS_DENIED_ERROR, MYF(0), sctx->priv_user, sctx->host_or_ip, dbname); general_log_print(thd,COM_INIT_DB,ER(ER_DBACCESS_DENIED_ERROR), @@ -2100,11 +2101,6 @@ void remove_status_vars(SHOW_VAR *list) } } -inline void make_upper(char *buf) -{ - for (; *buf; buf++) - *buf= my_toupper(system_charset_info, *buf); -} static bool show_status_array(THD *thd, const char *wild, SHOW_VAR *variables, @@ -2143,7 +2139,7 @@ static bool show_status_array(THD *thd, strnmov(prefix_end, variables->name, len); name_buffer[sizeof(name_buffer)-1]=0; /* Safety */ if (ucase_names) - make_upper(name_buffer); + my_caseup_str(system_charset_info, name_buffer); restore_record(table, s->default_values); table->field[0]->store(name_buffer, strlen(name_buffer), @@ -2270,6 +2266,323 @@ end: DBUG_RETURN(res); } +#ifdef COMPLEAT_PATCH_NOT_ADDED_YET +/* + Aggregate values for mapped_user entries by their role. + + SYNOPSIS + aggregate_user_stats + all_user_stats - input to aggregate + agg_user_stats - returns aggregated values + + RETURN + 0 - OK + 1 - error +*/ + +static int aggregate_user_stats(HASH *all_user_stats, HASH *agg_user_stats) +{ + DBUG_ENTER("aggregate_user_stats"); + if (hash_init(agg_user_stats, system_charset_info, + max(all_user_stats->records, 1), + 0, 0, (hash_get_key)get_key_user_stats, + (hash_free_key)free_user_stats, 0)) + { + sql_print_error("Malloc in aggregate_user_stats failed"); + DBUG_RETURN(1); + } + + for (uint i= 0; i < all_user_stats->records; i++) + { + USER_STATS *user= (USER_STATS*)hash_element(all_user_stats, i); + USER_STATS *agg_user; + uint name_length= strlen(user->priv_user); + + if (!(agg_user= (USER_STATS*) hash_search(agg_user_stats, + (uchar*)user->priv_user, + name_length))) + { + // First entry for this role. + if (!(agg_user= (USER_STATS*) my_malloc(sizeof(USER_STATS), + MYF(MY_WME | MY_ZEROFILL)))) + { + sql_print_error("Malloc in aggregate_user_stats failed"); + DBUG_RETURN(1); + } + + init_user_stats(agg_user, user->priv_user, name_length, + user->priv_user, + user->total_connections, user->concurrent_connections, + user->connected_time, user->busy_time, user->cpu_time, + user->bytes_received, user->bytes_sent, + user->binlog_bytes_written, + user->rows_sent, user->rows_read, + user->rows_inserted, user->rows_deleted, + user->rows_updated, + user->select_commands, user->update_commands, + user->other_commands, + user->commit_trans, user->rollback_trans, + user->denied_connections, user->lost_connections, + user->access_denied_errors, user->empty_queries); + + if (my_hash_insert(agg_user_stats, (uchar*) agg_user)) + { + /* Out of memory */ + my_free(agg_user, 0); + sql_print_error("Malloc in aggregate_user_stats failed"); + DBUG_RETURN(1); + } + } + else + { + /* Aggregate with existing values for this role. */ + add_user_stats(agg_user, + user->total_connections, user->concurrent_connections, + user->connected_time, user->busy_time, user->cpu_time, + user->bytes_received, user->bytes_sent, + user->binlog_bytes_written, + user->rows_sent, user->rows_read, + user->rows_inserted, user->rows_deleted, + user->rows_updated, + user->select_commands, user->update_commands, + user->other_commands, + user->commit_trans, user->rollback_trans, + user->denied_connections, user->lost_connections, + user->access_denied_errors, user->empty_queries); + } + } + DBUG_PRINT("exit", ("aggregated %lu input into %lu output entries", + all_user_stats->records, agg_user_stats->records)); + DBUG_RETURN(0); +} +#endif + +/* + Write result to network for SHOW USER_STATISTICS + + SYNOPSIS + send_user_stats + all_user_stats - values to return + table - I_S table + + RETURN + 0 - OK + 1 - error +*/ + +int send_user_stats(THD* thd, HASH *all_user_stats, TABLE *table) +{ + DBUG_ENTER("send_user_stats"); + + for (uint i= 0; i < all_user_stats->records; i++) + { + uint j= 0; + USER_STATS *user_stats= (USER_STATS*) hash_element(all_user_stats, i); + + table->field[j++]->store(user_stats->user, user_stats->user_name_length, + system_charset_info); + table->field[j++]->store((longlong)user_stats->total_connections,TRUE); + table->field[j++]->store((longlong)user_stats->concurrent_connections); + table->field[j++]->store((longlong)user_stats->connected_time); + table->field[j++]->store((double)user_stats->busy_time); + table->field[j++]->store((double)user_stats->cpu_time); + table->field[j++]->store((longlong)user_stats->bytes_received, TRUE); + table->field[j++]->store((longlong)user_stats->bytes_sent, TRUE); + table->field[j++]->store((longlong)user_stats->binlog_bytes_written, TRUE); + table->field[j++]->store((longlong)user_stats->rows_read, TRUE); + table->field[j++]->store((longlong)user_stats->rows_sent, TRUE); + table->field[j++]->store((longlong)user_stats->rows_deleted, TRUE); + table->field[j++]->store((longlong)user_stats->rows_inserted, TRUE); + table->field[j++]->store((longlong)user_stats->rows_updated, TRUE); + table->field[j++]->store((longlong)user_stats->select_commands, TRUE); + table->field[j++]->store((longlong)user_stats->update_commands, TRUE); + table->field[j++]->store((longlong)user_stats->other_commands, TRUE); + table->field[j++]->store((longlong)user_stats->commit_trans, TRUE); + table->field[j++]->store((longlong)user_stats->rollback_trans, TRUE); + table->field[j++]->store((longlong)user_stats->denied_connections, TRUE); + table->field[j++]->store((longlong)user_stats->lost_connections, TRUE); + table->field[j++]->store((longlong)user_stats->access_denied_errors, TRUE); + table->field[j++]->store((longlong)user_stats->empty_queries, TRUE); + if (schema_table_store_record(thd, table)) + { + DBUG_PRINT("error", ("store record error")); + DBUG_RETURN(1); + } + } + DBUG_RETURN(0); +} + +/* + Process SHOW USER_STATISTICS + + SYNOPSIS + mysqld_show_user_stats + thd - current thread + wild - limit results to the entry for this user + with_roles - when true, display role for mapped users + + RETURN + 0 - OK + 1 - error +*/ + +int fill_schema_user_stats(THD* thd, TABLE_LIST* tables, COND* cond) +{ + TABLE *table= tables->table; + int result; + DBUG_ENTER("fill_schema_user_stats"); + + if (check_global_access(thd, SUPER_ACL | PROCESS_ACL)) + DBUG_RETURN(1); + + /* + Iterates through all the global stats and sends them to the client. + Pattern matching on the client IP is supported. + */ + + pthread_mutex_lock(&LOCK_global_user_client_stats); + result= send_user_stats(thd, &global_user_stats, table) != 0; + pthread_mutex_unlock(&LOCK_global_user_client_stats); + + DBUG_PRINT("exit", ("result: %d", result)); + DBUG_RETURN(result); +} + +/* + Process SHOW CLIENT_STATISTICS + + SYNOPSIS + mysqld_show_client_stats + thd - current thread + wild - limit results to the entry for this client + + RETURN + 0 - OK + 1 - error +*/ + +int fill_schema_client_stats(THD* thd, TABLE_LIST* tables, COND* cond) +{ + TABLE *table= tables->table; + int result; + DBUG_ENTER("fill_schema_client_stats"); + + if (check_global_access(thd, SUPER_ACL | PROCESS_ACL)) + DBUG_RETURN(1); + + /* + Iterates through all the global stats and sends them to the client. + Pattern matching on the client IP is supported. + */ + + pthread_mutex_lock(&LOCK_global_user_client_stats); + result= send_user_stats(thd, &global_client_stats, table) != 0; + pthread_mutex_unlock(&LOCK_global_user_client_stats); + + DBUG_PRINT("exit", ("result: %d", result)); + DBUG_RETURN(result); +} + + +/* Fill information schema table with table statistics */ + +int fill_schema_table_stats(THD *thd, TABLE_LIST *tables, COND *cond) +{ + TABLE *table= tables->table; + DBUG_ENTER("fill_schema_table_stats"); + + pthread_mutex_lock(&LOCK_global_table_stats); + for (uint i= 0; i < global_table_stats.records; i++) + { + char *end_of_schema; + TABLE_STATS *table_stats= + (TABLE_STATS*)hash_element(&global_table_stats, i); + TABLE_LIST tmp_table; + size_t schema_length, table_name_length; + + end_of_schema= strend(table_stats->table); + schema_length= (size_t) (end_of_schema - table_stats->table); + table_name_length= strlen(table_stats->table + schema_length + 1); + + bzero((char*) &tmp_table,sizeof(tmp_table)); + tmp_table.db= table_stats->table; + tmp_table.table_name= end_of_schema+1; + tmp_table.grant.privilege= 0; + if (check_access(thd, SELECT_ACL | EXTRA_ACL, tmp_table.db, + &tmp_table.grant.privilege, 0, 0, + is_schema_db(tmp_table.db)) || + check_grant(thd, SELECT_ACL, &tmp_table, 1, UINT_MAX, + 1)) + continue; + + table->field[0]->store(table_stats->table, schema_length, + system_charset_info); + table->field[1]->store(table_stats->table + schema_length+1, + table_name_length, system_charset_info); + table->field[2]->store((longlong)table_stats->rows_read, TRUE); + table->field[3]->store((longlong)table_stats->rows_changed, TRUE); + table->field[4]->store((longlong)table_stats->rows_changed_x_indexes, + TRUE); + if (schema_table_store_record(thd, table)) + { + VOID(pthread_mutex_unlock(&LOCK_global_table_stats)); + DBUG_RETURN(1); + } + } + pthread_mutex_unlock(&LOCK_global_table_stats); + DBUG_RETURN(0); +} + + +/* Fill information schema table with index statistics */ + +int fill_schema_index_stats(THD *thd, TABLE_LIST *tables, COND *cond) +{ + TABLE *table= tables->table; + DBUG_ENTER("fill_schema_index_stats"); + + pthread_mutex_lock(&LOCK_global_index_stats); + for (uint i= 0; i < global_index_stats.records; i++) + { + INDEX_STATS *index_stats = + (INDEX_STATS*) hash_element(&global_index_stats, i); + TABLE_LIST tmp_table; + char *index_name; + size_t schema_name_length, table_name_length, index_name_length; + + bzero((char*) &tmp_table,sizeof(tmp_table)); + tmp_table.db= index_stats->index; + tmp_table.table_name= strend(index_stats->index)+1; + tmp_table.grant.privilege= 0; + if (check_access(thd, SELECT_ACL | EXTRA_ACL, tmp_table.db, + &tmp_table.grant.privilege, 0, 0, + is_schema_db(tmp_table.db)) || + check_grant(thd, SELECT_ACL, &tmp_table, 1, UINT_MAX, 1)) + continue; + + index_name= strend(tmp_table.table_name)+1; + schema_name_length= (tmp_table.table_name - index_stats->index) -1; + table_name_length= (index_name - tmp_table.table_name)-1; + index_name_length= (index_stats->index_name_length - schema_name_length - + table_name_length - 3); + + table->field[0]->store(tmp_table.db, schema_name_length, + system_charset_info); + table->field[1]->store(tmp_table.table_name, table_name_length, + system_charset_info); + table->field[2]->store(index_name, index_name_length, system_charset_info); + table->field[3]->store((longlong)index_stats->rows_read, TRUE); + + if (schema_table_store_record(thd, table)) + { + VOID(pthread_mutex_unlock(&LOCK_global_index_stats)); + DBUG_RETURN(1); + } + } + pthread_mutex_unlock(&LOCK_global_index_stats); + DBUG_RETURN(0); +} + /* collect status for all running threads */ @@ -4206,7 +4519,7 @@ int fill_schema_proc(THD *thd, TABLE_LIS DBUG_RETURN(1); } proc_table->file->ha_index_init(0, 1); - if ((res= proc_table->file->index_first(proc_table->record[0]))) + if ((res= proc_table->file->ha_index_first(proc_table->record[0]))) { res= (res == HA_ERR_END_OF_FILE) ? 0 : 1; goto err; @@ -4216,7 +4529,7 @@ int fill_schema_proc(THD *thd, TABLE_LIS res= 1; goto err; } - while (!proc_table->file->index_next(proc_table->record[0])) + while (!proc_table->file->ha_index_next(proc_table->record[0])) { if (store_schema_proc(thd, table, proc_table, wild, full_access, definer)) { @@ -5462,6 +5775,81 @@ struct schema_table_ref ST_SCHEMA_TABLE *schema_table; }; +ST_FIELD_INFO user_stats_fields_info[]= +{ + {"USER", USERNAME_LENGTH, MYSQL_TYPE_STRING, 0, 0, "User", SKIP_OPEN_TABLE}, + {"TOTAL_CONNECTIONS", MY_INT64_NUM_DECIMAL_DIGITS, MYSQL_TYPE_LONG, 0, 0, "Total_connections",SKIP_OPEN_TABLE}, + {"CONCURRENT_CONNECTIONS", MY_INT64_NUM_DECIMAL_DIGITS, MYSQL_TYPE_LONG, 0, 0, "Concurrent_connections",SKIP_OPEN_TABLE}, + {"CONNECTED_TIME", MY_INT64_NUM_DECIMAL_DIGITS, MYSQL_TYPE_LONG, 0, 0, "Connected_time",SKIP_OPEN_TABLE}, + {"BUSY_TIME", MY_INT64_NUM_DECIMAL_DIGITS, MYSQL_TYPE_DOUBLE, 0, 0, "Busy_time",SKIP_OPEN_TABLE}, + {"CPU_TIME", MY_INT64_NUM_DECIMAL_DIGITS, MYSQL_TYPE_DOUBLE, 0, 0, "Cpu_time",SKIP_OPEN_TABLE}, + {"BYTES_RECEIVED", MY_INT64_NUM_DECIMAL_DIGITS, MYSQL_TYPE_LONG, 0, 0, "Bytes_received",SKIP_OPEN_TABLE}, + {"BYTES_SENT", MY_INT64_NUM_DECIMAL_DIGITS, MYSQL_TYPE_LONG, 0, 0, "Bytes_sent",SKIP_OPEN_TABLE}, + {"BINLOG_BYTES_WRITTEN", MY_INT64_NUM_DECIMAL_DIGITS, MYSQL_TYPE_LONG, 0, 0, "Binlog_bytes_written",SKIP_OPEN_TABLE}, + {"ROWS_READ", MY_INT64_NUM_DECIMAL_DIGITS, MYSQL_TYPE_LONG, 0, 0, "Rows_read",SKIP_OPEN_TABLE}, + {"ROWS_SENT", MY_INT64_NUM_DECIMAL_DIGITS, MYSQL_TYPE_LONG, 0, 0, "Rows_sent",SKIP_OPEN_TABLE}, + {"ROWS_DELETED", MY_INT64_NUM_DECIMAL_DIGITS, MYSQL_TYPE_LONG, 0, 0, "Rows_deleted",SKIP_OPEN_TABLE}, + {"ROWS_INSERTED", MY_INT64_NUM_DECIMAL_DIGITS, MYSQL_TYPE_LONG, 0, 0, "Rows_inserted",SKIP_OPEN_TABLE}, + {"ROWS_UPDATED", MY_INT64_NUM_DECIMAL_DIGITS, MYSQL_TYPE_LONG, 0, 0, "Rows_updated",SKIP_OPEN_TABLE}, + {"SELECT_COMMANDS", MY_INT64_NUM_DECIMAL_DIGITS, MYSQL_TYPE_LONG, 0, 0, "Select_commands",SKIP_OPEN_TABLE}, + {"UPDATE_COMMANDS", MY_INT64_NUM_DECIMAL_DIGITS, MYSQL_TYPE_LONG, 0, 0, "Update_commands",SKIP_OPEN_TABLE}, + {"OTHER_COMMANDS", MY_INT64_NUM_DECIMAL_DIGITS, MYSQL_TYPE_LONG, 0, 0, "Other_commands",SKIP_OPEN_TABLE}, + {"COMMIT_TRANSACTIONS", MY_INT64_NUM_DECIMAL_DIGITS, MYSQL_TYPE_LONG, 0, 0, "Commit_transactions",SKIP_OPEN_TABLE}, + {"ROLLBACK_TRANSACTIONS", MY_INT64_NUM_DECIMAL_DIGITS, MYSQL_TYPE_LONG, 0, 0, "Rollback_transactions",SKIP_OPEN_TABLE}, + {"DENIED_CONNECTIONS", MY_INT64_NUM_DECIMAL_DIGITS, MYSQL_TYPE_LONG, 0, 0, "Denied_connections",SKIP_OPEN_TABLE}, + {"LOST_CONNECTIONS", MY_INT64_NUM_DECIMAL_DIGITS, MYSQL_TYPE_LONG, 0, 0, "Lost_connections",SKIP_OPEN_TABLE}, + {"ACCESS_DENIED", MY_INT64_NUM_DECIMAL_DIGITS, MYSQL_TYPE_LONG, 0, 0, "Access_denied",SKIP_OPEN_TABLE}, + {"EMPTY_QUERIES", MY_INT64_NUM_DECIMAL_DIGITS, MYSQL_TYPE_LONG, 0, 0, "Empty_queries",SKIP_OPEN_TABLE}, + {0, 0, MYSQL_TYPE_STRING, 0, 0, 0, 0} +}; + +ST_FIELD_INFO client_stats_fields_info[]= +{ + {"CLIENT", LIST_PROCESS_HOST_LEN, MYSQL_TYPE_STRING, 0, 0, "Client",SKIP_OPEN_TABLE}, + {"TOTAL_CONNECTIONS", MY_INT64_NUM_DECIMAL_DIGITS, MYSQL_TYPE_LONG, 0, 0, "Total_connections",SKIP_OPEN_TABLE}, + {"CONCURRENT_CONNECTIONS", MY_INT64_NUM_DECIMAL_DIGITS, MYSQL_TYPE_LONG, 0, 0, "Concurrent_connections",SKIP_OPEN_TABLE}, + {"CONNECTED_TIME", MY_INT64_NUM_DECIMAL_DIGITS, MYSQL_TYPE_LONG, 0, 0, "Connected_time",SKIP_OPEN_TABLE}, + {"BUSY_TIME", MY_INT64_NUM_DECIMAL_DIGITS, MYSQL_TYPE_DOUBLE, 0, 0, "Busy_time",SKIP_OPEN_TABLE}, + {"CPU_TIME", MY_INT64_NUM_DECIMAL_DIGITS, MYSQL_TYPE_DOUBLE, 0, 0, "Cpu_time",SKIP_OPEN_TABLE}, + {"BYTES_RECEIVED", MY_INT64_NUM_DECIMAL_DIGITS, MYSQL_TYPE_LONG, 0, 0, "Bytes_received",SKIP_OPEN_TABLE}, + {"BYTES_SENT", MY_INT64_NUM_DECIMAL_DIGITS, MYSQL_TYPE_LONG, 0, 0, "Bytes_sent",SKIP_OPEN_TABLE}, + {"BINLOG_BYTES_WRITTEN", MY_INT64_NUM_DECIMAL_DIGITS, MYSQL_TYPE_LONG, 0, 0, "Binlog_bytes_written",SKIP_OPEN_TABLE}, + {"ROWS_READ", MY_INT64_NUM_DECIMAL_DIGITS, MYSQL_TYPE_LONG, 0, 0, "Rows_read",SKIP_OPEN_TABLE}, + {"ROWS_SENT", MY_INT64_NUM_DECIMAL_DIGITS, MYSQL_TYPE_LONG, 0, 0, "Rows_sent",SKIP_OPEN_TABLE}, + {"ROWS_DELETED", MY_INT64_NUM_DECIMAL_DIGITS, MYSQL_TYPE_LONG, 0, 0, "Rows_deleted",SKIP_OPEN_TABLE}, + {"ROWS_INSERTED", MY_INT64_NUM_DECIMAL_DIGITS, MYSQL_TYPE_LONG, 0, 0, "Rows_inserted",SKIP_OPEN_TABLE}, + {"ROWS_UPDATED", MY_INT64_NUM_DECIMAL_DIGITS, MYSQL_TYPE_LONG, 0, 0, "Rows_updated",SKIP_OPEN_TABLE}, + {"SELECT_COMMANDS", MY_INT64_NUM_DECIMAL_DIGITS, MYSQL_TYPE_LONG, 0, 0, "Select_commands",SKIP_OPEN_TABLE}, + {"UPDATE_COMMANDS", MY_INT64_NUM_DECIMAL_DIGITS, MYSQL_TYPE_LONG, 0, 0, "Update_commands",SKIP_OPEN_TABLE}, + {"OTHER_COMMANDS", MY_INT64_NUM_DECIMAL_DIGITS, MYSQL_TYPE_LONG, 0, 0, "Other_commands",SKIP_OPEN_TABLE}, + {"COMMIT_TRANSACTIONS", MY_INT64_NUM_DECIMAL_DIGITS, MYSQL_TYPE_LONG, 0, 0, "Commit_transactions",SKIP_OPEN_TABLE}, + {"ROLLBACK_TRANSACTIONS", MY_INT64_NUM_DECIMAL_DIGITS, MYSQL_TYPE_LONG, 0, 0, "Rollback_transactions",SKIP_OPEN_TABLE}, + {"DENIED_CONNECTIONS", MY_INT64_NUM_DECIMAL_DIGITS, MYSQL_TYPE_LONG, 0, 0, "Denied_connections",SKIP_OPEN_TABLE}, + {"LOST_CONNECTIONS", MY_INT64_NUM_DECIMAL_DIGITS, MYSQL_TYPE_LONG, 0, 0, "Lost_connections",SKIP_OPEN_TABLE}, + {"ACCESS_DENIED", MY_INT64_NUM_DECIMAL_DIGITS, MYSQL_TYPE_LONG, 0, 0, "Access_denied",SKIP_OPEN_TABLE}, + {"EMPTY_QUERIES", MY_INT64_NUM_DECIMAL_DIGITS, MYSQL_TYPE_LONG, 0, 0, "Empty_queries",SKIP_OPEN_TABLE}, + {0, 0, MYSQL_TYPE_STRING, 0, 0, 0, 0} +}; + + +ST_FIELD_INFO table_stats_fields_info[]= +{ + {"TABLE_SCHEMA", NAME_LEN, MYSQL_TYPE_STRING, 0, 0, "Table_schema",SKIP_OPEN_TABLE}, + {"TABLE_NAME", NAME_LEN, MYSQL_TYPE_STRING, 0, 0, "Table_name",SKIP_OPEN_TABLE}, + {"ROWS_READ", MY_INT64_NUM_DECIMAL_DIGITS, MYSQL_TYPE_LONG, 0, 0, "Rows_read",SKIP_OPEN_TABLE}, + {"ROWS_CHANGED", MY_INT64_NUM_DECIMAL_DIGITS, MYSQL_TYPE_LONG, 0, 0, "Rows_changed",SKIP_OPEN_TABLE}, + {"ROWS_CHANGED_X_INDEXES", MY_INT64_NUM_DECIMAL_DIGITS, MYSQL_TYPE_LONG, 0, 0, "Rows_changed_x_#indexes",SKIP_OPEN_TABLE}, + {0, 0, MYSQL_TYPE_STRING, 0, 0, 0, 0} +}; + +ST_FIELD_INFO index_stats_fields_info[]= +{ + {"TABLE_SCHEMA", NAME_LEN, MYSQL_TYPE_STRING, 0, 0, "Table_schema",SKIP_OPEN_TABLE}, + {"TABLE_NAME", NAME_LEN, MYSQL_TYPE_STRING, 0, 0, "Table_name",SKIP_OPEN_TABLE}, + {"INDEX_NAME", NAME_LEN, MYSQL_TYPE_STRING, 0, 0, "Index_name",SKIP_OPEN_TABLE}, + {"ROWS_READ", MY_INT64_NUM_DECIMAL_DIGITS, MYSQL_TYPE_LONG, 0, 0, "Rows_read",SKIP_OPEN_TABLE}, + {0, 0, MYSQL_TYPE_STRING, 0, 0, 0,0} +}; /* Find schema_tables elment by name @@ -6683,6 +7071,8 @@ ST_SCHEMA_TABLE schema_tables[]= { {"CHARACTER_SETS", charsets_fields_info, create_schema_table, fill_schema_charsets, make_character_sets_old_format, 0, -1, -1, 0, 0}, + {"CLIENT_STATISTICS", client_stats_fields_info, create_schema_table, + fill_schema_client_stats, make_old_format, 0, -1, -1, 0, 0}, {"COLLATIONS", collation_fields_info, create_schema_table, fill_schema_collation, make_old_format, 0, -1, -1, 0, 0}, {"COLLATION_CHARACTER_SET_APPLICABILITY", coll_charset_app_fields_info, @@ -6707,6 +7097,8 @@ ST_SCHEMA_TABLE schema_tables[]= fill_status, make_old_format, 0, 0, -1, 0, 0}, {"GLOBAL_VARIABLES", variables_fields_info, create_schema_table, fill_variables, make_old_format, 0, 0, -1, 0, 0}, + {"INDEX_STATISTICS", index_stats_fields_info, create_schema_table, + fill_schema_index_stats, make_old_format, 0, -1, -1, 0, 0}, {"KEY_COLUMN_USAGE", key_column_usage_fields_info, create_schema_table, get_all_tables, 0, get_schema_key_column_usage_record, 4, 5, 0, OPEN_TABLE_ONLY}, @@ -6748,11 +7140,15 @@ ST_SCHEMA_TABLE schema_tables[]= get_all_tables, make_table_names_old_format, 0, 1, 2, 1, 0}, {"TABLE_PRIVILEGES", table_privileges_fields_info, create_schema_table, fill_schema_table_privileges, 0, 0, -1, -1, 0, 0}, + {"TABLE_STATISTICS", table_stats_fields_info, create_schema_table, + fill_schema_table_stats, make_old_format, 0, -1, -1, 0, 0}, {"TRIGGERS", triggers_fields_info, create_schema_table, get_all_tables, make_old_format, get_schema_triggers_record, 5, 6, 0, OPEN_TABLE_ONLY}, {"USER_PRIVILEGES", user_privileges_fields_info, create_schema_table, fill_schema_user_privileges, 0, 0, -1, -1, 0, 0}, + {"USER_STATISTICS", user_stats_fields_info, create_schema_table, + fill_schema_user_stats, make_old_format, 0, -1, -1, 0, 0}, {"VARIABLES", variables_fields_info, create_schema_table, fill_variables, make_old_format, 0, 0, -1, 1, 0}, {"VIEWS", view_fields_info, create_schema_table, === modified file 'sql/sql_table.cc' --- a/sql/sql_table.cc 2009-09-18 01:04:43 +0000 +++ b/sql/sql_table.cc 2009-10-19 17:14:48 +0000 @@ -7815,7 +7815,7 @@ bool mysql_checksum_table(THD *thd, TABL goto err; } ha_checksum row_crc= 0; - int error= t->file->rnd_next(t->record[0]); + int error= t->file->ha_rnd_next(t->record[0]); if (unlikely(error)) { if (error == HA_ERR_RECORD_DELETED) === modified file 'sql/sql_udf.cc' --- a/sql/sql_udf.cc 2009-05-15 12:57:51 +0000 +++ b/sql/sql_udf.cc 2009-10-19 17:14:48 +0000 @@ -567,10 +567,10 @@ int mysql_drop_function(THD *thd,const L goto err; table->use_all_columns(); table->field[0]->store(exact_name_str, exact_name_len, &my_charset_bin); - if (!table->file->index_read_idx_map(table->record[0], 0, - (uchar*) table->field[0]->ptr, - HA_WHOLE_KEY, - HA_READ_KEY_EXACT)) + if (!table->file->ha_index_read_idx_map(table->record[0], 0, + (uchar*) table->field[0]->ptr, + HA_WHOLE_KEY, + HA_READ_KEY_EXACT)) { int error; if ((error = table->file->ha_delete_row(table->record[0]))) === modified file 'sql/sql_update.cc' --- a/sql/sql_update.cc 2009-09-07 20:50:10 +0000 +++ b/sql/sql_update.cc 2009-10-19 17:14:48 +0000 @@ -142,7 +142,7 @@ static void prepare_record_for_error_mes /* Tell the engine about the new set. */ table->file->column_bitmaps_signal(); /* Read record that is identified by table->file->ref. */ - (void) table->file->rnd_pos(table->record[1], table->file->ref); + (void) table->file->ha_rnd_pos(table->record[1], table->file->ref); /* Copy the newly read columns into the new record. */ for (field_p= table->field; (field= *field_p); field_p++) if (bitmap_is_set(&unique_map, field->field_index)) @@ -1928,7 +1928,7 @@ int multi_update::do_updates() { if (thd->killed && trans_safe) goto err; - if ((local_error=tmp_table->file->rnd_next(tmp_table->record[0]))) + if ((local_error= tmp_table->file->ha_rnd_next(tmp_table->record[0]))) { if (local_error == HA_ERR_END_OF_FILE) break; @@ -1943,12 +1943,12 @@ int multi_update::do_updates() uint field_num= 0; do { - if((local_error= - tbl->file->rnd_pos(tbl->record[0], - (uchar *) tmp_table->field[field_num]->ptr))) + if ((local_error= + tbl->file->ha_rnd_pos(tbl->record[0], + (uchar*) tmp_table->field[field_num]->ptr))) goto err; field_num++; - } while((tbl= check_opt_it++)); + } while ((tbl= check_opt_it++)); table->status|= STATUS_UPDATED; store_record(table,record[1]); === modified file 'sql/sql_yacc.yy' --- a/sql/sql_yacc.yy 2009-09-07 20:50:10 +0000 +++ b/sql/sql_yacc.yy 2009-10-19 17:14:48 +0000 @@ -598,6 +598,7 @@ bool my_yyoverflow(short **a, YYSTYPE ** %token CHECK_SYM /* SQL-2003-R */ %token CIPHER_SYM %token CLIENT_SYM +%token CLIENT_STATS_SYM %token CLOSE_SYM /* SQL-2003-R */ %token COALESCE /* SQL-2003-N */ %token CODE_SYM @@ -744,6 +745,7 @@ bool my_yyoverflow(short **a, YYSTYPE ** %token IMPORT %token INDEXES %token INDEX_SYM +%token INDEX_STATS_SYM %token INFILE %token INITIAL_SIZE_SYM %token INNER_SYM /* SQL-2003-R */ @@ -985,6 +987,7 @@ bool my_yyoverflow(short **a, YYSTYPE ** %token SIGNED_SYM %token SIMPLE_SYM /* SQL-2003-N */ %token SLAVE +%token SLOW_SYM %token SMALLINT /* SQL-2003-R */ %token SNAPSHOT_SYM %token SOCKET_SYM @@ -1029,6 +1032,7 @@ bool my_yyoverflow(short **a, YYSTYPE ** %token TABLES %token TABLESPACE %token TABLE_REF_PRIORITY +%token TABLE_STATS_SYM %token TABLE_SYM /* SQL-2003-R */ %token TABLE_CHECKSUM_SYM %token TEMPORARY /* SQL-2003-N */ @@ -1076,6 +1080,7 @@ bool my_yyoverflow(short **a, YYSTYPE ** %token UPGRADE_SYM %token USAGE /* SQL-2003-N */ %token USER /* SQL-2003-R */ +%token USER_STATS_SYM %token USE_FRM %token USE_SYM %token USING /* SQL-2003-R */ @@ -10131,6 +10136,34 @@ show_param: { Lex->sql_command = SQLCOM_SHOW_SLAVE_STAT; } + | CLIENT_STATS_SYM + { + LEX *lex= Lex; + lex->sql_command= SQLCOM_SHOW_CLIENT_STATS; + if (prepare_schema_table(YYTHD, lex, 0, SCH_CLIENT_STATS)) + MYSQL_YYABORT; + } + | USER_STATS_SYM + { + LEX *lex= Lex; + lex->sql_command= SQLCOM_SHOW_USER_STATS; + if (prepare_schema_table(YYTHD, lex, 0, SCH_USER_STATS)) + MYSQL_YYABORT; + } + | TABLE_STATS_SYM + { + LEX *lex= Lex; + lex->sql_command= SQLCOM_SHOW_TABLE_STATS; + if (prepare_schema_table(YYTHD, lex, 0, SCH_TABLE_STATS)) + MYSQL_YYABORT; + } + | INDEX_STATS_SYM + { + LEX *lex= Lex; + lex->sql_command= SQLCOM_SHOW_INDEX_STATS; + if (prepare_schema_table(YYTHD, lex, 0, SCH_INDEX_STATS)) + MYSQL_YYABORT; + } | CREATE PROCEDURE sp_name { LEX *lex= Lex; @@ -10339,6 +10372,16 @@ flush_option: { Lex->type|= REFRESH_STATUS; } | SLAVE { Lex->type|= REFRESH_SLAVE; } + | SLOW_SYM QUERY_SYM LOGS_SYM + { Lex->type |= REFRESH_SLOW_QUERY_LOG; } + | CLIENT_STATS_SYM + { Lex->type|= REFRESH_CLIENT_STATS; } + | USER_STATS_SYM + { Lex->type|= REFRESH_USER_STATS; } + | TABLE_STATS_SYM + { Lex->type|= REFRESH_TABLE_STATS; } + | INDEX_STATS_SYM + { Lex->type|= REFRESH_INDEX_STATS; } | MASTER_SYM { Lex->type|= REFRESH_MASTER; } | DES_KEY_FILE @@ -11447,6 +11490,7 @@ keyword_sp: | CHAIN_SYM {} | CHANGED {} | CIPHER_SYM {} + | CLIENT_STATS_SYM {} | CLIENT_SYM {} | COALESCE {} | CODE_SYM {} @@ -11508,6 +11552,7 @@ keyword_sp: | HOSTS_SYM {} | HOUR_SYM {} | IDENTIFIED_SYM {} + | INDEX_STATS_SYM {} | INVOKER_SYM {} | IMPORT {} | INDEXES {} @@ -11631,6 +11676,7 @@ keyword_sp: | SIMPLE_SYM {} | SHARE_SYM {} | SHUTDOWN {} + | SLOW_SYM {} | SNAPSHOT_SYM {} | SOUNDS_SYM {} | SOURCE_SYM {} @@ -11650,6 +11696,7 @@ keyword_sp: | SUSPEND_SYM {} | SWAPS_SYM {} | SWITCHES_SYM {} + | TABLE_STATS_SYM {} | TABLES {} | TABLE_CHECKSUM_SYM {} | TABLESPACE {} @@ -11675,6 +11722,7 @@ keyword_sp: | UNKNOWN_SYM {} | UNTIL_SYM {} | USER {} + | USER_STATS_SYM {} | USE_FRM {} | VARIABLES {} | VIEW_SYM {} === modified file 'sql/structs.h' --- a/sql/structs.h 2009-06-26 19:57:42 +0000 +++ b/sql/structs.h 2009-10-19 17:14:48 +0000 @@ -76,6 +76,7 @@ typedef struct st_key { uint extra_length; uint usable_key_parts; /* Should normally be = key_parts */ uint block_size; + uint name_length; enum ha_key_alg algorithm; /* Note that parser is used when the table is opened for use, and @@ -88,6 +89,8 @@ typedef struct st_key { }; KEY_PART_INFO *key_part; char *name; /* Name of key */ + /* Unique name for cache; db + \0 + table_name + \0 + key_name + \0 */ + uchar *cache_name; /* Array of AVG(#records with the same field value) for 1st ... Nth key part. 0 means 'not known'. @@ -231,6 +234,111 @@ typedef struct user_conn { USER_RESOURCES user_resources; } USER_CONN; +typedef struct st_user_stats +{ + char user[max(USERNAME_LENGTH, LIST_PROCESS_HOST_LEN) + 1]; + // Account name the user is mapped to when this is a user from mapped_user. + // Otherwise, the same value as user. + char priv_user[max(USERNAME_LENGTH, LIST_PROCESS_HOST_LEN) + 1]; + uint user_name_length; + uint total_connections; + uint concurrent_connections; + time_t connected_time; // in seconds + double busy_time; // in seconds + double cpu_time; // in seconds + ulonglong bytes_received; + ulonglong bytes_sent; + ulonglong binlog_bytes_written; + ha_rows rows_read, rows_sent; + ha_rows rows_updated, rows_deleted, rows_inserted; + ulonglong select_commands, update_commands, other_commands; + ulonglong commit_trans, rollback_trans; + ulonglong denied_connections, lost_connections; + ulonglong access_denied_errors; + ulonglong empty_queries; +} USER_STATS; + +/* Lookup function for hash tables with USER_STATS entries */ +extern "C" uchar *get_key_user_stats(USER_STATS *user_stats, size_t *length, + my_bool not_used __attribute__((unused))); + +/* Free all memory for a hash table with USER_STATS entries */ +extern void free_user_stats(USER_STATS* user_stats); + +/* Intialize an instance of USER_STATS */ +extern void +init_user_stats(USER_STATS *user_stats, + const char *user, + size_t user_length, + const char *priv_user, + uint total_connections, + uint concurrent_connections, + time_t connected_time, + double busy_time, + double cpu_time, + ulonglong bytes_received, + ulonglong bytes_sent, + ulonglong binlog_bytes_written, + ha_rows rows_sent, + ha_rows rows_read, + ha_rows rows_inserted, + ha_rows rows_deleted, + ha_rows rows_updated, + ulonglong select_commands, + ulonglong update_commands, + ulonglong other_commands, + ulonglong commit_trans, + ulonglong rollback_trans, + ulonglong denied_connections, + ulonglong lost_connections, + ulonglong access_denied_errors, + ulonglong empty_queries); + +/* Increment values of an instance of USER_STATS */ +extern void +add_user_stats(USER_STATS *user_stats, + uint total_connections, + uint concurrent_connections, + time_t connected_time, + double busy_time, + double cpu_time, + ulonglong bytes_received, + ulonglong bytes_sent, + ulonglong binlog_bytes_written, + ha_rows rows_sent, + ha_rows rows_read, + ha_rows rows_inserted, + ha_rows rows_deleted, + ha_rows rows_updated, + ulonglong select_commands, + ulonglong update_commands, + ulonglong other_commands, + ulonglong commit_trans, + ulonglong rollback_trans, + ulonglong denied_connections, + ulonglong lost_connections, + ulonglong access_denied_errors, + ulonglong empty_queries); + +typedef struct st_table_stats +{ + char table[NAME_LEN * 2 + 2]; // [db] + '\0' + [table] + '\0' + uint table_name_length; + ulonglong rows_read, rows_changed; + ulonglong rows_changed_x_indexes; + /* Stores enum db_type, but forward declarations cannot be done */ + int engine_type; +} TABLE_STATS; + +typedef struct st_index_stats +{ + // [db] + '\0' + [table] + '\0' + [index] + '\0' + char index[NAME_LEN * 3 + 3]; + uint index_name_length; /* Length of 'index' */ + ulonglong rows_read; +} INDEX_STATS; + + /* Bits in form->update */ #define REG_MAKE_DUPP 1 /* Make a copy of record when read */ #define REG_NEW_RECORD 2 /* Write a new record if not found */ === modified file 'sql/table.cc' --- a/sql/table.cc 2009-09-09 21:06:57 +0000 +++ b/sql/table.cc 2009-10-19 17:14:48 +0000 @@ -1325,6 +1325,19 @@ static int open_binary_frm(THD *thd, TAB { uint usable_parts= 0; keyinfo->name=(char*) share->keynames.type_names[key]; + keyinfo->name_length= strlen(keyinfo->name); + keyinfo->cache_name= + (uchar*) alloc_root(&share->mem_root, + share->table_cache_key.length+ + keyinfo->name_length + 1); + if (keyinfo->cache_name) // If not out of memory + { + uchar *pos= keyinfo->cache_name; + memcpy(pos, share->table_cache_key.str, share->table_cache_key.length); + memcpy(pos + share->table_cache_key.length, keyinfo->name, + keyinfo->name_length+1); + } + /* Fix fulltext keys for old .frm files */ if (share->key_info[key].flags & HA_FULLTEXT) share->key_info[key].algorithm= HA_KEY_ALG_FULLTEXT; === modified file 'sql/table.h' --- a/sql/table.h 2009-09-15 10:46:35 +0000 +++ b/sql/table.h 2009-10-19 17:14:48 +0000 @@ -878,6 +878,7 @@ typedef struct st_foreign_key_info enum enum_schema_tables { SCH_CHARSETS= 0, + SCH_CLIENT_STATS, SCH_COLLATIONS, SCH_COLLATION_CHARACTER_SET_APPLICABILITY, SCH_COLUMNS, @@ -887,6 +888,7 @@ enum enum_schema_tables SCH_FILES, SCH_GLOBAL_STATUS, SCH_GLOBAL_VARIABLES, + SCH_INDEX_STATS, SCH_KEY_COLUMN_USAGE, SCH_OPEN_TABLES, SCH_PARTITIONS, @@ -905,8 +907,10 @@ enum enum_schema_tables SCH_TABLE_CONSTRAINTS, SCH_TABLE_NAMES, SCH_TABLE_PRIVILEGES, + SCH_TABLE_STATS, SCH_TRIGGERS, SCH_USER_PRIVILEGES, + SCH_USER_STATS, SCH_VARIABLES, SCH_VIEWS }; === modified file 'sql/tztime.cc' --- a/sql/tztime.cc 2009-09-07 20:50:10 +0000 +++ b/sql/tztime.cc 2009-10-19 17:14:48 +0000 @@ -1676,7 +1676,7 @@ my_tz_init(THD *org_thd, const char *def tz_leapcnt= 0; - res= table->file->index_first(table->record[0]); + res= table->file->ha_index_first(table->record[0]); while (!res) { @@ -1698,7 +1698,7 @@ my_tz_init(THD *org_thd, const char *def tz_leapcnt, (ulong) tz_lsis[tz_leapcnt-1].ls_trans, tz_lsis[tz_leapcnt-1].ls_corr)); - res= table->file->index_next(table->record[0]); + res= table->file->ha_index_next(table->record[0]); } (void)table->file->ha_index_end(); @@ -1865,8 +1865,8 @@ tz_load_from_open_tables(const String *t */ (void)table->file->ha_index_init(0, 1); - if (table->file->index_read_map(table->record[0], table->field[0]->ptr, - HA_WHOLE_KEY, HA_READ_KEY_EXACT)) + if (table->file->ha_index_read_map(table->record[0], table->field[0]->ptr, + HA_WHOLE_KEY, HA_READ_KEY_EXACT)) { #ifdef EXTRA_DEBUG /* @@ -1893,8 +1893,8 @@ tz_load_from_open_tables(const String *t table->field[0]->store((longlong) tzid, TRUE); (void)table->file->ha_index_init(0, 1); - if (table->file->index_read_map(table->record[0], table->field[0]->ptr, - HA_WHOLE_KEY, HA_READ_KEY_EXACT)) + if (table->file->ha_index_read_map(table->record[0], table->field[0]->ptr, + HA_WHOLE_KEY, HA_READ_KEY_EXACT)) { sql_print_error("Can't find description of time zone '%u'", tzid); goto end; @@ -1920,8 +1920,8 @@ tz_load_from_open_tables(const String *t table->field[0]->store((longlong) tzid, TRUE); (void)table->file->ha_index_init(0, 1); - res= table->file->index_read_map(table->record[0], table->field[0]->ptr, - (key_part_map)1, HA_READ_KEY_EXACT); + res= table->file->ha_index_read_map(table->record[0], table->field[0]->ptr, + (key_part_map)1, HA_READ_KEY_EXACT); while (!res) { ttid= (uint)table->field[1]->val_int(); @@ -1968,8 +1968,8 @@ tz_load_from_open_tables(const String *t tmp_tz_info.typecnt= ttid + 1; - res= table->file->index_next_same(table->record[0], - table->field[0]->ptr, 4); + res= table->file->ha_index_next_same(table->record[0], + table->field[0]->ptr, 4); } if (res != HA_ERR_END_OF_FILE) @@ -1991,8 +1991,8 @@ tz_load_from_open_tables(const String *t table->field[0]->store((longlong) tzid, TRUE); (void)table->file->ha_index_init(0, 1); - res= table->file->index_read_map(table->record[0], table->field[0]->ptr, - (key_part_map)1, HA_READ_KEY_EXACT); + res= table->file->ha_index_read_map(table->record[0], table->field[0]->ptr, + (key_part_map)1, HA_READ_KEY_EXACT); while (!res) { ttime= (my_time_t)table->field[1]->val_int(); @@ -2021,8 +2021,8 @@ tz_load_from_open_tables(const String *t ("time_zone_transition table: tz_id: %u tt_time: %lu tt_id: %u", tzid, (ulong) ttime, ttid)); - res= table->file->index_next_same(table->record[0], - table->field[0]->ptr, 4); + res= table->file->ha_index_next_same(table->record[0], + table->field[0]->ptr, 4); } /*

1 0

Re: [Maria-developers] Windows installer, part 2
by Steve Obbayi 19 Oct '09

19 Oct '09

Another thing to remember is when installation is already on the computer and the user attempts to install they should be given options of Repair, Modify or Remove.. I believe that would also be standard when the user opts to uninstall. SKYPE: sobbayi US: +1 202 297 6831 +1 202 470 0525 KE: +254 722 627 691 ----- Original Message ----- From: "Peter Laursen" <peter_laursen(a)webyog.com> To: "Bo Thorsen" <bo(a)sonofthor.dk> Cc: maria-developers(a)lists.launchpad.net Sent: Monday, October 19, 2009 5:41:07 AM GMT -05:00 US/Canada Eastern Subject: Re: [Maria-developers] Windows installer, part 2 I would suggest the basic workflow for the first build (not complete/final version): . 1) unpack archive (that is also a validation of the archive integrity) 2) prompt user for 'basedir' and 'datadir' 3) copy files accordingly 4) prompt user for choice of mini/midi/maxi base configuration and the service name to use 5) 'promote' (copy + rename) a template to my.ini as defined by user. 6) register service with --defaults-file and service name as defined, add items to 'Add/remove programs' in Control Panel, start menu, create uninstaller. 7) start service 8) prompt user for root password (and maybe also if remote root access should be allowed and if anonymous access should be allowed) 9) connect to server as root with empty password (using command line client or a small standalone executable client compiled with C-API), update mysql.user table as defined + flush privileges, disconnect 10) done . the steps 4)+5) above are of course those that should be elaborated further (port, default engine, charset etc - all what is established by updating/writing startup options in my.ini). But I like the top-down approach described here: first make a functional thing from start to end - next detail steps. . Peter Webyog On Mon, Oct 19, 2009 at 10:54, Bo Thorsen < bo(a)sonofthor.dk > wrote: Hi, I tried to send this from my askmonty.org address, so it might come twice. Fredag 16 oktober 2009 23:18:23 Sergey Petrunya wrote: > Thanks everyone for input on the input so far! I believe I've collected it > all and put into the WL entry: > http://askmonty.org/worklog/Server-RawIdeaBin/?tid=55 > > * If there is something missing there, please bring that argument up again. I would really like to see NSIS used for this package. It is a really good install system, and with all the features in part 1, I don't think anything else will be easier to use. It will be wasted effort to try and use anything else or build a script from a bat file or something like that. A simple zip file that doesn't do any installation is of course the easiest, but that's not what step #1 describes. I have used NSIS to build three installers before, and it has been powerful and easy. However, none of the installers I did used as many features as this installer will. The ones I did were fairly simple. One thing it also does, which I haven't seen mention here, is to produce an uninstaller script. This is almost as hard to do as an installer. It's not produced automatically, but it is part of the NSIS system to create one. At the end of this mail, I have copied some of the statements from an install script as examples of how easy it is to do standard things. > * Final call for objections/comments on Step#1. Please voice them now, > otherwise what is described in step #1 may end up implemented. One thing I don't see here is how to handle development files. Will there be one installer at 100MB or two packages? One way to handle it would be to have a separate installer that is installed with MariaDB, that pulls all the files from the net. Here are a couple of examples from an NSIS script: ; Install the license file if chosen !ifdef licensefile CreateShortCut "${startmenu}\License.lnk "$INSTDIR\${licensefile}" !endif ; Add a registry key with the install dir and the uninstaller WriteRegStr HKLM "${regkey}" "Install_Dir" "$INSTDIR" WriteRegStr HKLM "${uninstkey}" "UninstallString" '"$INSTDIR\${uninstaller}"' ; Example of an optional install IfFileExists "$EXEDIR/${java}" java nojava java: MessageBox MB_YESNO "Install the Java Runtime Environment?" IDNO nojava Exec "$EXEDIR/${java}" nojava: ; Files are added like this: File /a "${srcdir}\somefile" ; More files in a generated file with install lines !include "${files}" Bo. _______________________________________________ Mailing list: https://launchpad.net/~maria-developers Post to : maria-developers(a)lists.launchpad.net Unsubscribe : https://launchpad.net/~maria-developers More help : https://help.launchpad.net/ListHelp _______________________________________________ Mailing list: https://launchpad.net/~maria-developers Post to : maria-developers(a)lists.launchpad.net Unsubscribe : https://launchpad.net/~maria-developers More help : https://help.launchpad.net/ListHelp

1 0

Re: [Maria-developers] Windows installer, part 2
by Bo Thorsen 19 Oct '09

19 Oct '09

Hi, I tried to send this from my askmonty.org address, so it might come twice. Fredag 16 oktober 2009 23:18:23 Sergey Petrunya wrote: > Thanks everyone for input on the input so far! I believe I've collected it > all and put into the WL entry: > http://askmonty.org/worklog/Server-RawIdeaBin/?tid=55 > > * If there is something missing there, please bring that argument up again. I would really like to see NSIS used for this package. It is a really good install system, and with all the features in part 1, I don't think anything else will be easier to use. It will be wasted effort to try and use anything else or build a script from a bat file or something like that. A simple zip file that doesn't do any installation is of course the easiest, but that's not what step #1 describes. I have used NSIS to build three installers before, and it has been powerful and easy. However, none of the installers I did used as many features as this installer will. The ones I did were fairly simple. One thing it also does, which I haven't seen mention here, is to produce an uninstaller script. This is almost as hard to do as an installer. It's not produced automatically, but it is part of the NSIS system to create one. At the end of this mail, I have copied some of the statements from an install script as examples of how easy it is to do standard things. > * Final call for objections/comments on Step#1. Please voice them now, > otherwise what is described in step #1 may end up implemented. One thing I don't see here is how to handle development files. Will there be one installer at 100MB or two packages? One way to handle it would be to have a separate installer that is installed with MariaDB, that pulls all the files from the net. Here are a couple of examples from an NSIS script: ; Install the license file if chosen !ifdef licensefile CreateShortCut "${startmenu}\License.lnk "$INSTDIR\${licensefile}" !endif ; Add a registry key with the install dir and the uninstaller WriteRegStr HKLM "${regkey}" "Install_Dir" "$INSTDIR" WriteRegStr HKLM "${uninstkey}" "UninstallString" '"$INSTDIR\${uninstaller}"' ; Example of an optional install IfFileExists "$EXEDIR/${java}" java nojava java: MessageBox MB_YESNO "Install the Java Runtime Environment?" IDNO nojava Exec "$EXEDIR/${java}" nojava: ; Files are added like this: File /a "${srcdir}\somefile" ; More files in a generated file with install lines !include "${files}" Bo.

2 1