July 2017 - discuss - lists.mariadb.org

[Maria-discuss] MariaDB 10.2.7 and libmysql compatibility
by Reindl Harald 13 Jul '17

13 Jul '17

i am missing something (probably simple like symbol versioning however done with some hack) or is it really not possible by providing symlinks to avoid "libmysqlclient.so.18 => not found" and all clients linked against 10.1.x failing? attached my current rpm-spec - the server itself works fine and i really want to avoid building a "server-only" package for the whole lifetime of Fedora25/26 as well as not wait for Fedora 27 to get 10.2.x [root@testserver:/usr/lib64/mysql]$ ls insgesamt 308K lrwxrwxrwx 1 root root 13 2017-07-13 13:15 libmysqlclient.so.18.0.0 -> libmariadb.so lrwxrwxrwx 1 root root 13 2017-07-13 13:15 libmysqlclient.so.18 -> libmariadb.so -rwxr-xr-x 1 root root 305K 2017-07-13 13:15 libmariadb.so.3 lrwxrwxrwx 1 root root 15 2017-07-13 13:15 libmariadb.so -> libmariadb.so.3 lrwxrwxrwx 1 root root 17 2017-07-13 13:15 libmysqlclient_r.so -> libmysqlclient.so lrwxrwxrwx 1 root root 13 2017-07-13 13:15 libmysqlclient.so -> libmariadb.so [root@testserver:/usr/lib64/mysql]$ ldd /usr/sbin/dbmail-imapd linux-vdso.so.1 (0x00007fff2fdfd000) libgmime-2.6.so.0 => /lib64/libgmime-2.6.so.0 (0x00007f0b0fd94000) libgobject-2.0.so.0 => /lib64/libgobject-2.0.so.0 (0x00007f0b0fb42000) libglib-2.0.so.0 => /lib64/libglib-2.0.so.0 (0x00007f0b0f82f000) libevent-2.0.so.5 => /lib64/libevent-2.0.so.5 (0x00007f0b0f7e5000) libzdb.so.11 => /lib64/libzdb.so.11 (0x00007f0b0f7d0000) libdbmail.so.0 => /usr/lib64/dbmail/libdbmail.so.0 (0x00007f0b0f759000) libpthread.so.0 => /lib64/libpthread.so.0 (0x00007f0b0f53b000) libc.so.6 => /lib64/libc.so.6 (0x00007f0b0f175000) libgio-2.0.so.0 => /lib64/libgio-2.0.so.0 (0x00007f0b0ede2000) libz.so.1 => /lib64/libz.so.1 (0x00007f0b0ebcc000) libpcre.so.1 => /lib64/libpcre.so.1 (0x00007f0b0e95a000) libffi.so.6 => /lib64/libffi.so.6 (0x00007f0b0e750000) libmysqlclient.so.18 => not found libm.so.6 => /lib64/libm.so.6 (0x00007f0b0e447000) libdl.so.2 => /lib64/libdl.so.2 (0x00007f0b0e243000) libssl.so.10 => /lib64/libssl.so.10 (0x00007f0b0dfd1000) libcrypto.so.10 => /lib64/libcrypto.so.10 (0x00007f0b0db70000) libgmodule-2.0.so.0 => /lib64/libgmodule-2.0.so.0 (0x00007f0b0d96a000) libgthread-2.0.so.0 => /lib64/libgthread-2.0.so.0 (0x00007f0b0d768000) libmhash.so.2 => /lib64/libmhash.so.2 (0x00007f0b0d53c000) libevent_pthreads-2.0.so.5 => /lib64/libevent_pthreads-2.0.so.5 (0x00007f0b0d538000) libcrypt.so.1 => /lib64/libcrypt.so.1 (0x00007f0b0d302000) /lib64/ld-linux-x86-64.so.2 (0x0000559e18ffe000) libselinux.so.1 => /lib64/libselinux.so.1 (0x00007f0b0d0db000) libresolv.so.2 => /lib64/libresolv.so.2 (0x00007f0b0cebe000) libmount.so.1 => /lib64/libmount.so.1 (0x00007f0b0cc74000) libgomp.so.1 => /lib64/libgomp.so.1 (0x00007f0b0ca46000) libgssapi_krb5.so.2 => /lib64/libgssapi_krb5.so.2 (0x00007f0b0c7f9000) libkrb5.so.3 => /lib64/libkrb5.so.3 (0x00007f0b0c513000) libcom_err.so.2 => /lib64/libcom_err.so.2 (0x00007f0b0c30f000) libk5crypto.so.3 => /lib64/libk5crypto.so.3 (0x00007f0b0c0dc000) libfreebl3.so => /lib64/libfreebl3.so (0x00007f0b0bed9000) libblkid.so.1 => /lib64/libblkid.so.1 (0x00007f0b0bc96000) libuuid.so.1 => /lib64/libuuid.so.1 (0x00007f0b0ba91000) librt.so.1 => /lib64/librt.so.1 (0x00007f0b0b889000) libkrb5support.so.0 => /lib64/libkrb5support.so.0 (0x00007f0b0b678000) libkeyutils.so.1 => /lib64/libkeyutils.so.1 (0x00007f0b0b474000)

2 3

[Maria-discuss] 10.2.7: main.subselect_no_opts fails
by Reindl Harald 13 Jul '17

13 Jul '17

Completed: Failed 1/1865 tests, 99.95% were successful _________________________________________ [root@testserver:~]$ cat /usr/local/bin/mysql-test.sh #!/usr/bin/dash /usr/bin/su -c 'cd /usr/share/mysql-test; ./mysql-test-run.pl --parallel=4 --max-test-fail=0 --mysqld=--binlog-format=row --force --suites="main-,binlog-,csv-,funcs_1-,funcs_2-,handler-,heap-,maria-,optimizer_unfixed_bugs-,parts-,percona-,roles-,rpl-,sys_vars-,unit-,vcol-" --skip-test-list=/var/lib/mysql/skip-tests.txt' mysql rm -rf /usr/share/mysql-test/var/* _________________________________________ CURRENT_TEST: main.subselect_no_opts /usr/bin/mysqltest: Error on delete of '/usr/share/mysql-test/var/4/tmp/subselect.out.file.1' (Errcode: 2 "No such file or directory") mysqltest: In included file "/usr/share/mysql-test/t/subselect.test": included from /usr/share/mysql-test/t/subselect_no_opts.test at line 12: At line 5903: query 'Select (Select Sum(`TestCase`.Revenue) From mysql.slow_log E Where TestCase.TemplateID not in (Select 1 from mysql.slow_log where 2=2) ) As `ControlRev` From (Select 3 as Revenue, 4 as TemplateID) As `TestCase` Group By TestCase.Revenue, TestCase.TemplateID' failed: 1242: Subquery returns more than 1 row The result from queries just before the failure was: < snip > 2 SUBQUERY t2 ref b b 5 test.t1.a 2 Using index DROP TABLE t1,t2; # # MDEV-5991: crash in Item_field::used_tables # create table t1 (c int); select exists(select 1 from t1 group by `c` in (select `c` from t1)); exists(select 1 from t1 group by `c` in (select `c` from t1)) 0 drop table t1; # # MDEV-7565: Server crash with Signal 6 (part 2) # Select (Select Sum(`TestCase`.Revenue) From mysql.slow_log E Where TestCase.TemplateID not in (Select 1 from mysql.slow_log where 2=2) ) As `ControlRev` From (Select 3 as Revenue, 4 as TemplateID) As `TestCase` Group By TestCase.Revenue, TestCase.TemplateID; More results from queries before failure can be found in /usr/share/mysql-test/var/4/log/subselect_no_opts.log - saving '/usr/share/mysql-test/var/4/log/main.subselect_no_opts/' to '/usr/share/mysql-test/var/log/main.subselect_no_opts/' ***Warnings generated in error logs during shutdown after running tests: main.grant main.subselect3_jcl6 main.subselect_mat main.strict_autoinc_3heap main.func_time_hires main.failed_auth_3909 main.subselect_extra main.gis-json main.gis main.gis-precise main.subselect_no_opts main.explain main.get_diagnostics main.str_to_datetime_457 main.function_defaults main.execution_constants main.func_weight_string main.empty_server_name-8224 main.func_timestamp main.subselect_cache main.subselect_gis main.gcc296 main.flush_table 2017-07-13 15:35:31 140219388360448 [Warning] Sort aborted, host: localhost, user: root, thread: 164, query: select numeropost as a FROM t1 ORDER BY (SELECT 1 FROM t1 HAVING a=1) 2017-07-13 15:35:31 140219388360448 [Warning] Sort aborted, host: localhost, user: root, thread: 164, query: SELECT a FROM t1 GROUP BY a 2017-07-13 15:35:31 140219388360448 [Warning] Sort aborted, host: localhost, user: root, thread: 164, query: SELECT a FROM t1 GROUP BY a 2017-07-13 15:35:31 140219388360448 [Warning] Sort aborted, host: localhost, user: root, thread: 164, query: SELECT a FROM t1 2017-07-13 15:35:31 140219388360448 [Warning] Sort aborted, host: localhost, user: root, thread: 164, query: SELECT a FROM t1

1 0

Re: [Maria-discuss] 10.2.7 and xtradb in changelog
by Federico Razzoli 13 Jul '17

13 Jul '17

I don't find this extremely confusing (if it's a change to the code, it's not weird to see it in the changelog; filtering out things exposes us to bugs in the filtering process, which may hide important information) but I think that there is another confusing part. In this page: https://mariadb.com/kb/en/mariadb/why-does-mariadb-102-use-innodb-instead-o… I read: "Note, we still consider incorporating XtraDB optimizations, but as patches, rather than XtraDB as a whole" But then, it is not clear what exactly is or is not in MariaDB's InnoDB. Percona has a nice list of features: https://www.percona.com/doc/percona-server/LATEST/index.html It would be useful to have a list of which changes are applied to MariaDB's InnoDB. Hope it is not too hard to maintain. Cheers, Federico -------------------------------------------- Gio 13/7/17, Sergei Golubchik <serg(a)mariadb.org> ha scritto: Oggetto: Re: [Maria-discuss] 10.2.7 and xtradb in changelog A: "Reindl Harald" <h.reindl(a)thelounge.net> Cc: "Mailing-List mariadb" <maria-discuss(a)lists.launchpad.net> Data: Giovedì 13 luglio 2017, 12:35 Hi, Reindl! On Jul 13, Reindl Harald wrote: > Am 13.07.2017 um 11:55 schrieb Sergei Golubchik: > > On Jul 13, Reindl Harald wrote: > >> https://mariadb.com/kb/en/mariadb/mariadb-1027-changelog/ > >> > >> i thought xtradb is gone in 10.2.x > >> > >> is it back now with 10.2.7 or is the changelog a random wild mix of > >> commits with no context to the version of the release? > > > > Neither. Changelog is generated as "all commits that are present in the > > 10.2.7 history, but were not present in the 10.2.6 history". And all > > commits in 10.1 are merged into 10.2 eventually (because many 10.1 > > bugfixes apply to 10.2 too). > > > > So whatever was done to XtraDB in 10.1 it will be part of 10.2 revision > > history, even though XtraDB itself is disabled in 10.2 (and does not > > even compile there). > > > > I agree that it's confusing, but what can we do? Manually filter out all > > commits that contain the string "xtradb" from the changelog? > > well, at least some filters make sense because things like "Update > xtradb and innodb version to 5.6.36" are pure nonsense when innodb in > 10.2 AFAIK is 5.7 and so the complete changelog feels very useless and > instead beeing helpful it spreads doubt how that whole stuff works at > all when random pieces seems to get merged bewteen versions or at least > the changelog says so Okay. https://jira.mariadb.org/browse/MDEV-13312 Regards, Sergei Chief Architect MariaDB and security(a)mariadb.org _______________________________________________ Mailing list: https://launchpad.net/~maria-discuss Post to : maria-discuss(a)lists.launchpad.net Unsubscribe : https://launchpad.net/~maria-discuss More help : https://help.launchpad.net/ListHelp

4 3

[Maria-discuss] mariadb-covscan-stroverflow.patch (Fedora)
by Reindl Harald 13 Jul '17

13 Jul '17

how long does Fedora need to ship that patch which seems to apply still to 10.2.7 where i am asking mysqlf for years if i should include it in my own builds or trust upstream

2 1

[Maria-discuss] 10.2.7: innodb_flush_method = ALL_O_DIRECT
by Reindl Harald 13 Jul '17

13 Jul '17

it's one thing that xtradb is gone but when the server refuses to start instead just warn about no longer supported options this is *very bad* 2017-07-13 12:22:57 140560655323200 [ERROR] InnoDB: Unrecognized value ALL_O_DIRECT for innodb_flush_method 2017-07-13 12:22:57 140560655323200 [Note] InnoDB: Number of pools: 1 2017-07-13 12:22:57 140560655323200 [Note] InnoDB: Using SSE2 crc32 instructions 2017-07-13 12:22:57 140560655323200 [ERROR] InnoDB: Plugin initialization aborted with error Generic error 2017-07-13 12:22:57 140560655323200 [Note] InnoDB: Starting shutdown... 2017-07-13 12:22:57 140560655323200 [ERROR] Plugin 'InnoDB' init function returned error. 2017-07-13 12:22:57 140560655323200 [ERROR] Plugin 'InnoDB' registration as a STORAGE ENGINE failed. 2017-07-13 12:22:57 140560655323200 [ERROR] Unknown/unsupported storage engine: innodb _______________________________ BTW: "InnoDB: Cannot open '/Volumes/dune/mysql_dbmail/ib_buffer_pool' for reading: No such file or directory" - who cares when "innodb_buffer_pool_dump_at_shutdown = 0" is set on purpose because writing some GB buffer_pool at shutdown takes longer than the whole reboot of a typical server these days

1 0

[Maria-discuss] JSON in MariaDB 10.2.7 and MySQL
by Peter Laursen 13 Jul '17

13 Jul '17

-- MariaDB 10.2.7 CREATE TABLE `jsontst`( `id` INT NOT NULL AUTO_INCREMENT, `str` JSON, PRIMARY KEY (`id`) ) ENGINE=Aria CHARSET=latin1; SHOW CREATE TABLE `jsontst`; /* returns CREATE TABLE `jsontst2` ( `id` int(11) NOT NULL AUTO_INCREMENT, `str` text DEFAULT NULL, PRIMARY KEY (`id`) ) ENGINE=Aria DEFAULT CHARSET=latin1 PAGE_CHECKSUM=1 */ readings: https://jira.mariadb.org/browse/MDEV-9144 https://bugs.mysql.com/bug.php?id=86875 As you see except for JSON in MariaDB just being a synonym of TEXT there are two more significant differences 1) length: in MySQL JSON has the lenght of a LONGTEXT - in MariaDB it is a (plain) TEXT 2) encoding: in MySQL the utf8mb4 character set is enforced for JSON data - in MariaDB you may specify any character set. I think in particular 2nd point could cause problems (unicode garbling) when loading a dump with JSON data from MySQL. Basically because latin1 is still default character set in MariaDB. Are you aware of the 2 differences? I think the major reason for introducing JSON in MariaDB is compability concerns with MySQL. But the 2 points here are IMO incompabilites that could have been avoided?

2 4

[Maria-discuss] 10.2.7 and xtradb in changelog
by Reindl Harald 13 Jul '17

13 Jul '17

https://mariadb.com/kb/en/mariadb/mariadb-1027-changelog/ i thought xtradb is gone in 10.2.x is it back now with 10.2.7 or is the changelog a random wild mix of commits with no context to the version of the release? ______________________________ Revision #546a89ca58 2017-05-18 16:16:18 +0300 Update xtradb and innodb version to 5.6.36 Revision #f880200974 2017-05-20 17:47:07 +0300 After-merge fix: Correct the 32-bit XtraDB result diff. Revision #fe291c687d 2017-05-20 08:43:25 +0300 After-merge fix: Correct the XtraDB result diff. Revision #4df726e180 2017-07-10 00:05:06 +0300 storage_engine subsuites should also be disabled for XtraDB

3 5

[Maria-discuss] incomplete provide of symlinks with 10.2.7
by Reindl Harald 13 Jul '17

13 Jul '17

Fedora just started to work on MariaDB 10.2 packaging they create compat symlinks for a sane transition and i try to backport that to my F25 builds with change as less as possible on my existing "mariadb.spec" 10.2.7 seems to do the same now (Fedora last night was still on 10.2.6) well, "libmysqlclient.so.18" is missing in your upstream compat links while "libmysqlclient.so" and "libmysqlclient_r.so" gave an error "ln: failed to create symbolic link 'libmysqlclient.so': File exists" within rpmbuild https://koji.fedoraproject.org/koji/buildinfo?buildID=918138 ___________________________________________ pushd %{buildroot}%{_libdir}/mysql/ ln -s libmariadb.so libmysqlclient.so ln -s libmariadb.so libmysqlclient.so.18 ln -s libmariadb.so libmysqlclient_r.so popd ___________________________________________ + pushd /home/builduser/rpmbuild/BUILDROOT/mariadb-10.2.7-3.fc25.20170713.rh.x86_64/usr/lib64/mysql/ ~/rpmbuild/BUILDROOT/mariadb-10.2.7-3.fc25.20170713.rh.x86_64/usr/lib64/mysql ~/rpmbuild/BUILD/mariadb-10.2.7 + ln -s libmariadb.so libmysqlclient.so.18 + popd ~/rpmbuild/BUILD/

2 2

[Maria-discuss] MariaDB 10.2.7 now available
by Daniel Bartholomew 12 Jul '17

12 Jul '17

The MariaDB project is pleased to announce the immediate availability of MariaDB 10.2.7. This is a stable (GA) release. See the Release Notes and Changelogs for details. - - Links - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - MariaDB 10.2.7 - Release Notes: https://mariadb.com/kb/en/mdb-1027-rn/ - Changelog: https://mariadb.com/kb/en/mdb-1027-cl/ - Downloads: https://downloads.mariadb.org/mariadb/10.2.7 About MariaDB 10.2: - https://mariadb.com/kb/en/what-is-mariadb-102/ APT and YUM Repository Configuration Generator: - https://downloads.mariadb.org/mariadb/repositories/ - - MariaDB Webinars - - - - - - - - - - - - - - - - - - - - - - - - - There are several upcoming MariaDB-focused webinars and many previously held ones available to watch on an on-demand basis at: - https://mariadb.com/news-events/webinars - - MariaDB Books - - - - - - - - - - - - - - - - - - - - - - - - - - There is an ever-growing library of MariaDB books available to help you get the most out of MariaDB. See the MariaDB Books page for details and links: - https://mariadb.com/kb/en/mariadb/books/ - - User Feedback plugin - - - - - - - - - - - - - - - - - - - - - - - MariaDB includes a User Feedback plugin. This plugin is disabled by default. If enabled, it submits basic, completely anonymous MariaDB usage information. This information is used by the developers to track trends in MariaDB usage to better guide development efforts. If you would like to help make MariaDB better, please add "feedback=ON" to your my.cnf or my.ini file! See http://mariadb.com/kb/en/user-feedback-plugin for more information. - - Quality - - - - - - - - - - - - - - - - - - - - - - - - - - - - - The project always strives for quality, but in reality, nothing is perfect. Please take time to report any issues you encounter at: - http://mariadb.org/jira - - Support MariaDB - - - - - - - - - - - - - - - - - - - - - - - - - If you would like to contribute to the MariaDB Foundation, please see the "contributing" and "donations" pages. We also have merchandise available in a cafepress store. All proceeds go to support the MariaDB Foundation. - https://mariadb.com/kb/en/contributing - https://mariadb.org/donate/ We hope you enjoy MariaDB! -- MariaDB Website - http://mariadb.org Twitter - http://twitter.com/mariadbfdn - http://twitter.com/mariadb Google+ - http://google.com/+mariadb Facebook - http://fb.com/MariaDB.dbms Knowledge Base - http://mariadb.com/kb

1 0

[Maria-discuss] Galera cluster for databases replication over three servers
by Martin LEUSCH 11 Jul '17

11 Jul '17

Hi, I'm setting up a new mail infrastructure withe 3 servers running Debian Jessie, data can be read or write from these 3 servers at any time. I chosed Galera with with MariaDB 10.2 to share databases between servers for services like AMAVIS, dovecot, ... I had a lot of errors in my logs (almost each time a service access to a DB), for example: Jul 11 07:00:01 mail mysqld[1302]: 2017-07-11 7:00:01 140256990783232 [Warning] Aborted connection 14812 to db: 'sogo' user: 'sogo' host: 'localhost' (Got an error reading communication packets) Jul 11 10:15:17 mailfilter mysqld[3714]: 2017-07-11 10:15:17 139644624377600 [Warning] Aborted connection 399 to db: 'amavisd' user: 'amavisd' host: 'localhost' (Got timeout reading communication packets) I wonder if in this case Galera is a good choice for databases replication and if there is a way to avoid this kind of errors. Thanks -- Martin LEUSCH <martin.leusch(a)imrcom.fr>

1 0