- discuss - lists.mariadb.org

[Maria-discuss] New Question: Slow query log and long_query_time
by AskMonty KB 18 Dec '15

18 Dec '15

Hello, A new question has been asked in "Slow Query Log" by yu.d.sun_32615. Please answer it at http://mariadb.com/kb/en/slow-query-log-and-long_query_time/ as the person asking the question may not be subscribed to the mailing list. -------------------------------- In my understanding if we enable slow query log, MySQL will capture all SQL statements that take more than long_query_time seconds to execute into the log. If so below testing puzzled me: Server version: 10.0.21-MariaDB-wsrep-log MariaDB Server, wsrep_25.10.r4144 === in the mysql CLI ================== ip-10-74-2-218$ root@localhost[galeratest] MYSQL> select now(); +---------------------+ | now() | +---------------------+ | 2015-12-18 11:12:50 | +---------------------+ 1 row in set (0.00 sec) ip-10-74-2-218$ root@localhost[galeratest] MYSQL> show variables like 'long%'; +-----------------+-----------+ | Variable_name | Value | +-----------------+-----------+ | long_query_time | 10.000000 | +-----------------+-----------+ 1 row in set (0.00 sec) ip-10-74-2-218$ root@localhost[galeratest] MYSQL> select count(*) from test_table; +----------+ | count(*) | +----------+ | 13 | +----------+ 1 row in set (0.00 sec) ip-10-74-2-218$ root@localhost[galeratest] MYSQL> select sleep(5); +----------+ | sleep(5) | +----------+ | 0 | +----------+ 1 row in set (5.00 sec) ip-10-74-2-218$ root@localhost[galeratest] MYSQL> select sleep(11); +-----------+ | sleep(11) | +-----------+ | 0 | +-----------+ 1 row in set (11.00 sec) ip-10-74-2-218$ root@localhost[galeratest] MYSQL> select count(*) from test_table; +----------+ | count(*) | +----------+ | 13 | +----------+ 1 row in set (0.00 sec) ip-10-74-2-218$ root@localhost[galeratest] MYSQL> ==== from the slow query log ================== # Time: 151218 11:13:12 # User@Host: root[root] @ localhost [] # Thread_id: 125 Schema: galeratest QC_hit: No # Query_time: 0.000163 Lock_time: 0.000075 Rows_sent: 1 Rows_examined: 13 SET timestamp=1450465992; select count(*) from test_table; # Time: 151218 11:13:48 # User@Host: root[root] @ localhost [] # Thread_id: 125 Schema: galeratest QC_hit: No # Query_time: 11.002494 Lock_time: 0.000000 Rows_sent: 1 Rows_examined: 0 SET timestamp=1450466028; select sleep(11); # Time: 151218 11:14:10 # User@Host: root[root] @ localhost [] # Thread_id: 125 Schema: galeratest QC_hit: No # Query_time: 0.000159 Lock_time: 0.000073 Rows_sent: 1 Rows_examined: 13 SET timestamp=1450466050; select count(*) from test_table; ================ Question - why the 'select count(*) from test_table' statement is captured while it seems right that select sleep(11) got captured and sleep(5) did not? -------------------------------- To view or answer this question please visit: http://mariadb.com/kb/en/slow-query-log-and-long_query_time/

1 0

[Maria-discuss] New mariadb.org in the works
by Otto Kekäläinen 18 Dec '15

18 Dec '15

Hello! We are about to renew the mariadb.org website. The idea is to replace the current mariadb.org and blog.mariadb.org with one single site that includes both the basic information about the software project and the MariaDB Foundation, as well as the blog. Work in progress is visible at https://new.mariadb.org/ Please send us your feedback!

1 1

[Maria-discuss] Amazon Aurora: based on MySQL or MariaDB?
by Lukas Lehner 17 Dec '15

17 Dec '15

Hi https://aws.amazon.com/rds/aurora/ Did Amazon write everything on their own? Is it based on MySQL or MariaDB? Lukas

3 2

[Maria-discuss] CONNECT version maturity
by Federico Razzoli 17 Dec '15

17 Dec '15

Hi all In this page: https://mariadb.com/kb/en/mariadb/connect/ I see that CONNECT 1.03.0007 (that is in MariaDB 10.1.5) is gamma-quality, but 1.04.0003 (that is in 10.1.9) is beta. I am confused. Is this a documentation mistake? Or maybe many new bugs were found? Regards Federico

2 1

[Maria-discuss] New Question: Does MariaDB Galera Cluster 10.0.21 version support gmcast.segment?
by AskMonty KB 17 Dec '15

17 Dec '15

Hello, A new question has been asked in "MariaDB Galera Cluster" by yu.d.sun_32615. Please answer it at http://mariadb.com/kb/en/does-mariadb-galera-cluster-10021-version-support-… as the person asking the question may not be subscribed to the mailing list. -------------------------------- Hi, Server version: 10.0.21-MariaDB-wsrep-log MariaDB Server, wsrep_25.10.r4144 ip-10-74-2-218$ root@localhost[(none)] MYSQL> set global wsrep_provider_options="gmcast.segment=1"; ERROR 1210 (HY000): Incorrect arguments to SET I also tried to put wsrep_provider_options="gmcast.segment=1" in the my.cnf but cannot start the instance: 151217 7:59:58 [ERROR] WSREP: Unrecognized parameter '"gmcast.segment' 151217 7:59:58 [ERROR] WSREP: wsrep::init() failed: 7, must shutdown 151217 7:59:58 [ERROR] Aborting Denis -------------------------------- To view or answer this question please visit: http://mariadb.com/kb/en/does-mariadb-galera-cluster-10021-version-support-…

1 0

[Maria-discuss] max key length in import
by Warren H. Prince 17 Dec '15

17 Dec '15

I’m considering moving from MySql to MariaDB so I did a dump today and attempted an import. It failed with the “Specified key was too long; max key length is 767 bytes” error. I dealt with this in MySql and knew to change file format to Barracuda. I zapped the partial import, did a fresh mysql_install_db, restarted the service and tried again to no avail. To confirm my settings immediately after the failure: MariaDB [(none)]> SHOW VARIABLES LIKE 'innodb_file_%'; +--------------------------+-----------+ | Variable_name | Value | +--------------------------+-----------+ | innodb_file_format | Barracuda | | innodb_file_format_check | ON | | innodb_file_format_max | Barracuda | | innodb_file_per_table | ON | +--------------------------+—————+ Does anyone have any suggestions? The dump is 795M compressed so it’s difficult to find the exact place where the problem is (I used zcat for the import). My distribution is: 10.0.22-MariaDB running on Ubuntu 15.04 Thanks in advance.

2 2

[Maria-discuss] New Question: Galera Cluster expire_logs_days
by AskMonty KB 16 Dec '15

16 Dec '15

Hello, A new question has been asked in "MariaDB Galera Cluster" by diego.carrera. Please answer it at http://mariadb.com/kb/en/galera-cluster-expire_logs_days/ as the person asking the question may not be subscribed to the mailing list. -------------------------------- Is it safe to modify the "expire_logs_days" variable while the cluster is running? show variables like 'expire_log_days' is 14, I've not defined this under my.cnf or wsrep.cnf. What would be the proper procedure to reduce or increase binlog expiry? -------------------------------- To view or answer this question please visit: http://mariadb.com/kb/en/galera-cluster-expire_logs_days/

1 0

[Maria-discuss] Replication variables in my.cnf seem ignored if master connection does not exist
by Игорь Пашев 16 Dec '15

16 Dec '15

MariaDB 10.0.22 I saw weird things a couple of times: I start clean MariaDB with theses options in the configuration file: foo.replicate_wild_do_table = foo_live.% foo.replicate_ignore_table = foo_live.schema_updates Then I execute RESET SLAVE "foo" ALL just to be sure there is no such connection. Then I import the dump of master, and set master options like host, user, password, etc. Then I execute START SLAVE "foo" Then SHOW SLAVE "foo" reveals empty 'Replicate_Wild_Do_Table' and 'Replicate_Ignore_Table'. (And replication shortly breaks, because foo_live.schema_updates wasn't imported). I can fix this by restarting MariaDB. I'd expect the foo.replicate_* options are taking effect anytime. As a work-around I'd set those options dynamically after dump is imported. (I will fill MDEV in case it is confirmed as a bug)

1 1

[Maria-discuss] MariaDB Performance and Diagnostic Tools
by FERRETTI, ANDY 15 Dec '15

15 Dec '15

I'm looking for recommendations on MariaDB Performance Analysis tools. 1. I try MySQL workbench, but much of the capability is disabled. I get an error when I click on the Dashboard "This feature requires MySQL version 5.6.6 or newer". 2. I'm wondering if there are some good scripts to help identify key performance metrics from the performance_schema. I'd like to see a capability similar to an Oracle AWR report. I spent a fair amount of time doing google searches, and found some information on the performance_schema: https://php-built.com/enabling-performance-monitoring-for-mariadb-and-mysql/ , but I really think there should be something better. Thanks, Andy Ferretti Lead Principal-Technical Architect AT&T Technology Development - SD&E Cloud, Platform, Application and Data Layer Support (CPADS) Work: 205 989 0662 Cell: 205 478 9694 e-mail: andy.ferretti(a)att.com<mailto:andy.ferretti@att.com> This e-mail and any files transmitted with it are AT&T property, are confidential, and are intended solely for the use of the individual or entity to whom this emailis addressed. If you are not one of the named recipient(s) or otherwise have reason to believe that you have received this message in error, please notify the sender and delete this message immediately from your computer. Any other use, retention, dissemination, forwarding, printing, or copying of this e-mail is strictly prohibited.

5 4

[Maria-discuss] Galera Cluster: Cluster Blocked, when one node down?
by Benoit Panizzon 14 Dec '15

14 Dec '15

Hello We use MariaDB Galera Cluster for our email service platform. We decided to use Galera to create a high availability platform. After a year of operation, we start to relaize, that somehow Galera Failures seem to be the most common cause for outages we had in the past. So I wonder if others operating galera clusters also observe this situation: All our services using DB connections use a DNS round-robin name, to connect to one of our three galera instances. While testing this setup, we usualy killed one instance, or disconnected the node from the network to simulate an outage. In this situation, this works as expected. The client connect to the two remaining nodes, no service outage. When the node is re-started it is being re-synced quickly and service with three nodes is restored. Now we experienced a few galera cluster fails, which seem to happen this way: One of the nodes is getting a lot of load. DDOS Attacks, Memory Leaks or similar, which just renders the whole physical machine laggy for a short time. So the affected MariaDB node is being thrown out of the cluster by the two other nodes, probably for not syncing fast enough anymore. But as the node is not 'down' completely, it still accepts connections from the DB clients, but does not reply to them and seems to remain in a 'db locked' situation. Strangely this then also affects the two remaining nodes, who also go into 'locked' mode and do not reply to queries on the time expected by the application anymore. Of course this then causes more DB clients (IMAP, SMTP-Auth, etc) to spawn and to create DB connections worsening the whole situation. The situation seemingly can only be resolved by shuting down the MariaDB node that got thrown out of the cluster. Then the situations normalizes with the two remaining nodes and the third one can be restarted. Is this expected behaviour? Is there a way to tell a MariaDB node that got excluded from the cluster to shut himself down completely so it does NOT accept any more connections from clients, blocking the whole service? Regards -Benoît Panizzon- -- I m p r o W a r e A G - Leiter Commerce Kunden ______________________________________________________ Zurlindenstrasse 29 Tel +41 61 826 93 00 CH-4133 Pratteln Fax +41 61 826 93 01 Schweiz Web http://www.imp.ch ______________________________________________________

3 2