# mysql percona slowdown **URL:** \ **Category:** Other MySQL® Questions\ **Created:** [September 6, 2013, 4:06am UTC](https://forums.percona.com/t/mysql-percona-slowdown/2926 "2013-09-06T04:06:57Z")\ **Posts on this page:** 3\ **Page:** 1
**Author:** ![stormdust](https://avatars.discourse-cdn.com/v4/letter/s/6f9a4e/32.png) [@stormdust](https://forums.percona.com/u/stormdust)\ **Post date:** [September 6, 2013, 4:06am UTC](https://forums.percona.com/t/mysql-percona-slowdown/2926/1 "2013-09-06T04:06:57Z")
Hi there people. Im having a damn dificult problem with one mysql server that keeps getting slowing down when it gets spikes of heavy queries. We have our project mounted like this: -frontend in iis 7.5 and framework 4.5 -api in iis 7.5 with wcf and net.pipe(frontend) and mysql connector 6.3(DB) -Database with mysql Percona Server (GPL), Release 31.0 (5.5.32-rel31.0) When i have spikes on the api or when im deploying the project api sometimes everthing stops responding, when i go evaluate what is happening i see this: -frontend low cpu (large queue request) -api low cpu (large queue requests) -mysql slow down (many queries that run in 0 to 100 milliseconds → pass to 1second-5seconds) When this happen the last time i try the following: -i put the front end in schedule for not to receive any requests -i put the api in schedule for not to receive any requests -i have killed the front end w3wp process to clean up -i have killed the api w3wp process to clean up -i have killed all the processes on the mysql Try to start everthing but the mysql didnt recover, keeps slowing, the only way i have to stabilise the project is with a restart on the mysql server. Data / Requests per second -20 in average -by request i have 10-20 storeprocedures invokes -mysql server is with 3.5G´s RAM ocupied and swap is at 4Gb´s Configurations on mysql percona server Server virtualized 4 CPU Cores (6GHZ´s) 4Gb´s Ram 40Gb´s virtual disc one Database with 4Gb´s of information 200 tables 3000 Storeprocedures [mysql] [mysqld] # GENERAL group\_concat\_max\_len = 65535 user = mysql default\_storage\_engine = InnoDB socket = /var/run/mysqld/mysqld.sock pid\_file = /var/run/mysqld/mysqld.pid skip-external-locking wait\_timeout = 30 interactive\_timeout = 30 max\_connections = 50 thread\_concurrency = 8 # MyISAM key\_buffer\_size = 256M myisam\_recover = FORCE,BACKUP # SAFETY max\_allowed\_packet = 16M max\_connect\_errors = 1000000 # DATA STORAGE datadir = /mnt/data/mysql tmpdir = /tmp # BINARY LOGGING log\_bin = /mnt/data/mysql/mysql-bin binlog\_format = ‘ROW’ expire\_logs\_days = 1 sync\_binlog = 1 # CACHES AND LIMITS tmp\_table\_size = 128M max\_heap\_table\_size = 128M query\_cache\_type = 0 query\_cache\_size = 0 thread\_cache\_size = 50 open\_files\_limit = 65535 table\_definition\_cache = 4096 table\_open\_cache = 10240 myisam\_sort\_buffer\_size = 64M # INNODB innodb\_flush\_method = O\_DIRECT innodb\_file\_per\_table = 1 innodb\_flush\_log\_at\_trx\_commit = 1 innodb\_buffer\_pool\_size = 2G # LOGGING general-log-file = /var/log/mysql/mysql.log log\_error = /var/log/mysql/error.log slow\_query\_log\_file = /var/log/mysql/slow.log log\_queries\_not\_using\_indexes = 0 slow\_query\_log = 2 # TEXT SEARCH ft\_min\_word\_len = 2 # OTHERS key\_buffer = 32M read\_buffer\_size = 2M read\_rnd\_buffer\_size = 16M bulk\_insert\_buffer\_size = 64M max\_allowed\_packet = 16M [mysqldump] quick max\_allowed\_packet = 32M [myisamchk] key\_buffer\_size = 256M sort\_buffer\_size = 256M read\_buffer = 2M write\_buffer = 2M [mysqlhotcopy] interactive-timeout any thoughs where i should go to try catch what is causing this on my dabase server? ---
**Author:** ![przemek](https://sea1.discourse-cdn.com/flex019/user_avatar/forums.percona.com/przemek/32/3_2.png) [@przemek](https://forums.percona.com/u/przemek)\ **Post date:** [September 10, 2013, 5:25am UTC](https://forums.percona.com/t/mysql-percona-slowdown/2926/2 "2013-09-10T05:25:52Z")
So the database server lives on dedicated VM, right? What did you observe there? CPU bound, IO bound? What about memory, was the server swapping during the slowdown? Our tools like [pt-stalk](http://www.percona.com/doc/percona-toolkit/2.2/pt-stalk.html) run in critical moment could provide you many details about the situation. Another thing is some of your per session buffers really high. What was the reason you set : read\_buffer\_size = 2M read\_rnd\_buffer\_size = 16M ? Check this article: [URL][http://ronaldbradford.com/blog/dont-assume-per-session-buffers-2010-03-08/[/URL]](http://ronaldbradford.com/blog/dont-assume-per-session-buffers-2010-03-08/%5B/URL%5D) ---
**Author:** ![stormdust](https://avatars.discourse-cdn.com/v4/letter/s/6f9a4e/32.png) [@stormdust](https://forums.percona.com/u/stormdust)\ **Post date:** [January 25, 2016, 6:49am UTC](https://forums.percona.com/t/mysql-percona-slowdown/2926/3 "2016-01-25T06:49:49Z")
The problem was in the connector, i made some tweeks on it to improve it and this problem disapered.