How to enable log-slow-queries in mysql?

futurevision

Verified User
Joined
Nov 26, 2008
Messages
51
Hi,

I'm seeing some high loads on the server with mysql on top with 98% cpu usage and I want to know what the cause is. I suspect some crappy query in a non-optimized table from one of my clients. So, I read about the --log-slow-queries option but I can't figure out how to kill them mysql processes without breaking all websites on the server.

Versions
OS: CentOS 5.5
MySQL: Ver 5.0.67-community for pc-linux-gnu on i686 (MySQL Community Edition (GPL))

What I tried: /usr/sbin/mysqld --log-slow-queries (under root)
and gave this output:

---
120404 20:46:12 [Warning] Asked for 196608 thread stack, but got 126976
120404 20:46:12 [ERROR] Fatal error: Please read "Security" section of the manua l to find out how to run mysqld as root!

120404 20:46:12 [ERROR] Aborting

120404 20:46:12 [Note] /usr/sbin/mysqld: Shutdown complete
---

I think I'm quite close but I just don't understand how I can gracefully restart mysql with the log option without breaking all running sites. Any help is welcome, tnx :)
 
Thanks but:
- in what file do I put those lines?
- how do I properly restart mysqld?

The docs you point to are for mysql 5.5, I'm running 5.0 so I guess I have to do something different, according to the manual I have to start mysqld with --log-slow-queries but I dont know how.
 
for 5.5:
slow_query_log=1 # (1 = on, 0 = off)
slow_query_log_file=slow_query_log.log # (optinal, file name, not full path)
long_query_time=5 # (in seconds)

for 5.0:
log_slow_queries=slow_query_log.log
long_query_time=5 # (in seconds)

put this in your /etc/my.cnf

to restart mysql:
/etc/init.d/mysqld restart
 
Last edited:
Thought so, thanks anyway :)

BestBoard: I did that and everything seems to be working again, thanks! Now I hope I will see what queries cause the loads from time to time.
 
Back
Top