New on LowEndTalk? Please Register and read our Community Rules.
All new Registrations are manually reviewed and approved, so a short delay after registration may occur before your account becomes active.
All new Registrations are manually reviewed and approved, so a short delay after registration may occur before your account becomes active.
Memory management problems leading to Webmin going down on small VPS?
MTUser2012
Member
in Help
I have a small vps 512MB with 64mb swap with Virtualmin installed and a single WP site. Webmin and bind keep going down on this server. The main memory only shows around 250mb used, but the swap file always shows 64mb used.
Any suggestions to keep Webmin up and make the VPS use the main memory available and not the swap file?
Comments
Playing with
swappinessvalues may be useful here.Your case may be different but for me webmin barely runs flawless with my 1GB VPS...
I presume Mysql overtime with Default configs tends to go across 600-800mb, and then the Webmin/Virtualmin service stops itself.
Rebooting the server will fix it for some days... but for a permanent solution you must either do some real fine tuning to mysql (which I'm also learning) or get a bigger 2GB VPS.
Need apache, mysql and php optimizing.
Could be related to apache max clients, had a similar problem with WP on a small memory box. Apache kicks off too many child processes when traffic increases. Lower to maybe 10 to begin with.
I think mysql is the culprit, when you install webmin will ask you how much memory to allocate for it, go with the minimum. Even so, manual tuning is always better.
I think webmin not used a lot of RAM, i used webmin in my 256 RAM.
Try to convert innoDB into MyISAM
Until you optimize, as suggested above ... On the webmin menu select "System and Server Status" where you can add a webmin and bind monitor that will restart them after an oom event.
Have you tried optimising the server? - a complete overhaul and configuration maybe worth while
Use monit.
Check it for running processes every X seconds. If a process you need is not working, tell it to restart the process.
Also try playing with sysctl for swappiness.
Had this very problem (webmin being killed on a 1GB (!) OVZ VPS). You might want to try
and have a look at the last column (failcnt) for privvmpages. Explanation is here:
http://wiki.openvz.org/Resource_shortage
After the provider increased the privvmpages limit, everything worked fine for me.
This is just a shot in the dark, your problem might be completely unrelated....
Thank you for all the great suggestions. I'll check them out. I have a done a little mysql tuning previously using mysqltuner.pl.
Give this a try as well -- http://www.mysqlmymon.com/
extremely usefultool to find the bottleneck , it uses the same old unix commands , but presents them in a better manner for easier understanding.