Need post mortem help after server slows to a halt every 24 hours

jcase · 03-26-2007, 04:58 AM

Hi, the server is Plesk hosted and uname -r reports 2.6.10-1.771_FC2. Its running spamassassin and hosting websites for about 30 domains, some of which are development server-side java (resin).

Whats happening is it slowly grids to a halt every 24 hours, at which point no connections can be made:I can't even ssh in. I have to use the Plesk web panel to reboot, after which it flies (but only for about another day). I need broad advice on which logs to look at after this restart. Could abnormal spamassasin load be overloading it? Is a java webapp eating all the RAM? etc. Or, something I could put in place to watch for vital signs during its uptime. Very grateful for any suggestions, you may gather my hosting company have left me to it (its a dedicated server).

sandgroper · 03-26-2007, 10:24 AM

Quote:

Originally Posted by jcase

Hi, the server is Plesk hosted and uname -r reports 2.6.10-1.771_FC2. Its running spamassassin and hosting websites for about 30 domains, some of which are development server-side java (resin).

Whats happening is it slowly grids to a halt every 24 hours, at which point no connections can be made:I can't even ssh in. I have to use the Plesk web panel to reboot, after which it flies (but only for about another day). I need broad advice on which logs to look at after this restart. Could abnormal spamassasin load be overloading it? Is a java webapp eating all the RAM? etc. Or, something I could put in place to watch for vital signs during its uptime. Very grateful for any suggestions, you may gather my hosting company have left me to it (its a dedicated server).

Check the size of your log files.
I have heard of a similar problem with a server that slowly grinds to a halt and needs to be rebooted because of very large log files .

.