I am currently having very severe problems with my Linux CentOS dedicated server. For some peculiar reason, all of my memory is being consumed by apache and resultantly, my qmail process is crashing.
Firstly, why would qmail crash if apache is eating all of the memory?
How can I get to the bottom of what is causing this issue? I am being helped by the support team at Media Temple, but this is out of their scope of support so I am unsure of what else they can help me with.
I am really quite confused with this as it there seems to be no logical explanation for why the memory is being used up. I am not receiving any more traffic on my websites than usual
To give you a little info on the situation:
- It seems that all of this started when I updated Plesk and my container. However, I cannot be certain about this and it is likely I did other things on the same date to
- When I look at the beancounters, it says that there is a failcnt of 547225 in the tcpsndbuf. I do not know what this means but this is the reason for me believing the problem is with Apache. Can I research into this futher?
- Although probably irrelevant, a few weeks ago I used the Apache Benchmarking tool to simulate 1000 visits to one of my websites.
- A few weeks ago I also followed the Media Temple 'Making it better steps' to install Monit, Munin, Plesk Health Monitor and general Apache and MySQL tuning. These steps have undoubtedly had a reverse effect on the server as since then, my memory consumption has been much higher. I tried to revert to the previous conf files but this did not make a difference.
I have spent ages trying to get to the bottom of this but I think the issue is a little too advanced for me. Does any body have any advice?
I forgot to mention that every now and again, my CPU spikes. Not sure if that is relevant but thought I would include it.