This amount of failures …above, below or on average?

Question

It seems like chkservd is frequently having to restart services on our server and I was curious if the frequency was common or if we should be concerned with our VPS's reliability. I.E., on a "good" server are service restarts common every couple weeks or should they not even happen that often?

Here's a summary of the frequency of failures + restarts for the month of July:

July 30 named
July 20 httpd
July 19 httpd
July 16 httpd
July 10 httpd
July 8  imap, mysql, spamd
July 6  named
July 3  named

Server details:

cPanel 11.24.5-R37629 - WHM 11.24.2 - X 3.9
CENTOS 5.3 i686 virtuozzo on host
VPS Hybrid Server

Answer 1

I would hazard that memory is the culprit.

To confirm, check dmesg and /var/log/messages.

Look especially for oom-killer entries.

Answer 2

Services (Daemons really) going down is a bad thing sure, but they do sometimes do that, especially for these reasons: running out of memory, failing to get locks on files (disk space or nas issues), hard locking (often cpu cooling issues or really bleeding edge release use).

I suspect the first.

Answer 3

If your services are going down ever, something is wrong. Period. Why are they failing? Is there any more detail in the logs?