cppop dying

[nxds]

Over the past few nights the cppop daemon has died on a number of our servers. This has never happend before. Does anyone have any idea what might be causing this behaviour?

[randomuser]

Check /var/log/messages, always, and say what version of cPanel is being used on the problematic servers.

I saw this recently as well on someone's server that was running a very outdated version of cPanel. /var/log/messages showed something about a file in /etc/xinetd.d/ not being executable or something. /scripts/upcp resolved the issue automagically, so I didn't bother to investigate further.

Please say which version of cPanel this is happening on.
$ cat /usr/local/cpanel/version

[nxds]

This has happened on several servers running 10.9.0-RELEASE_34. This has only started to happen since last weekend.

I thought at first it was related to a faulty cpanel update (much like the one causing BIND to fail) but it has happened outside the time frame when upcp is run.

There is nothing in /var/log/messages or /var/log/maillog to indicate the cause of the problem.

I have enabled cpservd so that the service gets restarted; however I would like to know why it's dying. Some remote DOS attack? Remote exploit attempt?

[protocol]

I have seen this also one one Trustix server - other servers we have (CentOS) seem fine.

[nxds]

I have seen this also one one Trustix server - other servers we have (CentOS) seem fine.

Mine are all CentOS 4.4 with latest kernel

[chirpy]

cppop has never been particularly stable. I don't know if cPanel are even developing it any more now that it is deprecated. It's probably a good opportunity to consider migrating over to maildir and courier-imap (pop3) instead.