[SOLVED] Server crashed - Why?

[bertalanimre]

Dear all,

I have a VPS as webserver with NginX, PHP7, MySQL, etc. (if more detailed info is required, please ask, I'll type it down)

Oh 22nd Oct, We had a strange server crash. The website was unavailable and the VPS appeared turned off by the VPS provider. They have restarted it at 14:03. In the messages log I can see the following lines:

Code: Select all

....
Oct 22 03:51:03 server1 systemd: Starting Cleanup of Temporary Directories...
Oct 22 03:51:03 server1 systemd: Started Cleanup of Temporary Directories.
Oct 22 14:04:08 server1 rsyslogd: [origin software="rsyslogd" swVersion="7.4.7" x-pid="1028" x-info="http://www.rsyslog.com"] start
Oct 22 14:03:51 server1 kernel: Initializing cgroup subsys cpuset
Oct 22 14:03:51 server1 kernel: Initializing cgroup subsys cpu
Oct 22 14:03:51 server1 kernel: Initializing cgroup subsys cpuacct
....

I assume every log line after 14h is related to the start initialised by the VPS provider. But what could go wrong? There are other lines where the server is cleaning up the Temporary Directories and there was no issue there. Also, how come we had a log line added at 14:04:08 and the follofing lines are at 14:03:51 ? Isn't this a paradox?

If you need any further information, please just ask, I'll get it for you if possible.

Thanks for the help in advance!

////////////////////////////////////////////////////////////////////////////////////////////////////

There is no information or log file available that can tell what happened. it is a kernel panic or somebody unplugged the hardware power cord.

[TrevorH]

Your server stopped just after 03:51:03 but there's no info there to say how or why.

The discrepancy in timestamps is because the system caches log messages until rsyslog starts and is able to write them out. So you system started to log at 14:03:51 but rsyslog didn't start until 14:04:08 at which point it writes all of the messages out with their correct timestamps.

[bertalanimre]

Thanks Trevor,

Where would you try to look for possible error messages? I've checked messages, security, nginx and mysql. No issues found. Is it just me or it looks like when you simply pull the power cord out of a server?

[TrevorH]

That or a kernel panic/crash.

[bertalanimre]

Thank you. Marking as SOLVED