Logs & Lists to Save before httpd restart when web server freezes?

Linux and Unix-Like Red Hat
1

Hello...

We keep having a problem with the httpd freezing and users getting time outs.

I am not the Sys Admin but since I am the 1st one in, I have to restart the httpd server.

Is there a list of items I should collect before I restart in order to help with trouble shooting?

I currently do a log watch from a few days before and a ps -ef right before I restart.

What are some of the other things I should do?

Thanks,
~Donavon

2

Your Sys Admin is fine with you doing this? He/she should be working to resolve, or at least be directing you to what to look for before simply power-down/power-up.
So many things to think about...
# users connected
memory space in-use/free
disk space free
last user connected; last program run
cron jobs

and so on.
As you look through, probably several will be quickly ruled out. However, a system should not freeze without a reason.

3

Unfortunately it's pretty siloed with little to no feedback. :frowning:

Just trying to find out a list of information to collect.

All I can do is:

  • Collect as much information as possible to help with the trouble shooting process
  • restart services so users can access httpd

Thank you for the list.. It is a good place to start.

  • # users connected
  • memory space in-use/free
    cat /proc/meminfo
    free -m
    vmstat
  • disk space free
    df -h
  • last user connected; last program run
  • cron jobs (running???)
  • Current Running Services
    ps -ef
  • summary of current logs
    logwatch --detail High --archives --range
4 
strace -p <pid-of-a-stuck-httpd>

might give some helpful information.