Skip to main content

Server keeps crashing in last 3 days - Swap issue?

Comments

9 comments

  • heusdens
    Hi an update, the server just shut down again, it appears to be happening around 3am and 1pm. Load averages shot up to : 95.77 67.52 31.48 Perhaps its the cron jobs? I look at the cpanel cron job configuration and crikey moses, it's set to run at 1 and 3.30 which is pretty much in line with when the server crashes. This only started happening when WHM was upgraded to WHM 11.40.0 (build 19). Any suggestions on how this can be fixed without disabling it?
    0
  • kernow
    Your server is crashing because its run out of memory ( OOM). Check WHM >>> daily process log and confirm what's eating up your RAM
    0
  • cPanelMichael
    Hello :) Please review the following thread the next time your server load spikes: Troubleshooting High Load on Linux Systems It should help you to diagnose the cause of the high load. Thank you.
    0
  • heusdens
    Well what I've done is adjust the cpbackup config to run once a week. Therefore my assumption is that tomorrow morning at 3.30am my server won't crash. If my assumption is correct and my normal scheduled backups don't crash my server then why would cpbackup be crashing the server? How would one diagnose that. What logs would one check in?
    0
  • cPanelMichael
    You could manually generate backups via: /usr/local/cpanel/bin/backup --force
    Then, monitor the load and resource usage as backups generate to determine if that's the cause of the problem. Thank you.
    0
  • sahostking
    I have the same issue since 11.40, Going to try and schedule backups at a different time as it crashes at the same time each day. Will provide feedback in a 2 days.
    0
  • heusdens
    I've been monitoring since my last post, what I did was disable backups, and will run them manually every other day. Since then everything has been running smooth. However this is the weekend so the true test to see whether it was the cron jobs is yet to be determined as traffic will begin to pick up again during the week.
    0
  • heusdens
    Hi cPanelMichael I have an update, I did a force backup via /usr/local/cpanel/bin/backup --force and monitored server load using top to see if it was indeed contributing. The interesting thing I noticed was that the pigz process was consuming 100 to even 200% of cpu shooting the server average load to >2. I assume this is not normal and probably the reason why the server crashed. Do you have any advice?
    0
  • cPanelMichael
    A server load higher than 2 is not necessarily considered a high load. Was this slowing the server to the point where it was noticeable or failing to respond? Was the disk I/O (sar -u command) higher than normal? Thank you.
    0

Please sign in to leave a comment.