backlog limit exceeded

Hedloff

• February 01, 2017 13:39

In reference to thread: Backup crashed server / audit: backlog limit exceeded I just had the same issue on our dedicated servers. That solution helped me with those nasty freezes! Is there actually any need for audit? Are there anyone using it on hosting servers? This is default "on" on new RHEL systems today.

• 

  garconcn

  • February 01, 2017 21:35

  I was experiencing the same issue two days ago, I thought it was caused by R1soft because the server no responding just a few mins after R1soft replication started. I saw the backlog limit exceeded message in the server console. The server is cloudlinux 7.

  0
• 

  Hedloff

  • February 02, 2017 07:23

  Yes, especially on Cloudlinux 7.x and when R1Soft was running the server crashed all the time. We also changed the audit/backlog limit to 16.000 or something, but didn't help much.

  0
• 

  garconcn

  • February 02, 2017 17:54

  Do you see log like this on your server? /var/log/messages Jan 29 22:07:20 xxxx kernel: blk_update_request: I/O error, dev fd0, sector 0 Jan 29 22:07:20 xxxx kernel: blk_update_request: I/O error, dev fd0, sector 0 If yes, it's probability related to floppy device. When R1soft runs, it generates the above error on server. I only see this error on cloudlinux 7. I disabled it by this command: # rmmod floppy check floppy mod is loaded or not # lsmod|grep floppy floppy 69417 0 If you reboot the server, the floppy module will be back. I've tried to disable it by this but it didn't work. /etc/dracut.conf.d/nofloppy.conf omit_drivers+="floppy" and /etc/modprobe.d/nofloppy.conf blacklist floppy

  0
• 

  cPanelMichael

  • February 03, 2017 17:52

  Hello, You may also find the following third-party URL helpful: A Brief Introduction to auditd " Stack Exchange Security Blog Thank you.

  0

Please sign in to leave a comment.