0 Replies Latest reply on Nov 20, 2013 4:57 PM by TheGushi

    VMware blew away my crontab file?

    TheGushi Lurker

      Hey there,


      We've been using GhettoVCB for months now and love it. I apologize if this is a somewhat meta question as it mainly applies to cron, but as many people who use GhettoVCB will be modding their crontab in a similar (and totally unsupported) way, it seems like the right group to ask.


      Last week, it seems ESXI (5.1) decided to blow away my crontab file without reason and revert it to the system standard.  I didn't check how long crond was running or the datestamp on the crontab while debugging this.  I wanted to post here for two reasons.


      1) to ask if anyone else has hit this issue, and possibly ask the GhettoVCB author to warn people it can happen.

      2) to ask what a better alternative to trying to use cron might be.  I could have a system SSH in with a key and fire off these backups, perhaps, but I'm new to this.


      Uptime on the ESXI system was 61 days (still is).  No reboot occurred.  It's a standalone host.  Had I rebooted, my local.sh commands (as per the readme) would have worked:


      /bin/kill $(cat /var/run/crond.pid)

      /bin/echo "0 23 * * * /vmfs/volumes/Performance/ghettoVCB-master/ghettoVCB.sh -a > /dev/null" >> /var/spool/cron/crontabs/root



      Here's my info, if it matters:


      /etc/rc.local.d # uname -a

      VMkernel hostnameredacted.example.com 5.1.0 #1 SMP Release build-1065491 Mar 23 2013 10:12:46 x86_64 GNU/Linux


      Backups are happening over NFS to a Sun Thumper (I don't think it matters), but that's what was asked for in the sticky topic.


      There's no "version" string in my GhettoVCB script -- maybe there should be -- but it was current when I installed it on July 22nd.


      Thanks for reading!