HX bug CSCvh68059 ramdisk /var full

[gepperson]

This bug does not list HX version 2.6.1(d), but we are experiencing this issue in our environment. When this occurs the guest VMs do not function properly, become unresponsive, and cannot be rebooted, reset, or vmotioned off affected host. I would like some more background on why this ramdisk full condition occurs. We have two other clusters on 2.6.1(d) and are not experiencing this issue, but a third cluster experiences it often. Thank you. 

https://quickview.cloudapps.cisco.com/quickview/bug/CSCvh68059

[SMLunchen]

Here is a workaround I scripted:

SSH into esxi host.

Add the following lines to

 /etc/rc.local.d/local.sh

(before exit 0)

# Get the cron service pid and simply kill it
/bin/kill $(cat /var/run/crond.pid)

 
# The next line writes a cleanup cron to the crontab
/bin/echo "0 6 * * * /bin/rm /var/log/snapshots/*.log" >> /var/spool/cron/crontabs/root

 
# Finally start the cron-service again
/usr/lib/vmware/busybox/bin/busybox crond

This will make a reboot-surviving entry in the ESXi config to cleanup the logs at 6 am each day.

For applying the patch without reboot run additional the three lines from the script:

/bin/kill $(cat /var/run/crond.pid)
/bin/echo "0 6 * * * /bin/rm /var/log/snapshots/*.log" >> /var/spool/cron/crontabs/root
/usr/lib/vmware/busybox/bin/busybox crond