Updated 2015-10-06
VMware released a patch today that will fix this problem, KB available here
—————————
Updated 2015-10-01
Today VMware leased a KB 2133118 for the below described issue. No resolution included but a workaround.
————————
Original post
———————–
Today i came across this VMware Communities thread that explains different scenarios where vSphere ESXi 5.5 Update 3 can crash virtual machines (VMs) when deleting a snapshot. I have not seen this myself so i trust the communities thread in this case and i thought it was worth a quick blog post.
The initial example states that the VMs were not placed in the datastore root folder, they were using subfolders but further down in the thread it seems that VMs located in the datastore root folder also crashes.
The thread also states that VMware support has acknowledge this as a bug and hopefully a fix and a KB will be released soon.
I’ll update this blog post when i have more information.
The following information has been seen in the ESXi host log files:
- vcpu-0| I120: SNAPSHOT: SnapshotDiskTreeFind: Detected node change from ‘scsi0:0’ to ”.
- vcpu-0| W110: Caught signal 11 — tid 35868 (addr 98)
- Unexpected signal: 11. And [msg.panic.haveLog]
- Caught signal 11 — tid 296089 (addr 6B13458)