Weird ESX Boot Issue

[jamessa]

I wanted to see if anyone has had an issue like this before. I went to boot an ESX server the other night that had a Veeam NAS datastore on it from an instant recovery awhile back. Anyway I was not able to fully boot my ESX 4.1 v381591 box and I got an error about esxcfg-nas. Even after multiple reboots. Since we don't have any NAS datastores I knew this must have been the Veeam NAS datastore. I ran esxcfg-nas from Putty and deleted the Veeam datastore and the server was then able to boot. I am not able to reproduce this in my test environment.

Ideas?

[cdickerson]

Yeah I have experienced this same problem. VMware tries to restore the NFS mount locations. If the Veeam server is down, it can't contact the NFS server and thus stalls. I just always dismount the NFS mount point after doing any surebackup jobs.

[Gostev]

I have tons of inactive NFS mount points in my lab and they never prevent ESX from booting up. During the boot process, host does try to contact the mounted NFS shares (what Craig references as "stalls"), but after timeout the boot process resumes normally.

It would truly be a critical bug on VMware side if inavailability of some shared storage prevented host from booting up... hosts usually have multiple datastores/LUNs mounted, and it is perfectly normal for some of them not to be available. I am sure they test this scenario well.

[jamessa]

That is interesting that someone else has had the same issue. I could not find any documentation of this happening, and I was not able to reproduce it again myself. While we do not use NFS datastores we do use iSCSI datastores and ESX boots just fine when those are down.