Windows VM locks up on ESXi 5.0 during tape backup

[zzumbaugh]

I've had an ESXi 5.0 host running for almost 2 months now with Windows SBS 2011 running as a VM. I have an Adaptec 29320ALP SCSI card connect to a Quantum LTO4 tape drive. This server has locked up 2 or 3 times during a backup, in wich case I have to reboot ESXi. If I try to power down the guest OS, it gets to about 95% and freezes. Then if I try to reboot ESXi, it freezes as well.

Today I was able to reboot ESXi from the command line and it came up fine. I have attached the vmkernel.log from today. I have no clue how to fix this or why it is occurring at seemingly random intervals.

[Dave_Mishchenko]

You might want to check if this applies - http://kb.vmware.com/kb/1026157.

[zzumbaugh]

Thanks for the input Dave. I had to run that command on day one since ESXi 5.0 wouldn't see my tape drive at all. It shows up and works fine for 2 weeks or so, then freezes my system.

2011-10-27T15:22:41.271Z cpu2:2642)ScsiScan: 1098: Path 'vmhba4:C0:T4:L0': Vendor: 'QUANTUM '  Model: 'ULTRIUM 4       '  Rev: 'W53F'
2011-10-27T15:22:41.271Z cpu2:2642)ScsiScan: 1101: Path 'vmhba4:C0:T4:L0': Type: 0x1, ANSI rev: 5, TPGS: 1 (implicit only)

The error I get repeating through the vmkernel.log is this:

<6>0 :: Attempting to queue a TARGET RESET message:CDB:
2011-11-10T14:41:56.141Z cpu9:2157)scsi6: Device reset code sleeping
2011-11-10T14:41:58.145Z cpu6:2158)VSCSI: 2747: Retry 782 on handle 8193 still in progress after 2 seconds
2011-11-10T14:42:01.144Z cpu9:2157)scsi6: Device reset timer expired (active 3)
2011-11-10T14:42:01.144Z cpu9:2157)Recovery SCB completes
2011-11-10T14:42:01.144Z cpu9:2157)reset scb aborted from QINFIFO
2011-11-10T14:42:01.144Z cpu9:2157)scsi6: Device reset returning 0x2003

[zzumbaugh]

This has appeared to be resolved by running a bootable update disk on the server. It is an IBM x3500.