CAM status: SCSI Status Error

[jcpro]

I have a pfSense VM running on ESXi 6.0 and just started seeing these errors logged in dmesg:

Apr 21 02:57:53 	kernel 		(da0:mpt0:0:0:0): Retrying command
Apr 21 02:57:53 	kernel 		(da0:mpt0:0:0:0): SCSI status: Busy
Apr 21 02:57:53 	kernel 		(da0:mpt0:0:0:0): CAM status: SCSI Status Error
Apr 21 02:57:53 	kernel 		(da0:mpt0:0:0:0): READ(6). CDB: 08 02 08 8f 40 00 

The VM has an LSI Logic Parallel controller assigned to it.

Googling this there's a few people on FreeBSD 10.1-RELEASE that experienced it but commented that since upgrading to 10.2 they no longer experience the issue.  With pfSense on 10.3 I'm confused as to how to troubleshoot.

Has anybody seen this issue before?

[cmb]

Similar if not the exact same errors (don't recall for sure offhand) will come up when your ESX host is using shared storage and has issues or delays reaching it. That's the first thing I'd check, make sure the ESX host's access to storage isn't having issues.

[jcpro]

@cmb:

Similar if not the exact same errors (don't recall for sure offhand) will come up when your ESX host is using shared storage and has issues or delays reaching it. That's the first thing I'd check, make sure the ESX host's access to storage isn't having issues.

Wow that's exactly it.  My hosts lost their connections to the storage!  Fun times!
Nothing to see here, move along. :)

[gonzague]

hey … Same problem for me but on a machine that has no remote / shared storage, only a local RAID

For the moment I changed my LSI controller type from LSI Logic Parallel to SAS .. :)

It's quite strange though .. all my other VMs on the same host (not based on FreeBSD) dont report a problem with storage :s

Also, I'm keeping another PfSense running on this host with the LSI Logic Parallel controller if it can help diagnose something