Hello,
since we reconfigured one of our shared SAN-devices (Falconstor based) to Lun 100 we get the following error messages in the VMkernel-logs of all ESX 3.5hosts every five minutes: (Every host has two Qlogic2432 controller, each controller has one path to this Falconstor-device configured)
Apr 3 20:08:56 vm-esx-2 vmkernel: 9:08:20:02.457 cpu0:1043)WARNING: SCSI: 660: Could not get Redundant controller info using V5.3 for device vmhba0:8:0
Apr 3 20:08:56 vm-esx-2 vmkernel: 9:08:20:02.457 cpu0:1043)WARNING: SCSI: 911: Could not read sense data for vmhba0:8:0
Apr 3 20:08:56 vm-esx-2 vmkernel: 9:08:20:02.457 cpu0:1043)WARNING: SCSI: 660: Could not get Redundant controller info using V5.3 for device vmhba1:8:0
Apr 3 20:08:56 vm-esx-2 vmkernel: 9:08:20:02.457 cpu0:1043)WARNING: SCSI: 911: Could not read sense data for vmhba1:8:0
Apr 3 20:08:56 vm-esx-2 vmkernel: 9:08:20:02.457 cpu0:1043)WARNING: SCSI: 898: LUN 100 is too large for the v53 FAStT device at vmhba0:8:100
Apr 3 20:08:56 vm-esx-2 vmkernel: 9:08:20:02.457 cpu0:1043)WARNING: SCSI: 898: LUN 100 is too large for the v53 FAStT device at vmhba1:8:100
Apr 3 20:08:56 vm-esx-2 vmkernel: 9:08:20:02.489 cpu0:1244)WARNING: SCSI: 4508: Delaying failover to path vmhba0:8:100
Otherwise this device seems to be fully functional, all configured pathes are shown correctly in the storage adapter configuration (VI Client).
Is this a real problem? Should we reconfigure to a lesser LUN? Or simply ignore the warnings?
Looks to me like the paths are "flip-flopping" (Technical term). Are you using mru or fixed pathing? Is the falconstor an active-active or active-passive configuration? Usually this happens when the wrong pathing type is used (default is mru).
DB
VMware Communities Moderator
I am using the default "mru"-configuration and the falconstor is an active-passive configuration.
But i understand now what is happening:
I did reconfigure the falconstor from Lun 0 to Lun 100 and then a rescan . The Falconstor showed up with the new LUN and all seemed weel to me.
But now i have on one adapter for scsi target 8: vmhba0:8:0 AND vmhba0:8:100 for the falconstor . Only LUN100 is real now, LUN0 is only a "ghost" from before.
Only after i restarted one of my ESX hosts this "ghost" LUN was gone.
Seems that a rescan is not sufficient to get rid of the"old" LUN here and i will have to restart all my hosts eventually (next patchday is not far probably).
Thanks boydd.
