In combination of using iscsitarget and VMWare ESXi4, all my VM's freeze up. When I tried rebooting the complete VMWare server, It didn't recognize my storage anymore. After a restart of ietd, the storage was found again. I got these messages in /var/log/messages: Oct 12 18:51:22 rachel kernel: [442923.304444] iscsi_trgt: cmnd_abort(1167) b18d0200 1 0 42 512 0 0 Oct 12 18:51:22 rachel kernel: [442923.304450] iscsi_trgt: Abort Task (01) issued on tid:1 lun:0 by sid:845524445757952 (Unknown Task) Oct 12 18:51:30 rachel kernel: [442931.502678] iscsi_trgt: cmnd_abort(1167) c28d0200 1 0 42 4096 0 0 Oct 12 18:51:30 rachel kernel: [442931.502683] iscsi_trgt: Abort Task (01) issued on tid:1 lun:0 by sid:1126999422468608 (Unknown Task) Oct 12 18:51:30 rachel kernel: [442931.502975] iscsi_trgt: cmnd_abort(1167) c48d0200 1 0 42 512 0 0 Oct 12 18:51:30 rachel kernel: [442931.502979] iscsi_trgt: Abort Task (01) issued on tid:1 lun:0 by sid:1126999422468608 (Unknown Task) Some googling has led me to believe this is the issue that is fixed in 1.4.20 of iscsitarget (Fixed compliance issues around ABORT TASK/ABORT TASK SET). An update to the newest version is probably the best resolution.
Indeed, adjusting severity as a workaround exists; restarting iscsi might not be possible in every setup though that's why I leave it at "major". base-system, please adjust as needed.
This should be done with 1.4.20.2. Thanks for your patience and support. It is currently hard masked as I do not have the hardware. Please test and report back.