On Sat, Jun 1, 2013 at 10:37 PM, Doug Clow <doug.clow@xxxxxxxxxxx> wrote: > > I tried the new patch today and I'm running into an error. From Windows 2012 side I connect to the target. But it seems like it is having trouble connecting as the interface slows and sometimes it shows multiple copies of the same disk in Disk Management. From the Centos side I am getting this message repeated: > > Jun 1 12:06:44 iscsi1 tgtd: iscsi_scsi_cmd_tx_done(1896) target bug 0 > Jun 1 12:06:44 iscsi1 tgtd: conn_close(103) connection closed, 0x359bb18 2 > Jun 1 12:06:44 iscsi1 tgtd: conn_close(109) sesson 0x359fe10 1 > Jun 1 12:06:44 iscsi1 tgtd: conn_close(92) already closed 0x359bb18 1 > It looks like it works in the Windows' versions prior to Win2012. It also occurred to me that the refusal of Win iscsi initiator to recognize tgt LUNs because of the lack of VPD compliance was also a new "feature" of Win2012, as Win7 seemed to be happy with the VPD data we had supplied. The above error log : "iscsi_scsi_cmd_tx_done(1896) target bug 0" means that in the saved pdu of the command just returned on completion there is unexpected iscsi opcode, 0 which is NOP-OUT. This could be a pre-existing bug that has surfaced in Win2012 after we fixed the READ(10) stuff, or a bug in a specific timing for connection shutdown (while the connection gets closed because of other misbehavior). Anyway, it needs to be checked separately. I'm trying to get access to a Win2012 machine and conduct these tests. Tomo, it seems that the patches may be applied now, as they clearly fix the problems in some (or even most of) the setups. It would be nice to get more positive feedbacks of course. Teruaki, Youn-Sang, can you please verify that the patches allow Win7/8 to proceed normally? Thanx, Alexander -- To unsubscribe from this list: send the line "unsubscribe stgt" in the body of a message to majordomo@xxxxxxxxxxxxxxx More majordomo info at http://vger.kernel.org/majordomo-info.html
