iSCSI (istgt) error with GlobalSAN

[Rafael Henrique Faria]

I'm using the iSCSI Target (istgt-20100606) in FreeNAS 0.7.2, with Mac OS X
Snow Leopard 10.6.4, and GlobalSAN iSCSI Initiator (4.0.0.204).

I can transfer data to and from the volume (formated as HFS+) without
problems.

But, if I try to unzip a zipped file direct from the volume, I get a lot of
errors:

>From the istgt:

Aug 21 14:52:34 freenas istgt[8977]: Login from
iqn.1987.com.apple.snow-leopard (172.18.9.7) on
iqn.2007-09.org.freebsd.istgt:istor LU1 (172.18.9.2:3260,1),
ISID=800504000000, TSIH=2, CID=1029, HeaderDigest=on, DataDigest=on
Aug 21 14:52:48 freenas istgt[8977]:
istgt_iscsi.c:4084:istgt_iscsi_transfer_out: ***ERROR*** iscsi_read_pdu()
failed
Aug 21 14:52:48 freenas istgt[8977]: istgt_iscsi.c:5002:worker: ***ERROR***
iscsi_task_transfer_out() failed on
iqn.2007-09.org.freebsd.istgt:istor,t,0x0001(iqn.1987.com.apple.snow-leopard,i,0x800504000000)
Aug 21 14:52:48 freenas istgt[8977]:
istgt_lu_disk.c:4430:istgt_lu_disk_queue_start: ***WARNING*** transfer abort
CmdSN=1
Aug 21 14:52:48 freenas istgt[8977]: istgt_lu.c:2670:luworker: ***WARNING***
LU1: lu_disk_queue_start() aborted

After changing the Queue to 0 (disabling it), a different error appears:

Aug 21 16:11:18 freenas istgt[52845]: Login from
iqn.1987.com.apple.snow-leopard (172.18.9.7) on
iqn.2007-09.org.freebsd.istgt:istor LU1 (172.18.9.2:3260,1),
ISID=800a04000000, TSIH=1, CID=1034, HeaderDigest=on, DataDigest=on
Aug 21 16:15:16 freenas istgt[52845]:
istgt_iscsi.c:4084:istgt_iscsi_transfer_out: ***ERROR*** iscsi_read_pdu()
failed
Aug 21 16:15:16 freenas istgt[52845]:
istgt_lu_disk.c:1751:istgt_lu_disk_transfer_data: ***ERROR***
iscsi_transfer_out()
Aug 21 16:15:16 freenas istgt[52845]:
istgt_lu_disk.c:3589:istgt_lu_disk_lbwrite: ***ERROR***
lu_disk_transfer_data() failed
Aug 21 16:15:16 freenas istgt[52845]:
istgt_lu_disk.c:5275:istgt_lu_disk_execute: ***ERROR*** lu_disk_lbwrite()
failed

And logs from the GlobalSAN iSCSI Initiator:

21/08/10 16:13:42 kernel GLO Warning: Error 32 while receiving BHS.
21/08/10 16:13:42 kernel GLO Warning: Receiving thread has stopped with
error 32.
21/08/10 16:17:46 kernel GLO Warning: Timeout detected for connection
0x9af2d00 in state 2
21/08/10 16:17:46 kernel GLO Warning: Error 32 while receiving BHS.
21/08/10 16:17:53 kernel GLO Warning: Timeout occured while waiting Target
became ready.
...
21/08/10 16:21:05 kernel EMCDiskArrayInitiator4::probe fails
21/08/10 16:21:05 kernel SCSIPressurePathManager: new active path available,
checking, loops = 1
...
21/08/10 16:30:03 kernel GLO Warning: Receiving thread has stopped with
error 32.
21/08/10 16:30:13 kernel disk1s2: I/O error.

I can copy the same zipped file out of the volume, unzip it, and send the
file back to the volume, without errors... the error occurs only when I try
to unzip the file direct from the volume.

I have a lot of logs from the event, but it's all the same...


-- 
Rafael Henrique da Silva Faria