iSCSI target + MS initiator

Hi all,

I'm having problems using the MS initiator with iscsi enterprise. The target
machine is running iscsi enterprise 0.4.14. I partitioned a 250 gig SATA
into 4 partitions like so:

Disk /dev/sda: 250.0 GB, 250059350016 bytes
255 heads, 63 sectors/track, 30401 cylinders
Units = cylinders of 16065 * 512 = 8225280 bytes
   Device Boot      Start         End      Blocks   Id  System
/dev/sda1               1         600     4819468+  83  Linux
/dev/sda2             601        1200     4819500   83  Linux
/dev/sda3            1201        1800     4819500   83  Linux
/dev/sda4            1801        2400     4819500   83  Linux


and then used evms to build a RAID5 array out of sda1,sda2, and sda3. After
that I put LVM2 on top of md0 called Storage. Through evms I also created 2
LVM regions called testregion1 and testregion2. So at this point I have
lvm2/Storage/testregion1 and lvm2/Storage/testregion2. In the end I have
/dev/evms/lvm2/Storage/testregion1 /dev/evms/lvm2/Storage/testregion2.

I then edited the ietd.conf to look like this:

Target iqn.2007-01.com.hoober:dev.evms.lvm2.storage.testregion1
        Lun 0 Path=/dev/evms/lvm2/Storage/testregion1

Target iqn.2007-01.com.hoober:dev.evms.lvm2.storage.testregion2
       Lun 1 Path=/dev/evms/lvm2/Storage/testregion2

I (re)started the iscsi daemon which was successful. Now I installed the
latest iSCSI initiator from MS's website and connected to the target machine
described above. I am able to see both testregion1 and testregion2 from the
initiator software. I connected to testregion1 and formatted the disk as
NTFS. This all went smoothly, I even transferred a 3 gigabyte file just to
make sure that everything was working correctly.

Next I wanted to test the 'growing' ability of my new disk so I fired up
evms again and expanded the testregion1 to be just less than double it's
original size. I applied the changes and went back to the initiator and used
the disk management expand feature in windows XP. Once again this was fine.
Just to make sure that things were working ok I then rebooted the XP
initiator machine - this is where things went awry. Upon reboot and
reconnecting to testregion1 LUN I found that I could no longer see the
device from the disk management console of XP even though the initiator
software said that it was connected. I tested this on a production Windows
2003 machine as well, it connected to the LUN initially and won't connect
again.

Here is some dmesg output if it helps:

[42949432.350000] iSCSI Enterprise Target Software - version 0.4.14
[42949432.350000] iotype_init(90) register fileio
[42949432.350000] iotype_init(90) register nullio
[42949589.170000] sess_param(173) 1 1 1 8192 8192 262144 65536 2 20 1 1 1 0
1 1
[42949632.190000] execute_task_management(1205) 29 5 ffffffff
[42949692.190000] execute_task_management(1205) 2b 5 ffffffff
[42949767.200000] execute_task_management(1205) 2d 5 ffffffff
[42949857.200000] execute_task_management(1205) 2f 5 ffffffff
[42949962.210000] sess_param(173) 1 1 1 8192 8192 262144 65536 2 20 1 1 1 0
1 1
[42950047.220000] execute_task_management(1205) 2e 5 ffffffff
[42950091.220000] check_cmd_sn(534) sequence error (2d,2f)
[42950091.220000] cmnd_rx_start(1629) 2 30 -4
[42950091.220000] cmnd_skip_pdu(451) 30 2 0 0
[42950110.920000] check_cmd_sn(534) sequence error (2d,2f)
[42950110.930000] cmnd_rx_start(1629) 1 31 -4
[42950110.930000] cmnd_skip_pdu(451) 31 1 28 0
[42950225.230000] check_cmd_sn(534) sequence error (2e,2f)
[42950225.230000] cmnd_rx_start(1629) 2 32 -4
[42950225.230000] cmnd_skip_pdu(451) 32 2 0 0
[42950245.430000] check_cmd_sn(534) sequence error (2e,2f)
[42950245.430000] cmnd_rx_start(1629) 1 33 -4
[42950245.440000] cmnd_skip_pdu(451) 33 1 28 0
[42950375.240000] check_cmd_sn(534) sequence error (2d,2f)
[42950375.240000] cmnd_rx_start(1629) 2 34 -4
[42950375.240000] cmnd_skip_pdu(451) 34 2 0 0
[42950395.440000] check_cmd_sn(534) sequence error (2d,2f)
[42950395.440000] cmnd_rx_start(1629) 1 35 -4
[42950395.450000] cmnd_skip_pdu(451) 35 1 28 0
[42950472.550000] cdrom: open failed.
[42950540.260000] sess_param(173) 1 1 1 8192 8192 262144 65536 2 20 1 1 1 0
1 1
[42950685.260000] execute_task_management(1205) 2f 5 ffffffff
[42950714.980000] check_cmd_sn(534) sequence error (2e,30)
[42950714.980000] cmnd_rx_start(1629) 2 31 -4
[42950714.980000] cmnd_skip_pdu(451) 31 2 0 0
[42950734.970000] check_cmd_sn(534) sequence error (2e,30)
[42950734.970000] cmnd_rx_start(1629) 1 32 -4
[42950734.970000] cmnd_skip_pdu(451) 32 1 12 0
[42950743.980000] check_cmd_sn(534) sequence error (2f,30)
[42950743.980000] cmnd_rx_start(1629) 2 33 -4
[42950743.980000] cmnd_skip_pdu(451) 33 2 0 0

I have tried restarting both initiators and the target machines with no luck
- and only one initiator was ever connected to the LUN at a time. Is evms
screwing me up? I'm 95% sure that I'm doing something wrong but I can't
figure it out...any ideas?
Regards,

-- 
Scott Kaelin
Sitrof Technologies

-------------------------------------------------------------------------
Take Surveys. Earn Cash. Influence the Future of IT
Join SourceForge.net's Techsay panel and you'll get the chance to share your
opinions on IT & business topics through brief surveys - and earn cash
http://www.techsay.com/default.php?page=join.php&p=sourceforge&CID=DEVDEV

_______________________________________________
Evms-devel mailing list
[email protected]
To subscribe/unsubscribe, please visit:
https://lists.sourceforge.net/lists/listinfo/evms-devel