Re: [Scst-devel] qla2x00tgt : NPIV initiators get wrong LUNs

SourceForge Headquarters 225 Broadway Suite 1600 San Diego, CA 92101 +1 (858) 454-5900

I'll recreate locally.  I'll broadcast once I have something.

Regards,
Quinn Tran

From: Dave Butler <tea...@gm...<mailto:tea...@gm...>>
Date: Sunday, September 7, 2014 1:19 AM
To: "scs...@li...<mailto:scs...@li...>" <scs...@li...<mailto:scs...@li...>>
Subject: Re: [Scst-devel] qla2x00tgt : NPIV initiators get wrong LUNs

I found that after adding the 2nd vport (reverse order), then I did a rescan of the block device on initiator side, the disks became correct (size and data). but won't help if didn't add the 2nd vport.

but the disk inquiry info still showing incorrect disk :
[10:0:0:0]   disk    SCST_BIO sde               310  /dev/sdb
[11:0:0:0]   disk    SCST_BIO sde               310  /dev/sdc

On Sun, Sep 7, 2014 at 3:22 AM, Dave Butler <tea...@gm...<mailto:tea...@gm...>> wrote:
Hi Quinn,

Attached please find the log.

added vport :
50:00:00:00:00:00:00:00 (assigned 8GB disk)
50:00:00:00:00:00:00:11 (assigned 1GB disk)

Sep  7 02:52:08 : client reboot

Sep  7 02:58:31 : add vport 50:00:00:00:00:00:00:11 (reverse order)
got 8GB disk (incorrect disk)

Sep  7 02:59:28 : add vport 50:00:00:00:00:00:00:00
also got 8GB disk (correct disk)

There is another problem. When the first time I add vport, I can't get the disk until I call scstadmin -issue_lip.

Following is the log when adding vport for the first time (rebooted both client and target). and on initiator side, it keep sending reset and abort commands.

Sep  7 02:51:16 scstfc kernel: [689]: q2t_exec_sess_work:6771:Sending work cmd ffff88020e39db00 to SCST
Sep  7 02:51:16 scstfc kernel: [6140]: scst: scst_init_session:7053:Using security group "G1" for initiator "50:00:00:00:00:00:00:00" (target 21:00:00:24:ff:2e:17:cc)
Sep  7 02:51:16 scstfc kernel: [6140]: scst_alloc_add_tgt_dev:4643:Device sde on SCST lun=0
Sep  7 02:51:16 scstfc kernel: [6140]: scst: scst_alloc_set_UA:8463:Queuing new UA ffff8802132d1068 (6:29:0, d_sense 0) to tgt_dev ffff8802146d9ac0 (dev sde, initiator 50:00:00:00:00:00:00:00)
Sep  7 02:51:16 scstfc kernel: [6221]: scst: scst_cmd_thread:4944:Processing thread sde0_0 (PID 6221) started
Sep  7 02:51:16 scstfc kernel: [0]: q2t_sched_sess_work:1782:Scheduling work (type 0, prm ffff8802153d35c0) to find session for param ffff88021ed83d48 (size 8, tgt ffff88021347faa8)
Sep  7 02:51:16 scstfc kernel: [689]: q2t_sess_work_fn:6942:Sess work (tgt ffff88021347faa8)
Sep  7 02:51:16 scstfc kernel: [689]: q2t_exec_sess_work:6562:prm ffff8802153d35c0
Sep  7 02:51:17 scstfc kernel: [689]: q2t_exec_sess_work:6673:sess ffff8802133ce310 found
Sep  7 02:51:17 scstfc kernel: [689]: q2t_exec_sess_work:6692:Command or task mgmt received without FC-4 session sess ffff8802133ce310 loop_id 3 state ff deleted 0 logout_acc_pending 0. Terminate it.
Sep  7 02:51:17 scstfc kernel:
Sep  7 02:51:27 scstfc kernel: qla2xxx [0000:02:00.0]-1820:6: **** Failed mbx[0]=4005, mb[1]=8, mb[2]=0, mb[3]=0, cmd=72 ****.
Sep  7 02:51:27 scstfc kernel: qla2xxx [0000:03:00.0]-287d:8: FCPort state transitioned from ONLINE to LOST - portid=010700.
Sep  7 02:51:27 scstfc kernel: [6223]: q2t_fc_port_deleted:1635:sess ffff8802133ce670 id=10700 wwn=2100000e1e09994a
Sep  7 02:51:27 scstfc kernel: [6223]: q2t_schedule_sess_for_deletion:1011:Scheduling sess ffff8802133ce670 for deletion (schedule 1)
Sep  7 02:51:27 scstfc kernel: [6223]: q2t_schedule_sess_for_deletion:1025:qla2x00t(8): session for port 21:00:00:0e:1e:09:99:4a (loop ID 0) scheduled for deletion in 9 secs
Sep  7 02:51:27 scstfc kernel: qla2xxx [0000:03:00.0]-287d:8: FCPort state transitioned from ONLINE to LOST - portid=010600.

On Sat, Sep 6, 2014 at 2:06 AM, Quinn Tran <qui...@ql...<mailto:qui...@ql...>> wrote:
Jens, Dave,

Let me do some digging and get back with you.

Is this problem still true, with code base from
"git://git.qlogic.com/scst-qla2xxx.git<http://git.qlogic.com/scst-qla2xxx.git>"?  Qlogic have not up-keep/maintain
the "qla2x00t" version out no
"http://sourceforge.net/p/scst/svn/HEAD/tree/" in the past.  QLogic just
started introducing the "git://git.qlogic.com/scst-qla2xxx.git<http://git.qlogic.com/scst-qla2xxx.git>" version.
Both code base diverge quite a bit.

IF the new code is still a problem, can you capture log with the following
options in addition with what you have.

# set global trace option for qla2xxx_scst.ko.  Just one of the host is
sufficient.
echo "disc" > /sys/class/scsi_host/<host#>/trace_level

# set trace option for qla2x00t.ko
Echo "mgmt_dbg" > /sys/kernel/scst_tgt/targets/qla2x00t/trace_level

Regards,
Quinn Tran

On 9/5/14 1:28 AM, "Jens-U. Mozdzen" <jmo...@nd...<mailto:jmo...@nd...>> wrote:

>Hi Vlad & Quinn,
>
>Zitat von Vladislav Bolkhovitin <vs...@vl...<mailto:vs...@vl...>>
>> Hi guys,
>>
>> I've just talked to Quinn about this issue and he promised to take a
>> look at it.
>
>will this cover the symptoms I reported on Sep 1
>(http://sourceforge.net/p/scst/mailman/message/32783243/), too? It
>does look like a different issue, but within the same code area...
>
>Regards,
>Jens
>
>
>--------------------------------------------------------------------------
>----
>Slashdot TV.
>Video for Nerds.  Stuff that matters.
>http://tv.slashdot.org/
>_______________________________________________
>Scst-devel mailing list
>https://lists.sourceforge.net/lists/listinfo/scst-devel

________________________________

This message and any attached documents contain information from QLogic Corporation or its wholly-owned subsidiaries that may be confidential. If you are not the intended recipient, you may not read, copy, distribute, or use this information. If you have received this transmission in error, please notify the sender immediately by reply e-mail and then delete this message.

--
Dave Butler

--
Dave Butler

________________________________

This message and any attached documents contain information from QLogic Corporation or its wholly-owned subsidiaries that may be confidential. If you are not the intended recipient, you may not read, copy, distribute, or use this information. If you have received this transmission in error, please notify the sender immediately by reply e-mail and then delete this message.