RE: Getting iscsi offload on a M610 to work

2010-11-15 Thread Shyam_Iyer
(Ton I am copying the open-iscsi mailing list as some of the Broadcom iSCSI 
offload developers hang here as well and I believe I saw a fix for this issue )

Hi,

 I'm trying to get offload iscsi (rhel 5.5) to work on a PowerEdge M610
 and bnx2i.
 The thing is that I get it to work through the normal tcp/ip stack but
 not by specifying an iface which suppose to do offload iscsi.
 The discovery step works, but as I understood, this goes through the
 normal ip routing. The login step however just hangs.
 Here are some configuration details:

 # lspci:
 01:00.0 Ethernet controller: Broadcom Corporation NetXtreme II BCM5709S
 Gigabit Ethernet (rev 20)
 01:00.1 Ethernet controller: Broadcom Corporation NetXtreme II BCM5709S
 Gigabit Ethernet (rev 20)


 # ifconfig eth1
 eth1  Link encap:Ethernet  HWaddr 00:26:B9:FD:41:42
   inet addr:172.16.17.19  Bcast:172.16.17.255
 Mask:255.255.255.0
   inet6 addr: fe80::226:b9ff:fefd:4142/64 Scope:Link
   UP BROADCAST RUNNING MULTICAST  MTU:1500  Metric:1
   RX packets:604 errors:0 dropped:0 overruns:0 frame:0
   TX packets:61 errors:0 dropped:0 overruns:0 carrier:0
   collisions:0 txqueuelen:1000
   RX bytes:141966 (138.6 KiB)  TX bytes:8996 (8.7 KiB)
   Interrupt:24 Memory:dc00-dc012800

 # iscsiadm -m iface
 default tcp,empty,empty,empty,empty
 iser iser,empty,empty,empty,empty
 bnx2i.00:26:b9:fd:41:41
 bnx2i,00:26:b9:fd:41:41,172.16.17.21,empty,empty
 bnx2i.00:26:b9:fd:41:43
 bnx2i,00:26:b9:fd:41:43,172.16.17.20,empty,empty
 bnx2i.00:00:00:00:00:00 bnx2i,00:00:00:00:00:00,empty,empty,empty

 # iscsiadm -m iface -I bnx2i.00:26:b9:fd:41:43
 # BEGIN RECORD 2.0-871
 iface.iscsi_ifacename = bnx2i.00:26:b9:fd:41:43
 iface.net_ifacename = empty
 iface.ipaddress = 172.16.17.20
 iface.hwaddress = 00:26:b9:fd:41:43
 iface.transport_name = bnx2i
 iface.initiatorname = empty
 # END RECORD

 Then running:

 # iscsiadm -m discovery -t st -p 172.16.17.243 -I
 bnx2i.00:26:b9:fd:41:43
 172.16.17.243:3260,1 iqn.2002-10.com.infortrend:raid.sn7740362.101
 #

 and

 # iscsiadm -m node -T iqn.2002-10.com.infortrend:raid.sn7740362.101
 --login
 Logging in to [iface: bnx2i.00:26:b9:fd:41:43, target:
 iqn.2002-10.com.infortrend:raid.sn7740362.101, portal:
 172.16.17.243,3260]



 Do you have an offload key plugged into the motherboard. If not you
 might see offload not supported

 messages in /var/log/messages..

 Broadcom NetXtreme II CNIC Driver cnic v2.1.0 (Oct 10, 2009)
 cnic: Added CNIC device: eth0
 cnic: Added CNIC device: eth1
 Broadcom NetXtreme II iSCSI Driver bnx2i v2.1.0 (Dec 06, 2009)
 iscsi: registered transport (bnx2i)
 scsi1 : Broadcom Offload iSCSI Initiator
 scsi2 : Broadcom Offload iSCSI Initiator
 iscsi: registered transport (tcp)
 iscsi: registered transport (iser)
 iscsi: registered transport (be2iscsi)


 You will see the messages when the login is being attempted and not
 during driver load.


 Also see if you can restart iscsid. ( The difference is that iscsid
 reloads the bnx2i driver.)


 No luck. I'm afraid. A reboot hangs also because it tries to login to
 the target over this bnx2i interface
 Ton

1) Does the BIOS have Windows MBA mode Enabled.  This also can cause
 login to not occur.

 You can find this option in the CTRL+K option on the Broadcom Nic option
 ROM during post.

 don't seen anything about this at boot or in the bios.



Sorry. I should have been more explicit.

During POST you should see a CTRL+S option that lets you into the Broadcom 
Ethernet controller's firmware menu.

Inside this menu you will find a CTRL+K option. The Windows MBA mode in one 
of the settings inside this option. Disable that option

2) Do you see any segfault messages in the /var/log/messages for
 brcm_iscsiuio ?

 no

OK

 3) Using ps command can you determine if the brcm_iscsuio daemon is
 still running ?

 yes

Ok.



 4) If you could post your /var/log/messages and /var/log/brcm-iscsi.log
 to some paste-bin, equivalent that would be great. Or send me over
 email.

 with debugging on for brcm_iscsiuio gives this output:
 INFO  [Mon Nov 15 13:52:45 2010]Initialize logger using log
 file: /var/log/brcm-iscsi.log
 INFO  [Mon Nov 15 13:52:45 2010]Started BRCM iSCSI stack: Ver 0.5.15
 INFO  [Mon Nov 15 13:52:45 2010]Build date: Wed Sep 29 20:45:10 CEST
 2010
 INFO  [Mon Nov 15 13:52:45 2010]Debug mode enabled
 DBG   [Mon Nov 15 13:52:45 2010]Loaded nic library 'bnx2' Version:
 '0.5.15' build on Wed Sep 29 20:45:10 CEST 2010'
 DBG   [Mon Nov 15 13:52:45 2010]Added 'bnx2' nic library
 DBG   [Mon Nov 15 13:52:45 2010]Loaded nic library 'bnx2x' Version:
 '0.5.15' build on Wed Sep 29 20:45:10 CEST 2010'
 DBG   [Mon Nov 15 13:52:45 2010]Added 'bnx2x' nic library
 INFO  [Mon Nov 15 13:52:45 2010]uevent Listening for uevents
 INFO  [Mon Nov 15 13:52:45 2010]signal handling thread ready
 INFO  [Mon Nov 15 13:52:45 2010]nic_utils Found host[1]: host1
 INFO  [Mon Nov 15 13:52:45 2010]Done
 capturing 

RE: cannot logout

2010-11-15 Thread p1
 
Thanks for your reply Mike.
Since submitting that question, I've been able to bounce the server,
which cleared up the troublesome session and all is well now.

--
Paul

-Original Message-
From: Mike Christie [mailto:micha...@cs.wisc.edu] 
Sent: Saturday, November 13, 2010 2:30 PM
To: open-iscsi@googlegroups.com
Cc: p...@fhri.org
Subject: Re: cannot logout

On 11/10/2010 05:02 PM, p...@fhri.org wrote:
 I had a session logged in while structural changes were made to the
SAN.
 The session is no longer valid as target names and LUNs have changed.
 However, I cannot log the session out or even stop the iscsi service 
 on the server.  I can't bounce the server for several more days.  I 
 can stop the iscsid service which stops the barage of login errors I 
 get from the SAN, but I cannot end the session or remove the node.

 [db1: ~]# iscsiadm -m node -p 172.16.10.11 -o delete
 iscsiadm: This command will remove the record [iface: default, target:
 target0, portal: 172.16.10.11,3260], but a session is using it. Logout

 session then rerun command to remove record.
 iscsiadm: Could not execute operation on all records. Err 22.

Yeah, that should fail. It only deletes the record for the target
portal. It would not log it out.


 [db1: ~]# iscsiadm -m node --logoutall=all Logging out of session 
 [sid: 1, target: target0, portal:
 172.16.10.11,3260]
 iscsiadm: Could not logout of [sid: 1, target: target0, portal:
 172.16.10.11,3260]:
 iscsiadm: initiator reported error (9 - internal error)


This should work. Can you run

iscsiadm -m session -P 3

and send the output.

internal error normally means the session is just getting started and so
we cannot log it out yet.


 [db1: ~]# service iscsi stop
 Logging out of session [sid: 1, target: target0, portal:
 172.16.10.11,3260]
 iscsiadm: Could not logout of [sid: 1, target: target0, portal:
 172.16.10.11,3260]:
 iscsiadm: initiator reported error (9 - internal error)
 Stopping iSCSI daemon:  [  OK  ]

This runs the logout all command above so it fails for the same reason.


 [db1: ~]# iscsiadm -m session
 tcp: [1] 172.16.10.11:3260,4 target0

 CentOS 5.4  (2.6.18-164.11.1.el5)
 iscsi-initiator-utils.x86_646.2.0.871-0.16.el5

 Thanks for your help,
 Paul


 --
 ea926p




--
ea926p

-- 
You received this message because you are subscribed to the Google Groups 
open-iscsi group.
To post to this group, send email to open-is...@googlegroups.com.
To unsubscribe from this group, send email to 
open-iscsi+unsubscr...@googlegroups.com.
For more options, visit this group at 
http://groups.google.com/group/open-iscsi?hl=en.