I am having a performance issue when trying to connect to two iSCSI
targets.  When I have only 1 target connected and mounted, performance
is not an issue and the mounted LUN is useable.  When I mount the
second (and third) LUN from a second iSCSI device, performance gets
horrible.  Specifically, there seems to be an issue going between each
device.  When actively using one mounted LUN, performance is fine, but
there is a noticeable delay when trying to interact with the second
LUN.  After the delay, interaction is fine with the second LUN, but
going back to the first experiences the same delay.  Copying from one
LUN to the other is out of the question, because each time data is
copied, it constantly experiences the same delay.

I'm not sure if this is in any way related to the previous topics on I/
O stall, found 
http://groups.google.com/group/open-iscsi/browse_thread/thread/99fe4f746bd8cd13#
and 
http://groups.google.com/group/open-iscsi/browse_thread/thread/c13c9e141e7cf900#,
and I can't move to the newest build because I get the same error
mentioned
http://groups.google.com/group/open-iscsi/browse_thread/thread/2450096aced00852#

I have SLES10 SP2 x64, and have used YaST's built in open-iscsi
configuration utility.

afsbackup:~ # uname -a
Linux afsbackup 2.6.16.60-0.42.5-smp #1 SMP Mon Aug 24 09:41:41 UTC
2009 x86_64 x86_64 x86_64 GNU/Linux

afsbackup:~ # iscsiadm -V
iscsiadm version 2.0-868

I've tried:
 - with CHAP and without
 - moving the second iSCSI device on a second network, connected to a
second private interface.
 - Firewall on and off
 - Manually discovering and logging in via command prompt (no gui)

Configuration details:
Server: 192.168.0.10
iscsi1: 192.168.0.100
iscsi2: 192.168.0.110

Upon connecting the second device:
------------------------------
afsbackup:/backup # iscsiadm -m discovery -t sendtargets -p
192.168.0.110

192.168.0.110:3260,1 iqn.2002-10.com.infortrend:raid.sn7391357.701


afsbackup:~ # iscsiadm -m node -T iqn.
2002-10.com.infortrend:raid.sn7391357.701 -p 192.168.0.110 -l

Logging in to [iface: default, target: iqn.
2002-10.com.infortrend:raid.sn7391357.701, portal: 192.168.0.110,3260]

Login to [iface: default, target: iqn.
2002-10.com.infortrend:raid.sn7391357.701, portal:
192.168.0.110,3260]: successful
------------------------------

I get the following in my /var/log/messages:
---------------------------------------------------------------------
Sep 17 13:05:07 afsbackup kernel: ping timeout of 5 secs expired, last
rx 4316649957, last ping 4316649957, now 4316652457
Sep 17 13:05:07 afsbackup kernel:  connection4:0: iscsi: detected conn
error (1011)
Sep 17 13:05:07 afsbackup iscsid: semop up failed 22
Sep 17 13:05:26 afsbackup iscsid: semop up failed 22
Sep 17 13:06:46 afsbackup kernel: scsi10 : iSCSI Initiator over TCP/IP
Sep 17 13:06:46 afsbackup iscsid: semop up failed 22
Sep 17 13:06:46 afsbackup kernel:   Vendor: IFT       Model: A16E-
G2130-4      Rev: 364Q
Sep 17 13:06:46 afsbackup kernel:   Type:   Direct-
Access                      ANSI SCSI revision: 05
Sep 17 13:06:46 afsbackup kernel: sdc : very big device. try to use
READ CAPACITY(16).
Sep 17 13:06:46 afsbackup kernel: SCSI device sdc: 15623929856 512-
byte hdwr sectors (7999452 MB)
Sep 17 13:06:46 afsbackup kernel: sdc: Write Protect is off
Sep 17 13:06:46 afsbackup kernel: sdc: Mode Sense: 8f 00 00 08
Sep 17 13:06:46 afsbackup kernel: SCSI device sdc: drive cache: write
back
Sep 17 13:06:46 afsbackup kernel: sdc : very big device. try to use
READ CAPACITY(16).
Sep 17 13:06:46 afsbackup kernel: SCSI device sdc: 15623929856 512-
byte hdwr sectors (7999452 MB)
Sep 17 13:06:46 afsbackup kernel: sdc: Write Protect is off
Sep 17 13:06:46 afsbackup kernel: sdc: Mode Sense: 8f 00 00 08
Sep 17 13:06:46 afsbackup kernel: SCSI device sdc: drive cache: write
back
Sep 17 13:06:46 afsbackup kernel:  sdc: sdc1
Sep 17 13:06:46 afsbackup kernel: sd 10:0:0:0: Attached scsi disk sdc
Sep 17 13:06:46 afsbackup kernel: sd 10:0:0:0: Attached scsi generic
sg5 type 0
Sep 17 13:06:46 afsbackup kernel:   Vendor: IFT       Model: A16E-
G2130-4      Rev: 364Q
Sep 17 13:06:46 afsbackup kernel:   Type:   Direct-
Access                      ANSI SCSI revision: 04
Sep 17 13:06:46 afsbackup kernel: SCSI device sdd: 3904954368 512-byte
hdwr sectors (1999337 MB)
Sep 17 13:06:46 afsbackup kernel: sdd: Write Protect is off
Sep 17 13:06:46 afsbackup kernel: sdd: Mode Sense: 8f 00 00 08
Sep 17 13:06:46 afsbackup kernel: SCSI device sdd: drive cache: write
back
Sep 17 13:06:46 afsbackup kernel: SCSI device sdd: 3904954368 512-byte
hdwr sectors (1999337 MB)
Sep 17 13:06:46 afsbackup kernel: sdd: Write Protect is off
Sep 17 13:06:46 afsbackup kernel: sdd: Mode Sense: 8f 00 00 08
Sep 17 13:06:46 afsbackup kernel: SCSI device sdd: drive cache: write
back
Sep 17 13:06:46 afsbackup kernel:  sdd: sdd1
Sep 17 13:06:46 afsbackup kernel: sd 10:0:0:1: Attached scsi disk sdd
Sep 17 13:06:46 afsbackup kernel: sd 10:0:0:1: Attached scsi generic
sg6 type 0

----------------------------------------------

with the following repeating pretty frequently:
----------------------------------------------
Sep 17 13:09:56 afsbackup kernel: ping timeout of 5 secs expired, last
rx 4316722054, last ping 4316721118, now 4316724554
Sep 17 13:09:56 afsbackup kernel:  connection6:0: iscsi: detected conn
error (1011)
Sep 17 13:09:56 afsbackup iscsid: semop up failed 22
----------------------------------------------

I've read that the "semop failed" can be resolved if i were to enable
CHAP authentication and include the log-in information in the /etc/
iscsi/iscsi.conf file, but I'm trying to take CHAP out of the picture.

Any suggestions?  Would the newer version of open-iscsi solve the
issue?  If so, any idea when SLES10 will have an RPM, or the version
will be compilable?

Thanks for any help you can give me.
-Jeff

--~--~---------~--~----~------------~-------~--~----~
You received this message because you are subscribed to the Google Groups 
"open-iscsi" group.
To post to this group, send email to open-iscsi@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
-~----------~----~----~----~------~----~------~--~---

Reply via email to