[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
Google
  Web www.spinics.net

Re: [Iscsitarget-devel] Strange iSCSI / IET Behavior [Errors when Writing]



Thanks a lot for your input Ming.

In the meantime, here's my experimentation with the
core-iscsi initiator.  I was able to compile it as
well as the necessary kernel modules although I'm
using 2.6.9-22 smp while it recommends a minimum of
2.6.11 (probably due to all the backporting Red Hat
does in it's kernel).  The initiator was able to see
the target as shown from the log messages below:

When I run /etc/rc.d/init.d/initiator start, it starts
successfully and I'm able to partition the hard drive
and create a filesystem.

localhost kernel: iCHANNEL[0] - Incremented iSCSI
connection count to 1 to node:
iqn.2001-04.com.example:storage.disk2.sys1.xyz
localhost kernel: iCHANNEL[0] - Established iSCSI
session to node:
iqn.2001-04.com.example:storage.disk2.sys1.xyz
localhost kernel: iSCSI Core Stack[1] - Incremented
number of active iSCSI sessions to 1.
localhost kernel:   Vendor: IET       Model:
VIRTUAL-DISK      Rev: 0   
localhost kernel:   Type:   Direct-Access             
        ANSI SCSI revision: 04
localhost kernel: SCSI device sde: 703345408 512-byte
hdwr sectors (360113 MB)
localhost kernel: SCSI device sde: drive cache: write
back
localhost kernel:  sde: unknown partition table
localhost kernel: Attached scsi disk sde at scsi4,
channel 0, id 0, lun 0
localhost kernel: SCSI device sde: 703345408 512-byte
hdwr sectors (360113 MB)
localhost kernel: SCSI device sde: drive cache: write
back
localhost kernel:  sde: sde1
localhost kernel: SCSI device sde: 703345408 512-byte
hdwr sectors (360113 MB)
localhost kernel: SCSI device sde: drive cache: write
back
localhost kernel:  sde: sde1

I again did the same test as before: dd if=/dev/zero
of=/dev/sde bs=1024k count=5000

My /etc/sysconfig/initiator looks like this.

CHANNEL="0 1 lo 127.0.0.1 3260 0"

It would give me a lot of messages such as:

check_segment_length(1552) too lond data 15a 32768
8192
check_segment_length(1552) too lond data 15b 32768
8192
iscsi_session_push_cmnd(1531) too large cmd_sn (56,1)
... [and so forth]

before finally completing:

real 1m31.320s
user 0m0.001s
sys 0m0.139s

Sometimes it finishes while at other times, it gives
the errors above and then goes into a loop trying to
reconnect and then disconnecting again:

localhost kernel: iCHANNEL[0]_LU[0] - SCSI Task
Timeout with ITT: 0x00000264
localhost kernel: iCHANNEL[0]_LU[0] - Forcing Session
Reinstatement for timed-out SCSI Task to ONLINE
Logical Unit
localhost kernel: iCHANNEL[0] - Performing Cleanup for
failed iSCSI CID: 0 to
iqn.2001-04.com.example:storage.disk2.sys1.xyz
localhost kernel: iCHANNEL[0] - Decremented iSCSI
connection count to 0 to node:
iqn.2001-04.com.example:storage.disk2.sys1.xyz
localhost kernel: iCHANNEL[0] - released iSCSI session
to node:
iqn.2001-04.com.example:storage.disk2.sys1.xyz
localhost kernel: iSCSI Core Stack[1] - Decremented
number of active iSCSI sessions to 0
localhost kernel: iCHANNEL[0] - No defined iSCSI
Authentication Methods, skipping SecurityNegotiation
phase.
localhost kernel: iscsi_login_rx_data:54: ***ERROR***
rx_data returned -512, expecting 48.
localhost kernel:
iscsi_initiator_start_negotiation:754: ***ERROR***
iSCSI Login negotiation failed.
localhost kernel: iscsi_handle_authetication:108:
***ERROR*** sock_recvmsg() returned 0
localhost kernel: iCHANNEL[0] - No defined iSCSI
Authentication Methods, skipping SecurityNegotiation
phase.
localhost kernel: iscsi_login_rx_data:54: ***ERROR***
rx_data returned -512, expecting 48.
localhost kernel:
iscsi_initiator_start_negotiation:754: ***ERROR***
iSCSI Login negotiation failed.
localhost kernel: iscsi_handle_authetication:108:
***ERROR*** sock_recvmsg() returned 0
localhost kernel: iCHANNEL[0] - No defined iSCSI
Authentication Methods, skipping SecurityNegotiation
phase.
localhost kernel: iscsi_login_rx_data:54: ***ERROR***
rx_data returned -512, expecting 48.
localhost kernel:
iscsi_initiator_start_negotiation:754: ***ERROR***
iSCSI Login negotiation failed.
localhost kernel: iscsi_handle_authetication:108:
***ERROR*** sock_recvmsg() returned 0
localhost kernel: iCHANNEL[0] - No defined iSCSI
Authentication Methods, skipping SecurityNegotiation
phase.
localhost kernel: iscsi_login_rx_data:54: ***ERROR***
rx_data returned -512, expecting 48.
localhost kernel:
iscsi_initiator_start_negotiation:754: ***ERROR***
iSCSI Login negotiation failed.
localhost kernel: iscsi_handle_authetication:108:
***ERROR*** sock_recvmsg() returned 0
localhost kernel: iCHANNEL[0] - No defined iSCSI
Authentication Methods, skipping SecurityNegotiation
phase.
localhost kernel: iscsi_login_rx_data:54: ***ERROR***
rx_data returned -512, expecting 48.
localhost kernel:
iscsi_initiator_start_negotiation:754: ***ERROR***
iSCSI Login negotiation failed.
localhost kernel: iscsi_handle_authetication:108:
***ERROR*** sock_recvmsg() returned 0

PS. I'll check out open-iscsi tomorrow morning and
tell you how it goes.

--- Ming Zhang <mingz@xxxxxxxxxxx> wrote:

> forgot to say, the reason i ask u to check
> open-iscsi is that we tested
> open-iscsi + iet on same box before, it is not ok,
> then after a certain
> version, open-iscsi can run ok with iet on same box,
> but nobody do
> extensive tests.
> 
> ming
> 
> On Sun, 2006-03-26 at 12:37 -0500, Ming Zhang wrote:
> > frankly, i do not know. i assumed they should be
> ok no matter whether
> > they are on same box or not. but from experience,
> it was not.
> > 
> > maybe some kernel guru here can answer this
> question.
> > 
> > ming
> > 
> > 
> > 
> > On Sun, 2006-03-26 at 09:18 -0800, Rik Herrin
> wrote:
> > > I'm running them on the same machine to see the
> > > maximum throughput and to see what type of
> bottleneck
> > > the disk subsystem will give me.  I'm also
> waiting for
> > > a 4-port 1GbE Intel NIC to make sure that the
> network
> > > doesn't end up being the bottleneck.  Do you
> think
> > > there's any reason for them running on the same
> > > machine to cause a problem?
> > > 
> > > Rik
> > > 
> > > --- Ming Zhang <mingz@xxxxxxxxxxx> wrote:
> > > 
> > > > ps, any point to run them at same box? ;)
> > > > 
> > > > ming
> > > > 
> > > > On Sun, 2006-03-26 at 08:30 -0800, Rik Herrin
> wrote:
> > > > > I am going to try it with open-iscsi and
> > > > core-iscsi to
> > > > > see if this issue is initiator-specific or
> IET
> > > > > specific.  In the meantime, I configured a
> 4-disk
> > > > RAID
> > > > > 0 SATA configuration and set it as the
> device
> > > > exported
> > > > > by IET.  Without the iSCSI layer (without
> running
> > > > > either ietd or iscsi-linux), I saw write
> > > > throughput of
> > > > > about 212 MB/s.  When I started IET and
> > > > linux-iscsi
> > > > > and ran the sequential write test, I am now
> > > > getting
> > > > > the following errors:
> > > > > Mar 26 18:37:28 localhost kernel:
> > > > iscsi-sfnet:host5:
> > > > > Failing command cdb 0x2a task 445 with
> return code
> > > > =
> > > > > 0x20000
> > > > > Mar 26 18:37:28 localhost kernel:
> > > > iscsi-sfnet:host5:
> > > > > Failing command cdb 0x2a task 446 with
> return code
> > > > =
> > > > > 0x20000
> > > > > Mar 26 18:37:28 localhost kernel:
> > > > iscsi-sfnet:host5:
> > > > > Failing command cdb 0x2a task 447 with
> return code
> > > > =
> > > > > 0x20000
> > > > > Mar 26 18:37:28 localhost kernel:
> > > > iscsi-sfnet:host5:
> > > > > Failing command cdb 0x2a task 448 with
> return code
> > > > =
> > > > > 0x20000
> > > > > Mar 26 18:37:28 localhost kernel:
> > > > iscsi-sfnet:host5:
> > > > > Failing command cdb 0x2a task 449 with
> return code
> > > > =
> > > > > 0x20000
> > > > > Mar 26 18:37:28 localhost kernel:
> > > > iscsi-sfnet:host5:
> > > > > Failing command cdb 0x2a task 450 with
> return code
> > > > =
> > > > > 0x20000
> > > > > Mar 26 18:37:28 localhost kernel:
> > > > iscsi-sfnet:host5:
> > > > > Failing command cdb 0x2a task 451 with
> return code
> > > > =
> > > > > 0x20000
> > > > > Mar 26 18:37:28 localhost kernel:
> > > > iscsi-sfnet:host5:
> > > > > Failing command cdb 0x2a task 452 with
> return code
> > > > =
> > > > > 0x20000
> > > > > Mar 26 18:37:28 localhost kernel:
> > > > iscsi-sfnet:host5:
> > > > > Failing command cdb 0x2a task 453 with
> return code
> > > > =
> > > > > 0x20000
> > > > > Mar 26 18:37:28 localhost kernel:
> > > > iscsi-sfnet:host5:
> > > > > Failing command cdb 0x2a task 454 with
> return code
> > > > =
> > > > > 0x20000
> > > > > Mar 26 18:37:28 localhost kernel:
> > > > iscsi-sfnet:host5:
> > > > > Failing command cdb 0x2a task 455 with
> return code
> > > > =
> > > > > 0x20000
> > > > > Mar 26 18:37:28 localhost kernel:
> > > > iscsi-sfnet:host5:
> > > > > Failing command cdb 0x2a task 456 with
> return code
> > > > =
> > > > > 0x20000
> > > > > Mar 26 18:37:28 localhost kernel:
> > > > iscsi-sfnet:host5:
> > > > > Failing command cdb 0x2a task 457 with
> return code
> > > > =
> > > > > 0x20000
> > > > > Mar 26 18:37:28 localhost kernel:
> > > > iscsi-sfnet:host5:
> > > > > Failing command cdb 0x2a task 458 with
> return code
> > > > =
> > > > > 0x20000
> > > > > ... [a lot more of the above]
> > > > > 
> > > > > followed by:
> > > > > 
> > > > > Mar 26 18:37:25 localhost kernel:
> > > > > execute_task_management(1177) 1e2 1 bd010000
> > > > > Mar 26 18:37:25 localhost kernel:
> cmnd_abort(1114)
> > > > 1bd
> > > > > 1 0 42 131072 0 0
> > > > > Mar 26 18:37:25 localhost kernel:
> > > > iscsi-sfnet:host5:
> > > > > task mgmt itt 482 rejected (0x1)
> > > > > Mar 26 18:37:25 localhost kernel:
> > > > iscsi-sfnet:host5:
> > > > > Sending ABORT TASK for task itt 446
> > > > > Mar 26 18:37:25 localhost kernel:
> > > > iscsi-sfnet:host5:
> > > > > Waking tx_thread to send task mgmt function
> itt
> > > > 483
> > > > > Mar 26 18:37:25 localhost kernel:
> > > > > execute_task_management(1177) 1e3 1 be010000
> > > > > Mar 26 18:37:25 localhost kernel:
> cmnd_abort(1114)
> > > > 1be
> > > > > 1 0 42 131072 0 0
> > > > > Mar 26 18:37:25 localhost kernel:
> > > > iscsi-sfnet:host5:
> > > > > task mgmt itt 483 rejected (0x1)
> > > > > Mar 26 18:37:25 localhost kernel:
> > > > iscsi-sfnet:host5:
> > > > > Sending ABORT TASK for task itt 447
> > > > > Mar 26 18:37:25 localhost kernel:
> > > > iscsi-sfnet:host5:
> > > > > Waking tx_thread to send task mgmt function
> itt
> > > > 484
> > > > > Mar 26 18:37:25 localhost kernel:
> > > > > execute_task_management(1177) 1e4 1 bf010000
> > > > > Mar 26 18:37:25 localhost kernel:
> cmnd_abort(1114)
> > > > 1bf
> > > > > 1 0 42 131072 0 0
> > > > > Mar 26 18:37:25 localhost kernel:
> > > > iscsi-sfnet:host5:
> > > > > task mgmt itt 484 rejected (0x1)
> > > > > Mar 26 18:37:25 localhost kernel:
> > > > iscsi-sfnet:host5:
> > > > > Sending ABORT TASK for task itt 448
> > > > > Mar 26 18:37:25 localhost kernel:
> 
=== message truncated ===


__________________________________________________
Do You Yahoo!?
Tired of spam?  Yahoo! Mail has the best spam protection around 
http://mail.yahoo.com 


-------------------------------------------------------
This SF.Net email is sponsored by xPML, a groundbreaking scripting language
that extends applications into web and mobile media. Attend the live webcast
and join the prime developer group breaking into this new coding territory!
http://sel.as-us.falkag.net/sel?cmd=lnk&kid=110944&bid=241720&dat=121642
_______________________________________________
linux-iscsi-users mailing list
linux-iscsi-users@xxxxxxxxxxxxxxxxxxxxx
https://lists.sourceforge.net/lists/listinfo/linux-iscsi-users

[IP Storage]     [IETF]     [Linux SCSI]     [iSCSI Book]     [Linux Resources]     [Yosemite News]     [Photo]     [Home]     [IETF Announcements]     [IETF Discussion]     [SCSI Hardware]