FAQ Search Today's Posts Mark Forums Read
» Video Reviews

» Linux Archive

Linux-archive is a website aiming to archive linux email lists and to make them easily accessible for linux users/developers.


» Sponsor

» Partners

» Sponsor

Go Back   Linux Archive > Redhat > Red Hat Linux

 
 
LinkBack Thread Tools
 
Old 08-31-2010, 04:44 PM
Nicoli Stupinski
 
Default ISCSI disk goes read-only after network flooding

My ISCSI disk has been going into a read-only state due to some network
flooding issues. While I wait for the network team to fix things on their
end (and this may take some time) is it possible to have the disk come back
online read/write instead of read-only?

More details..

Here is what I'm seeing in the "messages" log...

Aug 30 06:49:00 node1 kernel: qla3xxx eth0: Link is down.
Aug 30 06:49:08 node1 kernel: connection1:0: ping timeout of 5 secs
expired, recv timeout 5, last rx 5988451077, last ping 5988456077, now
5988461077
Aug 30 06:49:08 node1 kernel: connection1:0: detected conn error (1011)
Aug 30 06:49:09 node1 kernel: qla3xxx eth0: Link is down.
Aug 30 06:49:09 node1 iscsid: Kernel reported iSCSI connection 1:0 error
(1011) state (3)
Aug 30 06:49:14 node1 iscsid: connect to 10.1.1.1:3260 failed (No route to
host)
Aug 30 06:51:08 node1 kernel: session1: session recovery timed out after
120 secs
Aug 30 06:51:08 node1 kernel: sd 9:0:0:0: SCSI error: return code =
0x000f0000
Aug 30 06:51:08 node1 kernel: end_request: I/O error, dev sdd, sector
1061945423
Aug 30 06:51:08 node1 kernel: printk: 4 messages suppressed.
Aug 30 06:51:08 node1 kernel: Buffer I/O error on device sdd1, logical block
132743170
Aug 30 06:51:08 node1 kernel: lost page write due to I/O error on sdd1
<similar stuff deleted>

And here is where the network must have came back online (although I never
did see a "Link up", but it's definitely up) and mounted the disk
read-only...

Aug 30 06:51:08 node1 kernel: ext3_abort called.
Aug 30 06:51:08 node1 kernel: EXT3-fs error (device sdd1):
ext3_journal_start_sb: Detected aborted journal
Aug 30 06:51:08 node1 kernel: Remounting filesystem read-only
Aug 30 06:51:08 node1 kernel: __journal_remove_journal_head: freeing
b_frozen_data
So my question is, to help get us through these trying times, is it possible
to make it so that the disk automatically comes back online read-write?
And/Or, is there a certain timeout value I can increase?

I'm running RHEL 5.5 and iscsi-initiator-utils.x86_64 6.2.0.871-0.16.el5

Thanks,
Nic
--
redhat-list mailing list
unsubscribe mailto:redhat-list-request@redhat.com?subject=unsubscribe
https://www.redhat.com/mailman/listinfo/redhat-list
 
Old 08-31-2010, 04:57 PM
Eugene Vilensky
 
Default ISCSI disk goes read-only after network flooding

On Tue, Aug 31, 2010 at 11:44 AM, Nicoli Stupinski <nstupinski@gmail.com> wrote:
> My ISCSI disk has been going into a read-only state due to some network
> flooding issues. *While I wait for the network team to fix things on their
> end (and this may take some time) is it possible to have the disk come back
> online read/write instead of read-only?
>
> More details..

Can you increase your timeout values so that instead of returning an
error, it waits for the disk?

>From Dell Equallogic's recommendations, for example:

"
For most Linux operating systems, use the following script to update
the timeout value:
for i in ‘ls /sys/class/scsi_disk‘;
do echo "60" > /sys/class/scsi_disk/$i/device/timeout;
done

For Red Hat operating systems, use the following script to update the
timeout value:
for i in ‘ls /sys/class/scsi_device‘;
do echo "60" > /sys/class/scsi_device/$i/device/timeout;
done
You can add the appropriate script to your system start up script. The
location of the startup script varies with
different Linux distributions; for example, /etc/rc.local, /etc/init.d/rc.local.
If devices are connected after startup, run this script again to
update the new devices’ timeout value.
'

--
redhat-list mailing list
unsubscribe mailto:redhat-list-request@redhat.com?subject=unsubscribe
https://www.redhat.com/mailman/listinfo/redhat-list
 
Old 09-02-2010, 01:06 PM
"Greaser, Tom"
 
Default ISCSI disk goes read-only after network flooding

You may want to ask you network team to create a iscsi vlan. I'm not sure what flooding on the network means to your team. To me. To much bcast traffic. Or to much traffic on that link causing your interface to time out with your storage array.

Eugene. Thanks for the info very cool
--
redhat-list mailing list
unsubscribe mailto:redhat-list-request@redhat.com?subject=unsubscribe
https://www.redhat.com/mailman/listinfo/redhat-list
 
Old 09-02-2010, 03:03 PM
Nicoli Stupinski
 
Default ISCSI disk goes read-only after network flooding

Thanks Tom and Eugene for the replies.

Eugene,

What about the timeout param in /etc/iscsi/iscsid.conf, such as...

# To specify the length of time to wait for session re-establishment
# before failing SCSI commands back to the application when running
# the Linux SCSI Layer error handler, edit the line.
# The value is in seconds and the default is 120 seconds.
node.session.timeo.replacement_timeout = 120
Should I change this as well?

Thanks,
Nic
--
redhat-list mailing list
unsubscribe mailto:redhat-list-request@redhat.com?subject=unsubscribe
https://www.redhat.com/mailman/listinfo/redhat-list
 

Thread Tools




All times are GMT. The time now is 06:35 AM.

VBulletin, Copyright ©2000 - 2014, Jelsoft Enterprises Ltd.
Content Relevant URLs by vBSEO ©2007, Crawlability, Inc.
Copyright 2007 - 2008, www.linux-archive.org