open-iscsi logout errors

2008-08-28 Thread gtdaqua

I am using open-iscsi-2.0-869.2 and even since the 2.0-868.x releases,
I am
facing this error when logging out. Out of the four targets (all go to
one
box), only one session successfully logs out. Other three fail. The 3
are sessions are still
open. Even killing iscsid process manually doesnt help.

The following is a continuous output:
*
[EMAIL PROTECTED]:~$ sudo /etc/init.d/open-iscsi stop
 * Disconnecting iSCSI targetsLogging out of session [sid: 2, target:
iqn.1984-05.com.sample:example.0061c23000ce22f047626207,
portal:
192.168.202.99,3260]
Logging out of session [sid: 3, target:
iqn.1984-05.com.sample:example.0061c23000ce22f047626207,
portal:
192.168.202.101,3260]
Logging out of session [sid: 4, target:
iqn.1984-05.com.sample:example.0061c23000ce22f047626207,
portal:
192.168.202.100,3260]
Logging out of session [sid: 5, target:
iqn.1984-05.com.sample:example.0061c23000ce22f047626207,
portal:
192.168.202.98,3260]
Logout of [sid: 2, target:
iqn.1984-05.com.sample:example.0061c23000ce22f047626207,
portal:
192.168.202.99,3260]: successful
iscsiadm: got read error (0/2), daemon died?
iscsiadm: Could not logout of [sid: 3, target:
iqn.1984-05.com.sample:example.0061c23000ce22f047626207,
portal:
192.168.202.101,3260]:
iscsiadm: initiator reported error (18 - could not communicate to
iscsid)
iscsiadm: got read error (0/2), daemon died?
iscsiadm: Could not logout of [sid: 4, target:
iqn.1984-05.com.sample:example.0061c23000ce22f047626207,
portal:
192.168.202.100,3260]:
iscsiadm: initiator reported error (18 - could not communicate to
iscsid)
iscsiadm: got read error (0/2), daemon died?
iscsiadm: Could not logout of [sid: 5, target:
iqn.1984-05.com.sample:example.0061c23000ce22f047626207,
portal:
192.168.202.98,3260]:
iscsiadm: initiator reported error (18 - could not communicate to
iscsid)
                                                                     
                                                                     
      [
OK ]
 * Stopping iSCSI initiator service                                  
                                                                     
     
[ OK ]
[EMAIL PROTECTED]:~$ sudo /etc/init.d/open-iscsi status
Current active iSCSI sessions:
tcp: [3] 192.168.202.101:3260,2
iqn.1984-05.com.sample:example.0061c23000ce22f047626207
tcp: [4] 192.168.202.100:3260,2
iqn.1984-05.com.sample:example.0061c23000ce22f047626207
tcp: [5] 192.168.202.98:3260,1
iqn.1984-05.com.sample:example.0061c23000ce22f047626207
[EMAIL PROTECTED]:~$

[EMAIL PROTECTED]:~$ sudo iscsiadm -m session
tcp: [3] 192.168.202.101:3260,2
iqn.1984-05.com.sample:example.0061c23000ce22f047626207
tcp: [4] 192.168.202.100:3260,2
iqn.1984-05.com.sample:example.0061c23000ce22f047626207
tcp: [5] 192.168.202.98:3260,1
iqn.1984-05.com.sample:example.0061c23000ce22f047626207
[EMAIL PROTECTED]:~$
*

However, there is no data loss. Strangely, every JFS/XFS file systems
have to
repaired and mounted during every restart of the server.

Best regards,

--
A.S. Arvind
--~--~-~--~~~---~--~~
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 [EMAIL PROTECTED]
For more options, visit this group at http://groups.google.com/group/open-iscsi
-~--~~~~--~~--~--~---



Re: Input/output error and strange problem ?

2008-08-28 Thread Chuanwen Wu

Hi, Konrad!
Thanks for you advice!

> Are you mounting that filesystem across many nodes (ie, through my 
> iscsi-initators)?
>
Yes, that's exactly what I had done. I  want to use iscsi to transmit
data in my distributed filesystem.

> If so, then this is expected. ext3 is not clustered-aware filesystem and won't
> behave properly. You would need to use GFS or GFS2 to do this.
GFS2 or GFS seems have some problem in Genoo across to bugzilla.
So you just think GFS/GFS2 is good enough ? Or is there any other
recommended clustered-aware filesystem?

-- 
wcw

--~--~-~--~~~---~--~~
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 [EMAIL PROTECTED]
For more options, visit this group at http://groups.google.com/group/open-iscsi
-~--~~~~--~~--~--~---



Re: Input/output error and strange problem ?

2008-08-28 Thread Konrad Rzeszutek

On Thu, Aug 28, 2008 at 11:21:19PM +0800, Chuanwen Wu wrote:
> 
> Hi,
> 
> I use open-iscsi-2.0.869.2 and iscsitarget-0.4.15-r1, and in my test,
> some strange errors occur.
> 
> In all my iscsi-target, I have a ext3 fs. And I mount these fs in the
> iscsi-initator.

Are you mounting that filesystem across many nodes (ie, through my 
iscsi-initators)?

If so, then this is expected. ext3 is not clustered-aware filesystem and won't
behave properly. You would need to use GFS or GFS2 to do this.


--~--~-~--~~~---~--~~
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 [EMAIL PROTECTED]
For more options, visit this group at http://groups.google.com/group/open-iscsi
-~--~~~~--~~--~--~---



Re: Input/output error and strange problem ?

2008-08-28 Thread Chuanwen Wu

I use gentoo, the kernel is 2.6.24-r4
-- 
wcw

--~--~-~--~~~---~--~~
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 [EMAIL PROTECTED]
For more options, visit this group at http://groups.google.com/group/open-iscsi
-~--~~~~--~~--~--~---



Input/output error and strange problem ?

2008-08-28 Thread Chuanwen Wu

Hi,

I use open-iscsi-2.0.869.2 and iscsitarget-0.4.15-r1, and in my test,
some strange errors occur.

In all my iscsi-target, I have a ext3 fs. And I mount these fs in the
iscsi-initator.


1, Input/output error

Once  I copied a file into iscsi-target,  I got the error
"Input/output error" after a "lseek" operation. And then I tried
copying again, I succeed this time. I still found more similar
problems in "open" operation.
Why sometimes it failed and sometimes succeed?

2, latency problem
Here is the log of the operation in the iscsi target:

$ ls
lost+found
$cat hello
hello
$ls
lost+found

As you can see, when I did "ls", I didn't see the file "hello", but
the strange thing is that I could do "cat hello" and I got the correct
result.
Why I can't "ls" but can "cat"?


Any help will be appreciated!
-- 
wcw

--~--~-~--~~~---~--~~
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 [EMAIL PROTECTED]
For more options, visit this group at http://groups.google.com/group/open-iscsi
-~--~~~~--~~--~--~---