| Author |
Message
|
| KeeferG |
 Posted: Tue Jun 20, 2006 8:01 am Post subject: AMQ9206 and AMQ9028 |
 |
|
Master
Joined: 15 Oct 2004
Posts: 215
Location: Basingstoke, UK
|
Hi all,
I am running MQ5.3.12 on Solaris
I keep getting channels failing with AMQ9026 and AMQ9208 messages. We get several a day in our cluster of 24 queue managers.
The AMQ9206 is below
-------------------------------------------------------------------------------
06/19/06 18:07:46
AMQ9206: Error sending data to host 10.252.18.33.
EXPLANATION:
An error occurred sending data over TCP/IP to 10.252.18.33. This may be due to a communications failure.
ACTION:
The return code from the TCP/IP(write) call was 32 X('20'). Record these values and tell your systems administrator.
The AMQ9208 is below
-------------------------------------------------------------------------------
06/19/06 18:07:00
AMQ9208: Error on receive from host ss44gfspar04 (10.252.18.24).
EXPLANATION:
An error occurred receiving data from ss44gfspar04 (10.252.18.24) over TCP/IP. This may be due to a communications failure.
ACTION:
The return code from the TCP/IP (read) call was 22 (X'16'). Record these values and tell the systems administrator.
Does anybody else get these messages on such a ragular basis.
Cheers
Keith
_________________
Keith Guttridge
-----------------
Using MQ since 1995 |
|
| Back to top |
|
 |
| jefflowrey |
| Posted: Tue Jun 20, 2006 8:02 am Post subject: |
|
|
Grand Poobah
Joined: 16 Oct 2002
Posts: 19981
|
Is there a firewall in the way?
_________________
I am *not* the model of the modern major general. |
|
| Back to top |
|
|
| KeeferG |
| Posted: Tue Jun 20, 2006 8:07 am Post subject: |
|
|
Master
Joined: 15 Oct 2004
Posts: 215
Location: Basingstoke, UK
|
We have firewalls between sites but we see the errors whether we go cross site or not.
_________________
Keith Guttridge
-----------------
Using MQ since 1995 |
|
| Back to top |
|
|
| wschutz |
| Posted: Tue Jun 20, 2006 10:38 am Post subject: |
|
|
Jedi Knight
Joined: 02 Jun 2005
Posts: 3316
Location: IBM (retired)
|
Hello Keith.... any FDC's thrown?
_________________
-wayne |
|
| Back to top |
|
|
| KeeferG |
| Posted: Wed Jun 21, 2006 1:18 am Post subject: |
|
|
Master
Joined: 15 Oct 2004
Posts: 215
Location: Basingstoke, UK
|
Hi Wayne,
We are getting no FDC's. I was hoping someone else had a similar issue.
We are losing about 7 channels a day this way so need to find the cause.
We do get an unusual error AMQ6090 before some of the AMQ9206 messages
06/20/06 17:30:00
06/20/06 17:17:38
AMQ6090
SP.AS01.MANAGER
7FFE
WebSphere MQ was unable to display an error message 7FFE.
MQ has attempted to display the message associated with return code hexadecimal '7FFE'. The return code indicates that there is no message text associated with the message. Associated with the request are inserts 825241138 : 892481073 : .21 : : . Use the standard facilities supplied with your system to record the problem identifier, and to save the generated output files. Contact your IBM support center. Do not discard these files until the problem has been resolved.
Have a PMR open but no joy as of yet.
I am currently collecting the log information to try and see any patterns so we can try a trace. Its a production system so can't leave trace on all machines.
_________________
Keith Guttridge
-----------------
Using MQ since 1995 |
|
| Back to top |
|
|
| KeeferG |
| Posted: Fri Sep 07, 2007 6:20 am Post subject: |
|
|
Master
Joined: 15 Oct 2004
Posts: 215
Location: Basingstoke, UK
|
Well, two years have gone by and still no solution. The only thing that has happened is people are very confused by it.
We have run snoops on the network and seen that there is a FIN being sent just before MQ receives its broken pipe errors. Tracing MQ shows that it isnt coming from it so I think that just leaves the solaris tcpip stack.
Networks are sayign that it isnt them. I'm saying that it isn't MQ and we cannot see into tcpip to see what is going on.
Since having the problem we have gone from 5.3.5 to 5.3.8 to 5.3.12 and now 6.0.2.1. We have applied several OS patches to our Solaris 5.8.
If anyone has any ideas it would be greatly appreciated.
_________________
Keith Guttridge
-----------------
Using MQ since 1995 |
|
| Back to top |
|
|
| jefflowrey |
| Posted: Fri Sep 07, 2007 7:10 am Post subject: |
|
|
Grand Poobah
Joined: 16 Oct 2002
Posts: 19981
|
Two years? It seems like 2006 was only last year... 
The PMR was no help?
You should ask the network folks if their firewalls are configured to shut down "idle" connections, or connections that have been open for a while.
I have seen other people run into problems because of this.
You should also look at the Solaris TCP stack KeepAlive parameters.
_________________
I am *not* the model of the modern major general. |
|
| Back to top |
|
|
| redhatsup |
| Posted: Wed Aug 11, 2010 8:38 pm Post subject: Same issue |
|
|
Newbie
Joined: 11 Aug 2010
Posts: 1
|
Does anyone have any update or solution for this problem? We are also facing the same error as mentioned in the thread.
Thanking in advance for your support. |
|
| Back to top |
|
|
| gbaddeley |
| Posted: Thu Aug 12, 2010 4:22 pm Post subject: |
|
|
Jedi Knight
Joined: 25 Mar 2003
Posts: 2538
Location: Melbourne, Australia
|
| Code: |
EXPLANATION:
An error occurred sending data over TCP/IP to 10.252.18.33. This may be due to a communications failure.
ACTION:
The return code from the TCP/IP(write) call was 32 X('20'). Record these values and tell your systems administrator.
The AMQ9208 is below
-------------------------------------------------------------------------------
06/19/06 18:07:00
AMQ9208: Error on receive from host ss44gfspar04 (10.252.18.24).
EXPLANATION:
An error occurred receiving data from ss44gfspar04 (10.252.18.24) over TCP/IP. This may be due to a communications failure.
ACTION:
The return code from the TCP/IP (read) call was 22 (X'16'). Record these values and tell the systems administrator. |
The return codes from TCP/IP are Unix errno values. There is a nice list at http://www.koders.com/c/fid53B3BE16088320EE31E67B12F901FE3FB9D1F480.aspx , and you can search for errno's on Internet for more information..
22 and 32 indicate there might be something wrong with the local TCP/IP stack or the network.
HTH
_________________
Glenn |
|
| Back to top |
|
|
|
|
