| Author |
Message
|
| cspodapati |
 Posted: Thu Mar 04, 2010 2:17 am Post subject: increase in Queue Depth |
 |
|
Novice
Joined: 12 Jan 2010
Posts: 20
|
Hi,
I am facing one problem,
We are using z/OS as source system and HP-UX as destination system.
In z/OS we are using 5.3 and we using 5x in HP-UX side
The problem is 200 messages were in transmission queue at z/OS side, and those were not transmitting to destination.
WE checked both the ends the corresponding channels are running fine,
Can any one tell me to resolve the problem, May i know why it is happening ?
Thanks in Advance |
|
| Back to top |
|
 |
| Mr Butcher |
| Posted: Thu Mar 04, 2010 2:58 am Post subject: |
|
|
Padawan
Joined: 23 May 2005
Posts: 1716
|
are the messages comitted on the z/OS xmitq? they will not be transferred unless comitted. or maybe thay are indoubt.... check the putting system / transaction / ......
if they are comitted, they should be transferred when the channel is running (assuming correct definitions and so on). you may also want to check (the log files) if there is a problem on the receiving end, and if the channel retry is used. this may also look like a running channel with messages not being transmitted .....
_________________
Regards, Butcher |
|
| Back to top |
|
|
| cspodapati |
| Posted: Thu Mar 04, 2010 4:38 am Post subject: |
|
|
Novice
Joined: 12 Jan 2010
Posts: 20
|
We got this problem twice, first we stopped the Requester channel and started again at that time all the messages are received,
this time first solution did not for me, and we reseted the channels at both the ends then the problem got resolved
I am not sure why it is happening like this |
|
| Back to top |
|
|
| Vitor |
| Posted: Thu Mar 04, 2010 6:07 am Post subject: |
|
|
Grand High Poobah
Joined: 11 Nov 2005
Posts: 26093
Location: Texas, USA
|
| cspodapati wrote: |
| this time first solution did not for me, and we reseted the channels at both the ends then the problem got resolved |
How did you arrive at that solution? Or did you just issue a reset in the hope it would work?
| cspodapati wrote: |
| I am not sure why it is happening like this |
And if you don't investigate the problem you will never be sure why it's happening. Of course, if you're content to reset the channels each time there's a problem that's not an issue.
You need to check the log output, channel status and other issues as my worthy associate has already commented. Something is interferring with the channel and you're be better served identifying the root cause.
_________________
Honesty is the best policy.
Insanity is the best defence. |
|
| Back to top |
|
|
| Mr Butcher |
| Posted: Thu Mar 04, 2010 7:00 am Post subject: |
|
|
Padawan
Joined: 23 May 2005
Posts: 1716
|
| Quote: |
We got this problem twice, first we stopped the Requester channel and started again at that time all the messages are received,
this time first solution did not for me, and we reseted the channels at both the ends then the problem got resolved |
you should check the log on both ends (even afterwards this is possible). if there is no sequence number error in there, then resetting the channels was of no help. you said the channels where in state "running" ?!? a channel, that is out of sequence will be in state retry.
i agree with vitor - if you are in the error situation, don't only try things, try to find out what the reason is.
did you check the receiving end for the retry - channel parameters? how about this scenario... you send form z/Os to HP-UX. maybe many data, or hp-ux is slow. so queue is full. but you use the retry of the receiver channel. so the channel will wait some time and then retry to put the messages. during that time, you stopped / started / resetted the channel, which looked like it helped, but that was not the reason. time passed, and messages where processed, and the channel was able to put the messages.
thats just a guess, but maybe worth to investigate. and dont reset if you are not out of sequence.
_________________
Regards, Butcher |
|
| Back to top |
|
|
| cspodapati |
| Posted: Mon Mar 08, 2010 8:08 pm Post subject: |
|
|
Novice
Joined: 12 Jan 2010
Posts: 20
|
thanks for your replies
and sorry for delay
I have gone through the log file, in that i saw " message sequence
number error " very frequently.
can you please clarify, if we get this type of errors, is it the problem with sender channel or receiver channel or both the ends ?
and one more thing i found the below info in IBM side
"" The rate at which
AMQ9526 is being reported will typically be directly
proportional to the rate at which memory is leaked. ""
The memory leak will happen at MQ level or at Operating System level.
Can any one clarify these things
Thanks in advance |
|
| Back to top |
|
|
| Vitor |
| Posted: Mon Mar 08, 2010 8:20 pm Post subject: |
|
|
Grand High Poobah
Joined: 11 Nov 2005
Posts: 26093
Location: Texas, USA
|
| cspodapati wrote: |
| can you please clarify, if we get this type of errors, is it the problem with sender channel or receiver channel or both the ends ? |
Could be either or both. You'll need more investigation to determine which.
| cspodapati wrote: |
and one more thing i found the below info in IBM side
"" The rate at which
AMQ9526 is being reported will typically be directly
proportional to the rate at which memory is leaked. ""
The memory leak will happen at MQ level or at Operating System level.
Can any one clarify these things |
If you tell us where you found this; the IBM site's a big place. Post the link.
_________________
Honesty is the best policy.
Insanity is the best defence. |
|
| Back to top |
|
|
| cspodapati |
| Posted: Mon Mar 08, 2010 9:48 pm Post subject: |
|
|
Novice
Joined: 12 Jan 2010
Posts: 20
|
|
| Back to top |
|
|
| bruce2359 |
| Posted: Tue Mar 09, 2010 6:06 am Post subject: |
|
|
Poobah
Joined: 05 Jan 2008
Posts: 9469
Location: US: west coast, almost. Otherwise, enroute.
|
The link you provided says that this issue (if it is your issue or not remains to be seen) is resolved at WMQ v6. You are at v5.
_________________
I like deadlines. I like to wave as they pass by.
ב''ה
Lex Orandi, Lex Credendi, Lex Vivendi. As we Worship, So we Believe, So we Live. |
|
| Back to top |
|
|
| cspodapati |
| Posted: Wed Mar 10, 2010 12:45 am Post subject: |
|
|
Novice
Joined: 12 Jan 2010
Posts: 20
|
| Vitor wrote: |
| Could be either or both. You'll need more investigation to determine which. |
Can you please guide me in doing investigation about previous line
Thanks
[/list][/code] |
|
| Back to top |
|
|
| Vitor |
| Posted: Wed Mar 10, 2010 6:06 am Post subject: |
|
|
Grand High Poobah
Joined: 11 Nov 2005
Posts: 26093
Location: Texas, USA
|
| cspodapati wrote: |
Can you please guide me in doing investigation about previous line
|
Well if the 2 ends of the channel are unable to resolve their differences automatically regarding sequence numbers something non-WMQ has happened. Could be manual intervention, could be network, could be application. You have 2 sequence numbers to work with; investigate with the external systems to determine which sequence number matches the actual state of the messages transmitted.
For instance, what happened just before the error? Was there maintenance? A command issued? An application error? A database error? The list goes on and on.
_________________
Honesty is the best policy.
Insanity is the best defence. |
|
| Back to top |
|
|
|
|
