| Author |
Message
|
| cicsprog |
 Posted: Thu Mar 30, 2017 11:32 am Post subject: SEQUENCE Number weirdness???? |
 |
|
Partisan
Joined: 27 Jan 2002
Posts: 347
|
I was moving over some Automation (AO) to a new site. The AO traps on CSQX526E and issues a RESET to fix the mismatch. It always worked at other sites.
When I try to create a SEQUENCE number mismatch between two ZOS MQs at v8, I can't get the error to come out. Tried
- I deleted one side of the channel pair after sending messages, then redefined. But still didn’t raise error
- Issued my own RESET to a large sequence number on one side of the channel. That didn’t work.
-Issued a NETSTAT DROP on the connection and some other stuff and that wouldn’t raise the error.
-edited SYNCQ and deleted msgs for the channels
Did v8 add some sort of auto fix or ignore Sequence number parameter? |
|
| Back to top |
|
 |
| PaulClarke |
| Posted: Thu Mar 30, 2017 11:57 am Post subject: |
|
|
Grand Master
Joined: 17 Nov 2005
Posts: 1002
Location: New Zealand
|
What type of channels are connecting these two Queue Managers ?
_________________
Paul Clarke
MQGem Software
www.mqgem.com |
|
| Back to top |
|
|
| cicsprog |
| Posted: Thu Mar 30, 2017 12:00 pm Post subject: |
|
|
Partisan
Joined: 27 Jan 2002
Posts: 347
|
|
| Back to top |
|
|
| PaulClarke |
| Posted: Thu Mar 30, 2017 12:20 pm Post subject: |
|
|
Grand Master
Joined: 17 Nov 2005
Posts: 1002
Location: New Zealand
|
Shame, was hoping that it was a Cluster pair since they don't check sequence numbers and it would have explained it.
_________________
Paul Clarke
MQGem Software
www.mqgem.com |
|
| Back to top |
|
|
| mqjeff |
| Posted: Thu Mar 30, 2017 12:37 pm Post subject: |
|
|
Grand Master
Joined: 25 Jun 2008
Posts: 17447
|
You have to issue the RESET on the sender side, I think, to get an out of sequence.
I believe the receiver causes the channels to resync.
Maybe it's the other way around?
Whichever way you tried to reset to a large number, try the other way.
_________________
chmod -R ugo-wx / |
|
| Back to top |
|
|
| PaulClarke |
| Posted: Thu Mar 30, 2017 12:39 pm Post subject: |
|
|
Grand Master
Joined: 17 Nov 2005
Posts: 1002
Location: New Zealand
|
It's the other way. You have to issue the RESET on the receiver end to 'cause' a problem. If the RESET is issued on the sender then it just tells the receiver the new start point when you next start the channel.
_________________
Paul Clarke
MQGem Software
www.mqgem.com |
|
| Back to top |
|
|
| hughson |
| Posted: Thu Mar 30, 2017 12:53 pm Post subject: Re: SEQUENCE Number weirdness???? |
|
|
Padawan
Joined: 09 May 2013
Posts: 1959
Location: Bay of Plenty, New Zealand
|
| cicsprog wrote: |
Tried
- I deleted one side of the channel pair after sending messages, then redefined. But still didn't raise error |
I think it matters whether the messages you sent were persistent. Try this one again ensuring that the messages you sent were persistent messages.
| cicsprog wrote: |
| - Issued my own RESET to a large sequence number on one side of the channel. That didn't work. |
As noted by others, this must be done on the RECEIVER end to work.
| cicsprog wrote: |
| -Issued a NETSTAT DROP on the connection and some other stuff and that wouldn't raise the error. |
Not sure this would be destructive enough to cause a sequence number error.
| cicsprog wrote: |
| -edited SYNCQ and deleted msgs for the channels |
This should have the same effect as deleting the channel definition. You have to know how to find the correct messages though. It's not the ones with the Msg/CorrelID of the channel name. You need to find the ones with the channel name in the body of the message.
Cheers
Morag
_________________
Morag Hughson @MoragHughson
IBM MQ Technical Education Specialist
Get your IBM MQ training here!
MQGem Software |
|
| Back to top |
|
|
| cicsprog |
| Posted: Thu Mar 30, 2017 1:34 pm Post subject: |
|
|
Partisan
Joined: 27 Jan 2002
Posts: 347
|
| Homer Doh....Ya...they needed to be PRES msgs...darn...my ISPF utility set to NON...nice catch Morag!!! My 58 year old brain is shot. |
|
| Back to top |
|
|
| PeterPotkay |
| Posted: Thu Mar 30, 2017 3:33 pm Post subject: |
|
|
Poobah
Joined: 15 May 2001
Posts: 7722
|
If sequence numbers could safely be reset automatically I think IBM would have coded MQ to do it on its own. Unless you are prone to frequent MQ channel sequence number mismatches for known reasons that allow for an automatic reset?
_________________
Peter Potkay
Keep Calm and MQ On |
|
| Back to top |
|
|
| cicsprog |
| Posted: Fri Mar 31, 2017 7:39 am Post subject: |
|
|
Partisan
Joined: 27 Jan 2002
Posts: 347
|
| PeterPotkay wrote: |
| If sequence numbers could safely be reset automatically I think IBM would have coded MQ to do it on its own. Unless you are prone to frequent MQ channel sequence number mismatches for known reasons that allow for an automatic reset? |
Like server admins that rebuild their servers and QManagers frequently you mean? I'm not defending the method only trying to stop the madness. |
|
| Back to top |
|
|
| PeterPotkay |
| Posted: Fri Mar 31, 2017 1:35 pm Post subject: |
|
|
Poobah
Joined: 15 May 2001
Posts: 7722
|
| cicsprog wrote: |
| PeterPotkay wrote: |
| If sequence numbers could safely be reset automatically I think IBM would have coded MQ to do it on its own. Unless you are prone to frequent MQ channel sequence number mismatches for known reasons that allow for an automatic reset? |
Like server admins that rebuild their servers and QManagers frequently you mean? I'm not defending the method only trying to stop the madness. |
 
Come on over here when you are done, plenty of madness to go around.
_________________
Peter Potkay
Keep Calm and MQ On |
|
| Back to top |
|
|
|
|
