| Author |
Message
|
| cicsprog |
 Posted: Fri Jun 01, 2007 1:22 pm Post subject: Any z/OS v6 Open APARs or bad contact admin going on? |
 |
|
Partisan
Joined: 27 Jan 2002
Posts: 347
|
My MVS team is shipping MQ v6 (on SYSRES) from our sandbox to the 100 z/OS MQM's on test and prod LPARS that I SysAdmin. I've taken it for a pretty good spin on the sandbox LPARs. The IBM mainteance level is RSU0704.
We are large users of MQ components that touch CICS, IMS, use CAF, CLUSTERing, etc. However, no SHARED Queues (yet).
Anybody have any open PMRs or big problems going on for z/OS MQ v6? I like my sleep. |
|
| Back to top |
|
 |
| PeterPotkay |
| Posted: Fri Jun 01, 2007 5:48 pm Post subject: |
|
|
Poobah
Joined: 15 May 2001
Posts: 7722
|
|
| Back to top |
|
|
| zpat |
| Posted: Sat Jun 02, 2007 2:27 am Post subject: |
|
|
Jedi Council
Joined: 19 May 2001
Posts: 5866
Location: UK
|
I have an open PMR which has been assigned an APAR number, but it's fairly obscure.
APAR PK46118 |
|
| Back to top |
|
|
| belchman |
| Posted: Fri Jul 13, 2007 11:59 am Post subject: |
|
|
Partisan
Joined: 31 Mar 2006
Posts: 386
Location: Ohio, USA
|
We had an incident where an MQ App was allowed to pass a negative MSG_LENGTH (via buffer overwrite) upon put which caused channel to abend. IBM fixed to prevent channel abend if negative value passed.
_________________
Make three correct guesses consecutively and you will establish a reputation as an expert. ~ Laurence J. Peter |
|
| Back to top |
|
|
| PeterPotkay |
| Posted: Fri Jul 13, 2007 6:54 pm Post subject: |
|
|
Poobah
Joined: 15 May 2001
Posts: 7722
|
6.0 on z/OS - If the messages are non persistent and the MQPUT happens without the syncpoint option while there is an outstanding MQGET with wait without syncpoint, the message is passed in memory. Better performance. But Queue Statistics are bypassed! You enqueue and dequeue counts don't get incremented.
Beware if you rely on queue statistics for chargeback based on message volume. MQ 6.0 on z/OS breaks it for the scenario described above! Its "not a bug, its by design". Uh, yeah.
But there is a fix you can ask for that makes it act like 5.3.
_________________
Peter Potkay
Keep Calm and MQ On |
|
| Back to top |
|
|
| cicsprog |
| Posted: Fri Aug 03, 2007 10:29 am Post subject: |
|
|
Partisan
Joined: 27 Jan 2002
Posts: 347
|
I get an intermittent 2055 from a CSQUTIL MAKEDEF in our nightly backups. This problem seems to have raised it head in several releases though.
But so far v6 has been clean for us. That should be the kiss of death. |
|
| Back to top |
|
|
| jeevan |
| Posted: Mon Aug 04, 2008 8:29 am Post subject: |
|
|
Grand Master
Joined: 12 Nov 2005
Posts: 1432
|
| PeterPotkay wrote: |
6.0 on z/OS - If the messages are non persistent and the MQPUT happens without the syncpoint option while there is an outstanding MQGET with wait without syncpoint, the message is passed in memory. Better performance. But Queue Statistics are bypassed! You enqueue and dequeue counts don't get incremented.
Beware if you rely on queue statistics for chargeback based on message volume. MQ 6.0 on z/OS breaks it for the scenario described above! Its "not a bug, its by design". Uh, yeah.
But there is a fix you can ask for that makes it act like 5.3. |
I have to do performance test ( find enqueue and dequeue count) on zos. Is there any utility /way to do that? This would be a great help. |
|
| Back to top |
|
|
|
|
