| Author |
Message
|
| Vitor |
 Posted: Fri Oct 21, 2011 8:30 am Post subject: |
 |
|
Grand High Poobah
Joined: 11 Nov 2005
Posts: 26093
Location: Texas, USA
|
| Gaya3 wrote: |
| my only suggestion to SeshaGiri is "Do not just escape from the issue..by resolving it by making Persistent messages, think logically, reset the channel and see does it work or not. |
I did wonder if the question was based on another questionable plan to make the magic come back.
_________________
Honesty is the best policy.
Insanity is the best defence. |
|
| Back to top |
|
 |
| Seshagiri |
| Posted: Fri Oct 21, 2011 8:43 am Post subject: |
|
|
Apprentice
Joined: 04 Oct 2011
Posts: 42
|
| Vitor wrote: |
Only to the limit you defined when you built the queue manager. Once those limits are reached the queue manager will consider the log full, and act accordingly. |
But you know there were 32 log files and when the application team made the messages as persistent we can see 37 log files. More over we have defined 31 as primary and 32 as secondary logs in q.ini file. |
|
| Back to top |
|
|
| Vitor |
| Posted: Fri Oct 21, 2011 8:50 am Post subject: |
|
|
Grand High Poobah
Joined: 11 Nov 2005
Posts: 26093
Location: Texas, USA
|
| Seshagiri wrote: |
| But you know there were 32 log files and when the application team made the messages as persistent we can see 37 log files. More over we have defined 31 as primary and 32 as secondary logs in q.ini file. |
I don't see the point you're making. So you've created more persistent messages and thus increased the number of log files by 5. This is not surprising.
Unless the point youre' making is that you don't understand the maximum number of log files is the sum of the primary and secondary, in your case 63. Given the lack of WMQ knowledge you've demonstrated thus far (in case of emergency reset the channel and see if it helps) I suppose that was to be expected.
_________________
Honesty is the best policy.
Insanity is the best defence. |
|
| Back to top |
|
|
| Vitor |
| Posted: Fri Oct 21, 2011 8:51 am Post subject: |
|
|
Grand High Poobah
Joined: 11 Nov 2005
Posts: 26093
Location: Texas, USA
|
| Seshagiri wrote: |
| But you know there were 32 log files |
How could I or anyones else on this forum possibly know that? We can't see your screen or anything! 
_________________
Honesty is the best policy.
Insanity is the best defence. |
|
| Back to top |
|
|
| mvic |
| Posted: Fri Oct 21, 2011 8:56 am Post subject: |
|
|
Jedi
Joined: 09 Mar 2004
Posts: 2080
|
| Seshagiri wrote: |
| But you know there were 32 log files and when the application team made the messages as persistent we can see 37 log files. |
Is that a problem? It doesn't sound like a problem to me.
By the way, I don't know why the suggestion has been made in this thread to NOT use persistent messages.
It sounds to me like you need persistent messages, if you cannot operate over a more reliable network.
You might also like to check the setting for NPMSPEED on your channels, if you do need to continue to use Non-Persistent messages. |
|
| Back to top |
|
|
| Vitor |
| Posted: Fri Oct 21, 2011 8:59 am Post subject: |
|
|
Grand High Poobah
Joined: 11 Nov 2005
Posts: 26093
Location: Texas, USA
|
| mvic wrote: |
| You might also like to check the setting for NPMSPEED on your channels, if you do need to continue to use Non-Persistent messages. |
As I suggested earlier.
_________________
Honesty is the best policy.
Insanity is the best defence. |
|
| Back to top |
|
|
| exerk |
| Posted: Sat Oct 22, 2011 5:05 am Post subject: |
|
|
Jedi Council
Joined: 02 Nov 2006
Posts: 6339
|
| Seshagiri wrote: |
| But you know there were 32 log files and when the application team made the messages as persistent we can see 37 log files. More over we have defined 31 as primary and 32 as secondary logs in q.ini file. |
How long running are your Units-of-Work? If you have defined 32 primaries, and are now cutting secondaries (32 primary + 5 secondary = 37) I'd say you have a problem with sizing.
_________________
It's puzzling, I don't think I've ever seen anything quite like this before...and it's hard to soar like an eagle when you're surrounded by turkeys. |
|
| Back to top |
|
|
| bruce2359 |
| Posted: Sat Oct 22, 2011 6:04 am Post subject: |
|
|
Poobah
Joined: 05 Jan 2008
Posts: 9469
Location: US: west coast, almost. Otherwise, enroute.
|
Sizing of logs is well-documented in the WMQ System Administration manual, and the equivalent InfoCenter pages.
Effective sizing of logs requires knowledge of your application messages. Sizing involves basic math skills. Or, you can just guess, and hope things work out ok.
_________________
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 |
|
|
| PeterPotkay |
| Posted: Sat Oct 22, 2011 12:41 pm Post subject: |
|
|
Poobah
Joined: 15 May 2001
Posts: 7722
|
| exerk wrote: |
| Seshagiri wrote: |
| But you know there were 32 log files and when the application team made the messages as persistent we can see 37 log files. More over we have defined 31 as primary and 32 as secondary logs in q.ini file. |
How long running are your Units-of-Work? If you have defined 32 primaries, and are now cutting secondaries (32 primary + 5 secondary = 37) I'd say you have a problem with sizing. |
...although if you have a misbehaving app that never commits its unit of work, it doesn't matter how accurate your sizing was - no number or size of logs will be big enough in the end.
_________________
Peter Potkay
Keep Calm and MQ On |
|
| Back to top |
|
|
| exerk |
| Posted: Sat Oct 22, 2011 2:51 pm Post subject: |
|
|
Jedi Council
Joined: 02 Nov 2006
Posts: 6339
|
| PeterPotkay wrote: |
| exerk wrote: |
| Seshagiri wrote: |
| But you know there were 32 log files and when the application team made the messages as persistent we can see 37 log files. More over we have defined 31 as primary and 32 as secondary logs in q.ini file. |
How long running are your Units-of-Work? If you have defined 32 primaries, and are now cutting secondaries (32 primary + 5 secondary = 37) I'd say you have a problem with sizing. |
...although if you have a misbehaving app that never commits its unit of work, it doesn't matter how accurate your sizing was - no number or size of logs will be big enough in the end. |
Go linear...
_________________
It's puzzling, I don't think I've ever seen anything quite like this before...and it's hard to soar like an eagle when you're surrounded by turkeys. |
|
| Back to top |
|
|
| Vitor |
| Posted: Sat Oct 22, 2011 4:51 pm Post subject: |
|
|
Grand High Poobah
Joined: 11 Nov 2005
Posts: 26093
Location: Texas, USA
|
| exerk wrote: |
| Go linear... |
Which still won't help with an application that never commits. Sooner or later you hit the limit.
_________________
Honesty is the best policy.
Insanity is the best defence. |
|
| Back to top |
|
|
| exerk |
| Posted: Sun Oct 23, 2011 3:00 am Post subject: |
|
|
Jedi Council
Joined: 02 Nov 2006
Posts: 6339
|
| Vitor wrote: |
| exerk wrote: |
| Go linear... |
Which still won't help with an application that never commits. Sooner or later you hit the limit. |
I suspect you're going to hit queue full long before the chance of full log space...
_________________
It's puzzling, I don't think I've ever seen anything quite like this before...and it's hard to soar like an eagle when you're surrounded by turkeys. |
|
| Back to top |
|
|
| mvic |
| Posted: Sun Oct 23, 2011 9:44 am Post subject: |
|
|
Jedi
Joined: 09 Mar 2004
Posts: 2080
|
| exerk wrote: |
| I suspect you're going to hit queue full long before the chance of full log space... |
Choosing linear or circular does not change the log space calculations.
Also, you cannot rely on a Q-full condition to protect you from a log-full condition. They happen independently of each other, for independent reasons. |
|
| Back to top |
|
|
| exerk |
| Posted: Sun Oct 23, 2011 9:59 am Post subject: |
|
|
Jedi Council
Joined: 02 Nov 2006
Posts: 6339
|
| mvic wrote: |
| exerk wrote: |
| I suspect you're going to hit queue full long before the chance of full log space... |
Choosing linear or circular does not change the log space calculations.
Also, you cannot rely on a Q-full condition to protect you from a log-full condition. They happen independently of each other, for independent reasons. |
My statement was merely the continuing philosophical discussion. I don't rely on a queue full condition to protect me from anything! After all, if an application never commits there are not going to be any messages anyway 
_________________
It's puzzling, I don't think I've ever seen anything quite like this before...and it's hard to soar like an eagle when you're surrounded by turkeys. |
|
| Back to top |
|
|
| Vitor |
| Posted: Mon Oct 24, 2011 4:38 am Post subject: |
|
|
Grand High Poobah
Joined: 11 Nov 2005
Posts: 26093
Location: Texas, USA
|
[quote="exerk"]
| mvic wrote: |
| My statement was merely the continuing philosophical discussion. I don't rely on a queue full condition to protect me from anything! After all, if an application never commits there are not going to be any messages anyway |
If you have 200,000 100mb uncommitted persistent messages in a queue then all those non-existent are going to fill your queue file quite effectively.
If they've all been put in a single unit of work on a queue manager using the default number of logs then it'll be out of log space well before that. Even with linear logging.
_________________
Honesty is the best policy.
Insanity is the best defence. |
|
| Back to top |
|
|
|
|
