ASG
IBM
Zystems
Cressida
Icon
Netflexity
 
  MQSeries.net
Search  Search       Tech Exchange      Education      Certifications      Library      Info Center      SupportPacs      LinkedIn  Search  Search                                                                   FAQ  FAQ   Usergroups  Usergroups
 
Register  ::  Log in Log in to check your private messages
 
RSS Feed - WebSphere MQ Support RSS Feed - Message Broker Support

MQSeries.net Forum Index » IBM MQ Performance Monitoring » High Queue Depth AM Report

Post new topic  Reply to topic
 High Queue Depth AM Report « View previous topic :: View next topic » 
Author Message
cicsprog
PostPosted: Wed Sep 21, 2005 6:46 am    Post subject: High Queue Depth AM Report Reply with quote

Partisan

Joined: 27 Jan 2002
Posts: 347

I admin 65+ z/OS MQM’s only. We have Omegamon XE but we are not doing any historical reporting (no money for centralized server to house this data, no mips, yada yada ). I would like to produce a consolidated (i.e. one report) morning report of QLOCALs exceeding a QUEUE DEPTH of 500. I am looking for ideas.

One hair-brained scheme is to put a step(s) in the nightly MQM backups to read all the QLOCALs in the MQM and create MQ messages for QUEUEs exceeding the depth boggy to send to a central queue for reporting. However, I don’t see anything in the MQ manuals for the MQI that allows sequential retrieval of QUEUE names to do a MQINQ on (looked at MQOPEN also). So only thought I have is to add a step with CSQUTIL to produce a list of QLOCALs as input to the program to interrogate QUEUE DEPTH. Or have CSQUTIL deliver QUEUE DEPTH and report from that file.

Any other ideas appreciated.
Back to top
View user's profile Send private message
EddieA
PostPosted: Wed Sep 21, 2005 8:20 am    Post subject: Reply with quote

Jedi

Joined: 28 Jun 2001
Posts: 2453
Location: Los Angeles

Quote:
I don’t see anything in the MQ manuals for the MQI that allows sequential retrieval of QUEUE names to do a MQINQ on

You can send a command to the Command Server, plain text for v5, PCF for v6, and get the list. It's a bit of a pain to parse the data returned for v5, but can be done.

Cheers,
_________________
Eddie Atherton
IBM Certified Solution Developer - WebSphere Message Broker V6.1
IBM Certified Solution Developer - WebSphere Message Broker V7.0
Back to top
View user's profile Send private message
jefflowrey
PostPosted: Wed Sep 21, 2005 8:56 am    Post subject: Reply with quote

Grand Poobah

Joined: 16 Oct 2002
Posts: 19981

Yeah, for v5 the best thing to do is send "dis ql(*) curdepth" to the command server.

That will return you the queue name and the current depth for all the queues on the qm.
_________________
I am *not* the model of the modern major general.
Back to top
View user's profile Send private message
RogerLacroix
PostPosted: Wed Sep 21, 2005 9:16 am    Post subject: Reply with quote

Jedi Knight

Joined: 15 May 2001
Posts: 3264
Location: London, ON Canada

Hi,

This isn't pretty but will work.

(1) On one queue manger (MQA1) create a local queue (MYSTATS.Q) with a large Max Q Depth (500,000).

(2) Now on every other queue manager create a remote queue pointing to MYSTATS.Q on MQA1 or create a queue manager alias pointing to MQA1.

(3) Now create a very simple program (COBOL/C/Rexx) to put a message to the SYSTEM.COMMAND.INPUT with the following message data:
Code:
display qlocal(*) curdepth

Set the MQMD fields
- Format to MQSTR
- MsgType to Request
- Reply-To-Q to MYQSTATS.Q
- Reply-To-QMgr to MQA1 (do not set if using remote queue).

At midnight, run this program against every single queue manager. Now you will get a message for every local queue from every queue manager sent to your stats queue.

Note: You get 2 extra messsages per command issued - a header message (CSQN205I) and a trailer message (CSQ9022I). You can discard these messages.

The only trick is how do you know where the message came from (which queue manager sent it to your stats queue). You have 2 places to check in the message's MQMD: Reply-To-QMgr field or the MsgID field. Both will have the originating queue manager name.

Hope that helps.

Regards,
Roger Lacroix
_________________
Capitalware: Transforming tomorrow into today.
Connected to MQ!
Twitter
Back to top
View user's profile Send private message Visit poster's website
fjb_saper
PostPosted: Wed Sep 21, 2005 3:31 pm    Post subject: Reply with quote

Grand High Poobah

Joined: 18 Nov 2003
Posts: 20756
Location: LI,NY

We do it twice a day through QPasa and distribute per email to the relevant user community.

Trigger is a crontab job that puts a message on our report trigger queue. It clears the queue after 1 min.

We have different rules and a total of over 70 queues are checked accross 3 different plattforms (AIX, Solaris, zOS). Our criteria is to enter the queue into the message with its depth if the curdepth is over 0. You could as well set it 500.

The filtering is done via a perl script. Of course all the queues checked are being monitored by QPasa.

Works like a charm.

Ask your Omegamon support to help you set it up under the motto if QPasa can do it, show me how to do it with Omegamon...

And by the way this is part of instant monitoring and has nothing to do with history... History reporting involves reports over a period of time not instant alerts...


Enjoy
Back to top
View user's profile Send private message Send e-mail
hopsala
PostPosted: Wed Sep 21, 2005 3:46 pm    Post subject: Reply with quote

Guardian

Joined: 24 Sep 2004
Posts: 960

I don't see how creating such a report every morning is a good solution, if you had a problem during the day you would only know about it the day after...

So, another suggestion, which I find slightly better, in steps:

1. Set performance events with a high and low threshold for every application queue.
2. If you wish to collect this auditing data somewhere else than your z/OS, Delete local SYSTEM.ADMIN.PERFM.EVENT queue and create a parallel remote queue that point to the qm you use to collect all data. Conversly write the prog in step 3 as client (a better solution I believe)
3. Write a small prog (or d/l sp) that reads from this queue and reports any infractions.

I always find a trigger-oriented solution (push technology) to be better than a sampliing solution (pull technology) for many undestandable reasons.
You can also trigger this application on FIRST on the PERFM queue, thus getting an online report whenever some queue reaches high depth, and possibly send an SNMP record to Omegamon (if it accepts them).

(p.s Just a note- what you're talking about is not historical reports, but online reports...)
Back to top
View user's profile Send private message
Display posts from previous:   
Post new topic  Reply to topic Page 1 of 1

MQSeries.net Forum Index » IBM MQ Performance Monitoring » High Queue Depth AM Report
Jump to:  



You cannot post new topics in this forum
You cannot reply to topics in this forum
You cannot edit your posts in this forum
You cannot delete your posts in this forum
You cannot vote in polls in this forum
Protected by Anti-Spam ACP
 
 


Theme by Dustin Baccetti
Powered by phpBB © 2001, 2002 phpBB Group

Copyright © MQSeries.net. All rights reserved.