| |
|
RSS Feed - WebSphere MQ Support
|
RSS Feed - Message Broker Support
|
  |
|
| PING a retrying channel on Mainframe ... |
« View previous topic :: View next topic » |
| Author |
Message
|
| tkaravind |
 Posted: Sat May 01, 2004 6:16 pm Post subject: PING a retrying channel on Mainframe ... |
 |
|
Acolyte
Joined: 24 Jul 2001
Posts: 64
|
Dear All,
We are facing this strange situation on an MQSeries sender channel on mainframe.
There is an automated PING activity on this channel to healthcheck.
The moment this starts the channel seems to be giving up its RETRY operation ! (The logs do not have the RETRYs recorded any more after the time the PING starts - and later, even when the remote end comes up the channel does not connect automatically and has to be manually started).
Is this to be expected ? Can PING be so messy ?
Can anybody throw light on this ?
Thanks & Regards,
Aravind |
|
| Back to top |
|
 |
| mqonnet |
| Posted: Wed May 05, 2004 8:55 am Post subject: |
|
|
Grand Master
Joined: 18 Feb 2002
Posts: 1114
Location: Boston, Ma, Usa.
|
You should gather more evidence as to what happened. I doubt all this mess would have been because of the ping channel(from your post this is what i gathered, i mean the command used) command.
When a ping channel is issued it uses the same channel defs in a simulated channel start. So all the resources used in establishing a successful connection of a channel are used when you issue ping as well.
So, that would mean, if you have a ping channel issued while a channel is in retry, the resources, xmitq, channels at both ends(if we can infact communicate with remote end), are busy because of the ping command. And that would in turn mean that at this point you *cannot* issue a start channel, be it automatic or manual. And you would get channel in use error, if you do try it out.
But yes, once the ping comes back, either successful or unsuccessful, you could reissue start channel command. If all is well, channel would come up.
So, IMHO, the problem that you described doesnt look like it has any bearing with the Ping channel command. It could just be a co-incidence that you used this command while you had a channel in retry state. May be to find out the state.
Look in the error logs and corelate the sequence of events.
BTW, is this a repeatable scenario or happened this one time. If repeatable, can you post the steps followed to do so.
Cheers
Kumar |
|
| Back to top |
|
|
| tkaravind |
| Posted: Thu May 06, 2004 11:31 pm Post subject: |
|
|
Acolyte
Joined: 24 Jul 2001
Posts: 64
|
Hi Kumar,
This scenario is definitely repeating everytime we bring the receiver box (AIX) down i.e whenever the sender channel gets into a retrying state
and when the receiver box is brought up AFTER the PING schedule starts.
To further ascertain things we tried to bring up the AIX server before the PING activity starts and this time the channel connected automatically i.e RETRY was successful.
That is the main reason why we suspect the PING and in the next test we would stop PING altogether and let the channel connect by itself.
Although all this does not make sense but from the evidence this is what seems to be the case.
We have already raised a query with IBM but they are yet to get back on this.
Please let me know what you feel about this.
Regards,
Aravind |
|
| Back to top |
|
|
| mqonnet |
| Posted: Mon May 10, 2004 8:18 am Post subject: |
|
|
Grand Master
Joined: 18 Feb 2002
Posts: 1114
Location: Boston, Ma, Usa.
|
Could all this just be a timing related thing. You have to remember that channels when in retry would only try to reconnect depending on the value set in hbint. And it could be very much possible that the "one" time when you started the queue manager, this channel went into running state instead of hanging around in the retry state.
Also, i did not ask this earlier. Did you get any error messages in the error logs. Logs could throw some more light.
Cheers
Kumar |
|
| Back to top |
|
|
| PeterPotkay |
| Posted: Mon May 10, 2004 4:13 pm Post subject: |
|
|
Poobah
Joined: 15 May 2001
Posts: 7722
|
| Quote: |
You have to remember that channels when in retry would only try to reconnect depending on the value set in hbint.
|
Not quite true. The channel will retry when the Retry Interval (either short or if expired, then long) goes by. The HBInterval plays no part in a channel that is retrying already, although it certainly plays a part in getting a channel to start retrying.
_________________
Peter Potkay
Keep Calm and MQ On |
|
| Back to top |
|
|
| tkaravind |
| Posted: Tue May 11, 2004 12:04 am Post subject: |
|
|
Acolyte
Joined: 24 Jul 2001
Posts: 64
|
All,
Yes. I did check the mainframe (OS390) MQSeries logs. This happens to be the Sender side of the channel.
Curiously enough till the PING messages started showing , the log was full of RETRY messages getting recorded every 20 mins (which is the Long retry interval set) ...But they simply disappeared after the PING messages started coming up !!!
This was infact what led us to suspect the PING activity - that it was interfering with the normal functioning of the channel which should be simply to retry until it is able to connect. (The LONG retry count itself seems to have been set at its default which is 9 times the number 9 , a very large number. So no hiccups there too)
Regards,
Aravind |
|
| Back to top |
|
|
| mqonnet |
| Posted: Tue May 11, 2004 5:10 am Post subject: |
|
|
Grand Master
Joined: 18 Feb 2002
Posts: 1114
Location: Boston, Ma, Usa.
|
Peter, thanks for correcting me there. I meant retry(short/long) as you pointed out, and instead wrote down hbint. :).
"This was infact what led us to suspect the PING activity - that it was interfering with the normal functioning of the channel which should be simply to retry until it is able to connect"
---As i mentioned earlier. If you issue ping channel command, then ping is using all the resources and hence the sdr channel wont start at that time. Only after the ping stops, sdr channel can again go ahead and retry.
From what all you posted here, i dont gather enough evidence that ping is causing all this. It could be, but more info is needed.
You could either have ping channel or sdr channel running at one point in time.
Cheers
Kumar
_________________
IBM Certified WebSphere MQ V5.3 Developer
IBM Certified WebSphere MQ V5.3 Solution Designer
IBM Certified WebSphere MQ V5.3 System Administrator |
|
| Back to top |
|
|
| tkaravind |
| Posted: Wed May 12, 2004 5:34 am Post subject: |
|
|
Acolyte
Joined: 24 Jul 2001
Posts: 64
|
Hi,
That is the whole point. I can understand that a running channel cannot be pinged. But why should a channel in RETRY mode stop retrying after the first ping ???
The PING happens every 15 mins by some automated mechanism - so between the first and the next PING there is a big gap of 15 mins by which time atleast one retry should go through. This is what never seems to happen.
Regards,
Aravind |
|
| Back to top |
|
|
|
|
|
| |
|
Page 1 of 1 |
|
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
|
|
|
|
