[Netarchivesuite-users] The message limit (maxNumMsgs) of 100000 has been reached

Bjarne Andersen bja at statsbiblioteket.dk
Wed Sep 16 09:23:18 CEST 2009


The current latest stable version (3.8.2) handles some auto-reconnection to the JMS-broker in case of lost connection - but the next stable version should handle such error scenarios much more smoothly. So even a restart of the JMS-broker would make applications auto-reconnect - a scenario that the current version does not fully or always support.

I could say that in Netarchive.dk we have set up a number of very small cronjobs to check different aspects of the JMS-queues
 - number of listeners on different queues (e.g. number of bitarchive-machines listening, number of listeners on HIGH_PRIORITY harvester queues (must not be 0) - etc. )
 - number of messages in different queues (e.g. number of messages on the MONITOR_QUEUE should not exeed 5000 messages)

The script counts the expected number of different things from the deploy-xml-file so that we dynamically always e.g. check for the right number of expected bitarchive machines.

If you like a copy of this script I can mail it to you.

-
Bjarne
________________________________________
Fra: netarchivesuite-users-bounces at lists.gforge.statsbiblioteket.dk [netarchivesuite-users-bounces at lists.gforge.statsbiblioteket.dk] På vegne af aponb at gmx.at [aponb at gmx.at]
Sendt: 16. september 2009 08:49
Til: NetarchiveSuite-users
Emne: Re: [Netarchivesuite-users] The message limit (maxNumMsgs) of     100000 has been reached

Thanks for your reply!
I did restart the broker. After that the messages in queue
ONB_COMMON_MONITOR was empty.
After that I had to restart all Harvester-Instances because they didn't
get new scheduled jobs.
What I tried before restarting the broker was to inspect the queue, but
without success. With which tool are you doing this? imqcmd doesn't
offer this possibilty. Probably this has to be done with an own written
code. The question for me is now, is it normal to get to limit of the
maximal number of message in the Monitor Queue and should it be a normal
process to delete this queue?

Thanks for your help
a.

> Hi
>
> The people who know most about the JMS broker are not here today.
> I don't know what the 'right' way of solving this problem, but the easy solution would be restarting the JMS broker.
>
>
> Best regards
> Jonas
>
> -----Oprindelig meddelelse-----
> Fra: netarchivesuite-users-bounces at lists.gforge.statsbiblioteket.dk <https://lists.gforge.statsbiblioteket.dk/mailman/listinfo/netarchivesuite-users> [mailto:netarchivesuite-users-bounces at lists.gforge.statsbiblioteket.dk <https://lists.gforge.statsbiblioteket.dk/mailman/listinfo/netarchivesuite-users>] På vegne af aponb at gmx.at <https://lists.gforge.statsbiblioteket.dk/mailman/listinfo/netarchivesuite-users>
> Sendt: 15. september 2009 11:46
> Til: NetarchiveSuite-users
> Emne: [Netarchivesuite-users] The message limit (maxNumMsgs) of 100000 has been reached
>
> I've got the warning message in my JMS-Log and no more jobs will be
> scheduled.
> Bug 1275 reports also this error and the last follow up says "Making
> sure a queue doesn't fill up is a different matter, and
> has to do with monitoring the JMS broker". Till now I was convinced that
> messages will be removed if not needed anymore. Obviously this should be
> part of the monitoring of the JMS-Broker. Do you have any recommendation
> when to delete which messages?
> I assume after deleting a certain number of messages the system will
> accept new messages and will continue to schedule new jobs.
> Please let me know about any things I can do!
>
> Thanks in advance
> a.

_______________________________________________
NetarchiveSuite-users mailing list
NetarchiveSuite-users at lists.gforge.statsbiblioteket.dk
https://lists.gforge.statsbiblioteket.dk/mailman/listinfo/netarchivesuite-users




More information about the NetarchiveSuite-users mailing list