qpid-users mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From Rob Godfrey <rob.j.godf...@gmail.com>
Subject Re: monitoring and alarms/alerts
Date Wed, 16 May 2012 20:14:16 GMT
There are folks who have written a small amount of code against our
JMX APIs to regularly query the info that is made available.

In general we probably have seen more people using the log file
information. Of the cases you mention, I don't think we have anything
that specifically covers the "empty queue" case (since the queue being
empty is for many use cases the "expected" state)... Also I'm not
quite sure what you are wanting to measure with "slowness".  The Java
Broker can be configured to log out inbound/outbound rates (msg and
bytes) at regular intervals... but those numbers can be low if the
clients are slow / have nothing to say...

-- Rob

On 16 May 2012 21:56, Gary Ogden <gogden@gmail.com> wrote:
> Sorry. I should have said I wish to use the Java broker.
>
> Did you write a custom JMX tool?
>
> On 16 May 2012 16:47, Rob Godfrey <rob.j.godfrey@gmail.com> wrote:
>
>> I'm not sure from your question whether you are talking of the Java or
>> C++ Broker.
>>
>> We use the Java Broker in-house and use a mixture of log scraping and
>> JMX.  The Java Broker's QMF support need a bit of an overhaul and
>> probably doesn't work with all of the QMF toolset at the moment.
>>
>> For the C++ Broker QMF is obviously an option, along with the logging,
>> as you mentioned.  I don't think the QMF-JMX bridge is maintained
>> anymore and I believe we no longer release it.
>>
>> -- Rob
>>
>> On 16 May 2012 20:29, oggie <gogden@gmail.com> wrote:
>> > While evaluating QPid, I also need to determine if it can be monitored
>> > easily. Our current monitoring tool is nagios.
>> >
>> > We need to be able to monitor the queue for things such as:
>> > 1 - empty queue
>> > 2 - full queue
>> > 3 - old messages
>> > 4 - slowness
>> > etc.
>> >
>> > As far as I can tell, there's three options:
>> > 1 - Add the startup option to have qpid write to the system log and then
>> > just get nagios to montor that log
>> > 2 - Use QMF and write a nagios plugin in python to use QMF.
>> > 3 - Write a jmx client plugin for nagios that uses the jmx interface in
>> > qpid.
>> >
>> > Does anyone have any experience with this? Anything they'd like to
>> share? Is
>> > one option better than the others?
>> >
>> > I think the #1 and #2 would be the quickest to implement.
>> >
>> > Thanks!
>> >
>> >
>> > --
>> > View this message in context:
>> http://qpid.2158936.n2.nabble.com/monitoring-and-alarms-alerts-tp7562412.html
>> > Sent from the Apache Qpid users mailing list archive at Nabble.com.
>> >
>> > ---------------------------------------------------------------------
>> > To unsubscribe, e-mail: users-unsubscribe@qpid.apache.org
>> > For additional commands, e-mail: users-help@qpid.apache.org
>> >
>>
>> ---------------------------------------------------------------------
>> To unsubscribe, e-mail: users-unsubscribe@qpid.apache.org
>> For additional commands, e-mail: users-help@qpid.apache.org
>>
>>

---------------------------------------------------------------------
To unsubscribe, e-mail: users-unsubscribe@qpid.apache.org
For additional commands, e-mail: users-help@qpid.apache.org


Mime
View raw message