Skip Menu |

This queue is for tickets about the jmx4perl CPAN distribution.

Report information
The Basics
Id: 67772
Status: resolved
Priority: 0/
Queue: jmx4perl

People
Owner: roland [...] cpan.org
Requestors: work [...] paul.dubuc.org
Cc:
AdminCc:

Bug Information
Severity: Normal
Broken in: 0.90
Fixed in: (no value)



Subject: check_jmx4perl reports wrong error count
For some tests the plugin doesn't include them in the total failures. For example: Status Information: CRITICAL - 0 of 21 failed: [ 2] CRITICAL neomon_jms_queue_count: JMS Queue BillingErrorQueue Count : Threshold '5' failed for value 5098 [18] CRITICAL memory_gc_time_base: PS MarkSweep : 59073.21 GC ms/minute ... (all other tests are OK) my definition of neomon_jms_queue count uses the definition for jboss provided to set threshold values: <Check neomon_jms_queue_count> Use = jboss_jms_queue_count($0) Critical = ${1:10} Warning = ${2:5} </Check> Perl version: 5.10.0 SuSE 11.1 uname -a Linux mrldev312 2.6.32.24-0.2-default #1 SMP 2010-10-29 16:39:49 +0200 x86_64 x86_64 x86_64 GNU/Linux Is there something in the way I'm configuring the plugin that could be causing this? Some tests bump up the count and others don't. Another example: Status Information: CRITICAL - 1 of 21 failed: , [ 2] CRITICAL neomon_jms_queue_count: JMS Queue BillingErrorQueue Count : Threshold '5' failed for value 22825 [ 7] CRITICAL neomon_jms_queue_rate: JMS Queue BillingOfflineQueue : Message count rate -60.00 messages/minute [18] CRITICAL memory_gc_time_base: PS MarkSweep : 59807.31 GC ms/minute [10] WARNING memory_heap: Heap-Memory: 95.63% used (1.79 GB / 1.87 GB) ... I'm pretty sure it's the memory_heap test that is being counted here and the others are not. Thanks.
Hi,

Show quoted text
> <Check neomon_jms_queue_count> 
> Use = jboss_jms_queue_count($0)
> Critical = ${1:10}
> Warning = ${2:5}
> </Check>

Sorry, I cant reproduce this behavior yet. Could you send me the definition of the multi-check
as well where this check is contaied in ? One thing that 's a bit strange is, that the criticial
threshold is defined with '10' as default but fails wit a value of '5'. Do you provide the
concrete critical  value from the outside ?

...roland
Subject: Re: [rt.cpan.org #67772] check_jmx4perl reports wrong error count
Date: Wed, 27 Apr 2011 13:49:02 -0400
To: bug-jmx4perl [...] rt.cpan.org
From: "Paul M. Dubuc" <work [...] paul.dubuc.org>
Roland Huss via RT wrote: Show quoted text
> <URL: https://rt.cpan.org/Ticket/Display.html?id=67772> > > Hi, >
>> <Check neomon_jms_queue_count> >> Use = jboss_jms_queue_count($0) >> Critical = ${1:10} >> Warning = ${2:5} >> </Check>
> > Sorry, I cant reproduce this behavior yet. Could you send me the definition of > the multi-check > as well where this check is contaied in ? One thing that 's a bit strange is, > that the criticial > threshold is defined with '10' as default but fails wit a value of '5'. Do you > provide the > concrete critical value from the outside ? > > ...roland >
I'm sorry for not providing more information sooner. Yes, I was overriding the default thresholds in that one test. Here are the Multicheck definitions I was using. neomon_bss is the check that is being passed to the plugin: include memory.cfg include threads.cfg include jboss.cfg <MultiCheck neomon_bss> MultiCheck = neomon_jms_queues_bss MultiCheck = neomon_common </MultiCheck> <MultiCheck neomon_common> MultiCheck = neomon_memory MultiCheck = neomon_threads Check = jboss_cpool_available("DefaultDS") </Multicheck> <MultiCheck neomon_memory> Check = memory_heap(98,95) Check = memory_pool_base("PS Eden Space",100,100) Check = memory_pool_base("PS Survivor Space",100,100) Check = memory_pool_base("PS Old Gen") Check = memory_pool_base("PS Perm Gen") Check = memory_gc_count_base("PS Scavenge") Check = memory_gc_count_base("PS MarkSweep") Check = memory_gc_time_base("PS Scavenge") Check = memory_gc_time_base("PS MarkSweep") </MultiCheck> <MultiCheck neomon_threads> Check = thread_count Check = thread_deadlock </MultiCheck> <Check neomon_jms_queue_count> Use = jboss_jms_queue_count($0) Critical = ${1:10} Warning = ${2:5} </Check> <Check neomon_jms_queue_rate> Use = jboss_jms_queue_rate($0) Critical = ${1:200} Warning = ${2:100} Label = JMS Queue $0 : Message count rate %2.2f messages/minute </Check> <MultiCheck neomon_jms_queues_bss> Check = neomon_jms_queue_count("LogoffQueue") Check = neomon_jms_queue_count("BillingErrorQueue",5,1) Check = neomon_jms_queue_count("EventErrorQueue",5,1) Check = neomon_jms_queue_count("DLQ",5,1) Check = neomon_jms_queue_count("ExpiryQueue",5,1) Check = neomon_jms_queue_count("BillingOfflineQueue",1200,1000) Check = neomon_jms_queue_rate("BillingOfflineQueue") Check = neomon_jms_queue_count("EventOfflineQueue",1200,1000) Check = neomon_jms_queue_rate("EventOfflineQueue") </MultiCheck>
Thank you for the details, I will try to reproduce the issue 
this weekend and provide a fix for 0.91
I found the issue and fixed it for 0.91 (scheduled for this weekend). It affected
all non-relative checks within multi checks which simply were forgotten to be counted.

Thanks for spotting this.
Subject: Re: [rt.cpan.org #67772] check_jmx4perl reports wrong error count
Date: Fri, 06 May 2011 13:17:29 -0400
To: bug-jmx4perl [...] rt.cpan.org
From: "Paul M. Dubuc" <work [...] paul.dubuc.org>
Roland Huss via RT wrote: Show quoted text
> <URL: https://rt.cpan.org/Ticket/Display.html?id=67772> > > I found the issue and fixed it for 0.91 (scheduled for this weekend). It > affected > all non-relative checks within multi checks which simply were forgotten to be > counted. > > Thanks for spotting this. >
Thank you very much for fixing this and the others. You have a great package here! Much good work and thought went into it.
On Fri May 06 13:17:40 2011, PMDubuc wrote:

Show quoted text
> Thank you very much for fixing this and the others. You have a great
> package
> here! Much good work and thought went into it.

And thank you for using jmx4perl ;-). I just released 0.91 final and will close 
the open tickets at the end of next week except you find an issue 
with the fixes.

bye, nice weekend ...
... roland

 finally closed.