topic Re: How to create Health rule for WebSphere MQ ??? in Splunk AppDynamics

How to create Health rule for WebSphere MQ ???

CommunityUser — Tue, 21 Aug 2018 20:01:12 GMT

Team,

Below are the requirements for my MQ, please guide!

1. I am trying to create health rule for more than 100 channels , whenever the status of any of these channels is not equal to 3 i want to be alerted . Can you let me know if there's any quick way to add the channels together in one condition??

2. For the queues, i would like to partion them into local queues, transmitting queues and dead queues based on the naming from the queues and set alerts for the same.

Thanks,

Teja.

Re: How to create Health rule for WebSphere MQ ???

Yogesh_Chouk — Mon, 27 Aug 2018 04:12:16 GMT

Hi Teja,

You can try using a wildcard as specified in the below documents.

https://docs.appdynamics.com/display/PRO45/Using+Wildcards+in+Metric+Definitions

However, you cannot specify condition such as "not equal to 3". You may need to specify 2 conditions to achieve this scenario

1. Value<4 AND

2. Value>2

Let me know if that helps.

Thanks,

Yogesh

Re: How to create Health rule for WebSphere MQ ???

CommunityUser — Tue, 28 Aug 2018 20:04:16 GMT

Hi Yogesh,

It isn't working as expected with wildcards. Is there any other way in which we can setup alerts?

Thanks,

Teja!

Re: How to create Health rule for WebSphere MQ ???

Yogesh_Chouk — Wed, 29 Aug 2018 05:21:26 GMT

Hi Teja,

With multiple metrics, you need to get this work with wildcards only. There is no other easy way. What issue are you seeing using the wildcard? I request you to share the screenshots of how you are configuring the health rules.

Thanks,

Yogesh

Re: How to create Health rule for WebSphere MQ ???

CommunityUser — Wed, 29 Aug 2018 17:30:23 GMT

Hi Yogesh,

Please find the below steps that i am using, i can't share the screenshot considering the confidentiality.

Overview - Overall Application performance, Affects ,Critical Condition - Specify A Relative Metric path, which i have copied from the Metric browser which looks like below:

After this i specify the critical condition as value <2 and value >2 and any condition match and saved the health rule.

I don't see the Current Evaluation Status are showing the "?"

This is working when i enter the below metric path and i don't think it's feasible to select each MQ status manually.

This alert is for MQ availability. The channel also is similar to this with different alert condition.

Thanks,

Teja.

Re: How to create Health rule for WebSphere MQ ???

Yogesh_Chouk — Wed, 29 Aug 2018 17:48:24 GMT

Hi Teja,

We see that the metric path provided is 'Application Performance Infrastructure|MCAG...'. Can you try to change it starting from Custom Metrics like below and let me know if that helps?

Custom Metrics|WebsphereMQ|GTLOD01|Status

If that does not work, try changing the Category in overview section to Tier/Node health...

For more info, please follow

https://docs.appdynamics.com/display/PRO44/Specify+Metrics+in+Multiple+Entities+Using+a+Wildcard#SpecifyMetricsinMultipleEntitiesUsingaWildcard-DefinetheRelativeMetricPath

Thanks,

Yogesh

Re: How to create Health rule for WebSphere MQ ???

CommunityUser — Wed, 29 Aug 2018 19:14:53 GMT

Hi Yogesh,

The below works but we are not able to add multiple MQ instances

Custom Metrics|WebsphereMQ|GTLOD01|Status - Working

Custom Metrics|WebsphereMQ|GTLOD01|Status - We want to add multiple GTLOD[1,2,3,4]

Custom Metrics|WebsphereMQ|*|Status - Not Working

On changing to Category in overview section to Tier/Node health and following the instructions as per the link shared below. We are still not able to see anything on current evaluation status

Thanks,

Teja

Re: How to create Health rule for WebSphere MQ ???

Yogesh_Chouk — Thu, 30 Aug 2018 04:29:10 GMT

Hi Teja,

What is the controller version you are using? We can schedule a webex session to take this further.

Thanks,

Yogesh

Re: How to create Health rule for WebSphere MQ ???

Ashley_Lewis1 — Mon, 27 Jan 2020 13:09:01 GMT

Hi Yogesh,

Were you able to figure this out.We have a similar requirement to apply a global threshold on all uncommitted messages.

I have created a rule to evaluate this based on baseline standard deviation and the following relative metric path. Not sure if it will work yet though