Expand my Community achievements bar.

Don’t miss the AEM Skill Exchange in SF on Nov 14—hear from industry leaders, learn best practices, and enhance your AEM strategy with practical tips.
SOLVED

how to access jmx console(s) in a production cloud env?

Avatar

Level 9

my understanding of JMX is that its per VM.  With AEM cloud, they will scale up VMs and servers as needed.  In this case, how does one access and/or aggregate the jmx info?  As Adobe dont offer any sort of monitoring / alerting solution, being able to query the number of errors, or the average time for API responses via jmx would be a good partial solution.

 

The jmx console seems to be available on local dev instances, but not on any. cloud instances (gives 403)

 

1 Accepted Solution

Avatar

Correct answer by
Community Advisor

New Relic One has following features, which might help you and closer to your requirements:

  • Direct access to a dedicated New Relic One account

  • Instrumented New Relic One APM agent that shows exact method calls with line numbers, including external dependencies and databases

  • Holistic performance optimization by combining key metrics from infrastructure-level monitoring and application (Adobe Experience Manager) monitoring

  • Exposure of AEM as a Cloud Service JMX Mbeans and health checks directly within New Relic Insights metrics, allowing for deep inspection of application stack performance and health metrics.

 

References: https://experienceleague.adobe.com/docs/experience-manager-cloud-service/content/implementing/using-...

 


Aanchal Sikka

View solution in original post

9 Replies

Avatar

Employee Advisor

 

The console are not available, that's right. What information do you need/want to monitor?

Avatar

Level 9
We want to monitor servlet response times, number of calls per minute, number of errors, and the same for calls which our servlets make to 3rd party apis. Without this, we are blind to problems or slowdowns.

Avatar

Correct answer by
Community Advisor

New Relic One has following features, which might help you and closer to your requirements:

  • Direct access to a dedicated New Relic One account

  • Instrumented New Relic One APM agent that shows exact method calls with line numbers, including external dependencies and databases

  • Holistic performance optimization by combining key metrics from infrastructure-level monitoring and application (Adobe Experience Manager) monitoring

  • Exposure of AEM as a Cloud Service JMX Mbeans and health checks directly within New Relic Insights metrics, allowing for deep inspection of application stack performance and health metrics.

 

References: https://experienceleague.adobe.com/docs/experience-manager-cloud-service/content/implementing/using-...

 


Aanchal Sikka

Avatar

Employee Advisor
That is a very good usecase for it, although I would question that you need direct access to the JMX console for it. You rather an interface for a monitoring system, which can pick up data created by this integration. Can you create a feature request for it, at best here in the Forums and as well as via Support.

Avatar

Employee Advisor
Besides that the only way I can imagine is to create a servlet which you can query from an external system and which provides that data (in whatever format). I agree that there should be a more standardized way to achieve this.

Avatar

Level 9
Im thinking of integrating either New Relic Event + log API, or Application Insights (aka Azure monitoring). When we signed up for AEM, the Adobe sales and technical team told us we would get New Relic for server side and custom event monitoring and alerting, then once we paid, they told us it had been withdrawn. Not good.

Avatar

Employee Advisor
I cannot comment on NewRelic, but I would suggest you to push that topic to get a definitive answer.

Avatar

Level 9
is system/console/configMgr also blocked for Cloud? If so, how does create or edit osgi configurations?

Avatar

Level 2

Is this an adobe hosted cloud environment, or self hosted AWS/Azure environment?  The JMX console will be most likely secured and inaccessible by design.  I would deploy the prod packages to a lower environment and try connecting there.