Showing results for 
Show  only  | Search instead for 
Did you mean: 

This product reached the end of support date on March 31, 2021.

How can we use measures/incidents to detect that an app server is hung?


Please advise how we can detect that an app server (JBoss and Tomcat) instance is hung, using dynatrace measures/incidents.

Our intent is to then link that measure violation (incident) to an action plugin (e.g. Generic Execution Plugin) to recycle that app server, in order to have a "self-healing" capability.

If possible, please be specific as to what measure(s) to use, and what threshold value to define.

BTW, we are on dynatrace 6.2

thanks! - mike d.



To clarify.. Websphere has a measure "thread declared hung count" which can be used to detect hung threads.. However I see no such corresponding measure for Tomcat (I reviewed all the JMX MBean metrics viewable via dynatrace as well). So is this even possible for Tomcat (or is this a fool's errand) ?

Dynatrace Guru
Dynatrace Guru

Hi Mike

We have some ongoing internal research towards detecting stuck threads as well as stuck JVMs. I just learned from the engineering team that for detecting stuck JVMs we will see some level of support in the upcoming 6.3 release. So - please stay tuned.

Sounds good! Thanks!

Great to hear!

Thanks for the update!