Intermittent Monitoring

[elisag]

Hi,

In my environment dynatrace managed only for two AIX hosts I have infrastructure monitoring with holes as shown in the image.

Active gate and hosts are on the same network.

We don't know how we have this beahaviour.

That situation don't help to monitoring the hosts.

what could it be?

Thank you a lot

Elisa GAiti

elisa.gaiti@alten.it

[pahofmann]

I assume the hosts are running normally though? 

It could be resource congestion on the host, so that the OneAgent can't capture properly, or a network Issue so the data doesn't reach the cluster. 

Id seems to happen pretty often, so I'd check the Ressource consumption and logs when there is no data in dynatrace. 

If that doesn't help open a support ticket. 

[Mizső]

Hi @elisag,

To be honest I do not have AIX host monitoring in my managed environments. I have met similar patterns when the hosts have performance issues/peaks. But in my case oneagent was on the hosts.

If I were you, I would raise a support ticket about it.

I hope anyone else with activegate based AIX monitoring experience in the community will have a better idea for you.

As I see @y_buccellato  has an AIX monitoring experience maybe he could help to you.

Solved: AIX host monitoring - CPU Usage - problem detection - Dynatrace Community

Have a nince day.

Best regards,

Mizső 

[y_buccellato · Accepted solution]

Ciao Elisa,

here cause @Mizső made my name 🙂 We do monitor a wide variety of AIX system and I never had such a issue (metrics coming and going).

If I'd start fresh on the subject (not knowing what was done on Dynatrace before I was asked to install Oneagent on vm's) I would first check the fundamentals:

• on which architecture the AIX is running? (in my experience people that are running AIX moreoften are using old power architecture formally not supported by Dynatrace)
• on which AIX version did I install Oneagent?
• did I made a clean installation to the last version available or was it just an update? (in the case of update I'd simply try to stop and start or removing completly the old agent and go for reinstallation)

Once I did that, if problem persist, I'd verify health status of the agent via Oneagent Diagnostic:

In the end I'd check network routing of your agent to your AGs and cluster node ip and fqdn.
At the really end, if problem persist, I'd open a ticket like many other suggested,

hope you are able to solve this soon,

Yann