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

Integration with OMI

jose-antonio_ra
Inactive

Hi all.

One of our clients needs to integrate dynatrace with OMI (HPE operation Manager)

Does anyone have experience or any reference on how to do it?


Thanks in advanced

Jose A

5 REPLIES 5

I think you should check few things. First check OMI API, there should be some option to send problems from dynatrace via webhook. Another option is how to determine which Dynatrace entity is which one in OMI. In such case you may need to add some tags to entities and then use it as a key in payload. I don't have experience with this tool, but this is how this is done for service now (for SN there is embedded integration, for OMI you should create one via webhook).

Sebastian

Hi Sebastian,

What do you mean with "In such case you may need to add some tags to entities and then use it as a key in payload."?

For example, we've seen severity entity on OMI is very different to Dynatrace. So, do you mean we should add this tag? In this case, how can we do that?


Thanks in advance.

Regards,

Elena.

For severity, you need to define how to handle it, since as you said OMI and DT don't work with the same level of Severity.

For example... all the Resources alarms as Critical, you need to map that.

In our case, the Webhook Payload didn't work since we need it to pass different levels of severity (an other things) so the problem information needs it to be "worked" before sending it to the destination. Unless that once into the destination, you can "work" the event that was sent (I think OMI can do that with a little work).

The way to use the Tags, is for a better Entity match with those from OMI, so you send the tags and in OMI you reads those tags and then match it with XX or YYY entity node/application.

ADPaoloni
Mentor

Hey, we are moving away from OMi but a few months ago i started to work to integrate for OMi, so it shows the dynatrace problems in the console.

It uses the opcmsg component of OMi, the python scripts execute in a collector with the OMi agent installed and gather the problems feed and pass it to the OMi console with a few variables. That's the way we started to doit, something like this (is a mockup):

#subprocess.call(["/path/to/command", "arg1", "arg2"])
#opcmsg
#severity=major
#application="key"
#object="entityName"
#print ('msg_text="Se alarmo el componente"' + entityName)
#msg_grp="Test"
#-option IDDYNA="id"
#-option STATUS=status
#-option TIPO_EVENTO=eventType
#-option SEVERITY="severityLevel"
#-option MASDATA="masdata"
#-option LOQUESEMECANTE="lo que quiera enviar"

We ended doing something similar but for the CA SOI platform.

elena_renedo1
Participant

Thanks for your answer, Dante.

But how can we define that? Could you send us any link with documentation about it? Where should we define this: on Dynatrace or on OMI?

Thanks,

Elena.