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

Windows Sevice Availability

mrc15816
Helper

Hello All,

We are very excited about the Windows Service Availability intro in 1.196.

Little about my current situation. Our current monitoring tool provides alerts when 'Window Services' Startup type is set to 'Automatic' and the service is not 'Running'. Dynatrace 1.196 release doesn't offer this at the moment, but once a service is created it does alert - would be a pain to define each and every COTS product services.

We also would like to know if Dynatrace would allow to use tags in alerting message in that way we can define support group - example below

Splunk service running as a Windows service and any issues with service will go to Splunk Support group, like that we have quite a few different services that are supported by different group.

Any suggestion anyone?

Thank you in advance!

Raj

4 REPLIES 4

ChadTurner
Leader

You could always leverage the metadata at a host level and define out certain things that you feel would be useful for splunk. For example, In your metadata you could define out Application = Easy Travel, Owner/Team=John's team or Easy Travel team, Location= City/state or On Prem/Cloud,.. and so on. With that metadata you could then create automatic tagging rules that trickle down from the host, to the process, to the services and then supply the tags of each entities in your alert payload which splunk could then parse out/look for.

As for the 196 release, I have not worked with that yet but I will update you with any additional solutions that I may find along the way.

-Chad

mrc15816
Helper

@Chad T. We are already using hostcustomproperties.conf and autohost.conf files. We have also configured Autotags at the host level and process level, but the Windows Service alerting wouldn't touch the process level and during the incident assignment it would be assigned to the wrong group. We are now looking at another alternative by hard coding the groups to the Windows Service message and use that value to assign the right group. Thank you for your suggestion.

ct_27
Organizer

We are currently transitioning off of our current monitoring solution (OpsView). It too has the ability to automatically monitor all Windows Services that are set to 'automatic'. This is a fantastic feature because any Service set to run automatically is important enough to ensure 1) IT starts on system startup 2) remains running.

It would be completely impractical to have to continuously update some list of Services to monitor because it's an async human interaction that is highly likely to breakdown.

Could Dyntrace consider adding this ability. To automatically monitor the availability of all windows Services where 'Startup Type' = 'automatic'. Generate an Event that we can alert on should the process not be running.' Not be running', includes the scenario that it does not start on system startup, if it at any time its Status != 'Running' because it crashed or was manually stopped. If the 'Startup Type' were to be changed to any value other than Automatic then DT does not need to alert on it Status any longer.

jimge
Observer

ct_27 with another insight nugget. We are on pace to transistion off of SolarWinds to DT by 2032 at current pace. This would save us about 8 years.