Skip to content
This repository was archived by the owner on Nov 15, 2019. It is now read-only.

About how to proceed with alerts #99

Open
aljesusg opened this issue Nov 6, 2017 · 7 comments
Open

About how to proceed with alerts #99

aljesusg opened this issue Nov 6, 2017 · 7 comments

Comments

@aljesusg
Copy link
Member

aljesusg commented Nov 6, 2017

Now when you create a alert of a type in Miq, hawkular is not creating the trigger for each Datasource. I have little knowledge about hawkular and I didn't know this. So here is the problem and some proposal.

Datasource

MiQ is looking for
MI~R~[0b84c5546795/Local DMR~~]~MT~Datasource Pool Metrics~In Use Count

Hawkular is providing metrics in
MI~R~[92da52a9b841/Local DMR~/subsystem=datasources/data-source=ExampleDS]~MT~Datasource Pool Metrics~In Use Count

/subsystem=datasources/data-source=ExampleDS

So If yo wanna trigger the alert you need to push data

client.metrics.gauges.push_data("MI~R~[0b84c5546795/Local DMR~~]~MT~Datasource Pool Metrics~In Use Count",{:value=>37})

Mesagging

MiQ is looking
MI~R~[0b84c5546795/Local~~]~MT~JMS Topic Metrics~Subscription Count

Hawkular is providing in
MI~R~[92da52a9b841/Local~/subsystem=messaging-activemq/server=default/jms-topic=HawkularAlertsActionsTopic]~MT~JMS Topic Metrics~Subscription Count"

/subsystem=messaging-activemq/server=default/jms-topic=HawkularAlertsActionsTopic

With Web Sessions there isn't any problem.

Problems

  • MiQ provider is not creating the trigger for each Datasource/Mesagging type in MWServer

Proposal Solutions

  1. fix de manageiq-providers-hawkular to create the trigger for each DataSource in the MiddlewareServer selected
  2. As @lucasponce told us in Fix problem getting live metrics when trigger is created with Datasource and Mesagging types (bugzilla 1505343) #88 make changes in MiQ UI to select Datasource type, Mesagging...

cc @abonas , @mtho11, @jdoyleoss

@abonas
Copy link
Member

abonas commented Nov 6, 2017

@jdoyleoss I think this is a functional decision first. so please see above :)

@jshaughn
Copy link
Contributor

The comments in #88 are correct, and the options above make sense. My question is whether this is going to be resolved for TP3. Because if not then we should wait for Prometheus migration and then proceed.

As for the approach, I'd probably suggest option 2, being able to assign to specific DS or Messaging entities. The alternative could result in a lot of unnecessary triggers and possibly unwanted alerts.

@aljesusg
Copy link
Member Author

I am working in the second option I think is the best too.

@jdoyleoss
Copy link

jdoyleoss commented Nov 10, 2017 via email

@abonas
Copy link
Member

abonas commented Nov 15, 2017

I agree with the second option. this is NOT for gaprindashvili though.

@abonas
Copy link
Member

abonas commented Nov 15, 2017

@aljesusg / @jdoyleoss please log an appropriate JMAN4 as this is a functional enhancement

@miq-bot
Copy link
Member

miq-bot commented May 21, 2018

This issue has been automatically marked as stale because it has not been updated for at least 6 months.

If you can still reproduce this issue on the current release or on master, please reply with all of the information you have about it in order to keep the issue open.

Thank you for all your contributions!

Sign up for free to subscribe to this conversation on GitHub. Already have an account? Sign in.
Projects
None yet
Development

No branches or pull requests

5 participants