-
Notifications
You must be signed in to change notification settings - Fork 58
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
Maintain datadog monitors via code #1673
Comments
Just a note: you're using the word alarm, which has two definitions: Application alarm that allows the application to be reactive. Metric alarm. |
My intention is [alarm_word_placeholder] that would notify engineer to react on. I think datadog use the word monitor. Just I am used to use the word alarm previously 😅 |
Just to expand Ino's comment a bit. The in-app alarms right now are mostly the ones that signals that the service is not in good shape to accept new transactions, e.g. The only one that seems out of place is |
If we want to simplify a bit and standardize everything, we can make those But....might not bring much benefit as we already have those code in place and is working fine. Just some burden to be able to see what are all the monitors we have. |
Ah I see. I think we can send an But I concur that it's not high priority given other things we have in hand at the moment. |
lol my personally feeling is nobody would take this up at all hahaha unless gold team board become cleaner 😅 But can we pick this up in new services like chch v2 to be a standard approach for new service ?! @InoMurko @achiurizo |
Why
We are having more and more alarms/metric/monitors in data dog now. Currently it is a messy manual process. Also, there are 2 possible ways to trigger an alarm. Could be via application code directly or via data dog monitor.
We have some alarm that is directly triggered in application code. However, it would be great if all alarm can be just in data dog and the application only need to emit metric. Currently for most cases we are emitting both metric and alarm via application code.
But even not doing the simplification, using some code as infra to maintain our current messy monitors might be already beneficial enough.
Note
One potential tool is terraform: https://www.terraform.io/docs/providers/datadog/r/monitor.html
Seems like supporting datadog too.
The text was updated successfully, but these errors were encountered: