-
Notifications
You must be signed in to change notification settings - Fork 2.3k
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
[synthetic-monitoring-agent] fix deployment not starting on update/auto-scaling #3070
base: main
Are you sure you want to change the base?
Conversation
Signed-off-by: Iridias <[email protected]>
Signed-off-by: Iridias <[email protected]>
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
@Iridias Could you please resolve the Conflicts
Signed-off-by: MH <[email protected]>
Signed-off-by: MH <[email protected]>
@zanhsieh what do you think about this ? I mean there are other charts as well which are using hpa, how are they resolving this ? |
I'd assume, they don't have to solve this - and that the apps, they're for, are HA-capable. That's still true with the current version 0.31.0 |
Retry #2994
The implementation of the agent does not seem to allow for concurrency.
Thus scaling the deployment - either on helm-upgrade or auto-scaling - will result in the new PODs to never become ready!
So any helm-upgrade will run into a timeout and then abort.
In the logs you'll find the following messages (if debug is enabled):
With emphasis on:
response: probe already exists
To fix that, I changed the
Deployment
to aStatefulSet
, as k8s ensures, that the old POD is killed/deleted before spawning the new one.I also removed all the autoscaling-resources, as they're not useful anyway.
And of course, I also successfully tested the changes on one of our clusters.