Skip to content
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

Ability to use intra.broker.goals in anomaly detection / self healing #1264

Open
amuraru opened this issue Jul 14, 2020 · 5 comments · May be fixed by #1721 or adobe/cruise-control#6
Open

Ability to use intra.broker.goals in anomaly detection / self healing #1264

amuraru opened this issue Jul 14, 2020 · 5 comments · May be fixed by #1721 or adobe/cruise-control#6
Labels
functionality A feature request. good first issue A good fit as a first issue in the project.

Comments

@amuraru
Copy link
Contributor

amuraru commented Jul 14, 2020

Currently the intra.broker.goals goals can only be used in rebalance API when rebalance_disk=true
In environments where goal-violation based anomaly detection and self healing is enabled it would be useful to be able to define/configure these intra.broker.goals as well.

ref: https://gitter.im/kafka-cruise-control/Lobby?at=5eb9c02903af9f38ff6790df

@efeg efeg added functionality A feature request. good first issue A good fit as a first issue in the project. labels Jul 14, 2020
@efeg
Copy link
Collaborator

efeg commented Jul 14, 2020

@amuraru Thanks for submitting the feature request. There is no strong reason for not having this feature. One design option is to add another dedicated anomaly detector similar to goal_violation detector to check for intra broker goal violations and fix them when needed.

@amuraru
Copy link
Contributor Author

amuraru commented Jul 14, 2020

That could work but was wondering if the regular interbroker goals can be unified with these intra brokers goals. One edge case in your proposal would be when say DiskCapacityGoal proposal is executed but as it is not aware of individual broker disk may reassign partitions to already at-capcity individual disk

@ecojan
Copy link

ecojan commented Oct 4, 2021

@efeg can you please provide some insight on the above question?

@ev-hines
Copy link

ev-hines commented Oct 4, 2021

@efeg I'd be happy to look into submitting a PR to add this functionality. Same design question as amuraru though. Would unifying these goals be better so as to not cause race conditions between the other non-inter broker disk anomaly detectors.

If not, what would be the expected way to synchronize these components?

@imans777
Copy link

imans777 commented Jan 4, 2025

related: #2234,

I'd be very happy this PR gets reviewed and merged.

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment