Skip to content

[helm] scheduling primitives#3154

Merged
polyzos merged 2 commits into
apache:mainfrom
fresh-borzoni:helm-scheduling-support
Apr 23, 2026
Merged

[helm] scheduling primitives#3154
polyzos merged 2 commits into
apache:mainfrom
fresh-borzoni:helm-scheduling-support

Conversation

@fresh-borzoni

@fresh-borzoni fresh-borzoni commented Apr 21, 2026

Copy link
Copy Markdown
Member

closes #3153

Exposes affinity, nodeSelector, tolerations, and topologySpreadConstraints on both coordinator and tablet. Required to spread replicas across zones/nodes under replication factor 3 - without these, a single node failure can take out all tablet replicas for segments not yet tiered.

@fresh-borzoni

Copy link
Copy Markdown
Member Author

@morazow @affo @polyzos PTAL 🙏

@affo affo left a comment

Copy link
Copy Markdown
Contributor

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Loving where we are going with the Helm charts, thanks @fresh-borzoni for this one, another step towards full flexibility 🚀

I just have a question, why do you mention "With replication factor 3, a single node failure could take out all replicas simultaneously..."? I mean, this would hold with any replication factor if nodes are co-located, isn't it?

@morazow morazow left a comment

Copy link
Copy Markdown
Contributor

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Minor nit from my side in the values file 🤝

Thanks, shaping nicely 🚀

Comment thread helm/values.yaml Outdated
@fresh-borzoni

Copy link
Copy Markdown
Member Author

I just have a question, why do you mention "With replication factor 3, a single node failure could take out all replicas simultaneously..."? I mean, this would hold with any replication factor if nodes are co-located, isn't it?

Well, it's just me missing "Even" before with, bc we had exactly this scenario and the first reaction was - but how all 3 simultaneously :)

@polyzos polyzos merged commit 56c490e into apache:main Apr 23, 2026
3 checks passed
polyzos added a commit to polyzos/fluss that referenced this pull request Apr 23, 2026
* [helm] scheduling primitives

* nits

(cherry picked from commit 56c490e)
polyzos added a commit that referenced this pull request Apr 23, 2026
* [helm] scheduling primitives

* nits

(cherry picked from commit 56c490e)

Co-authored-by: ipolyzos <ipolyzos@apache.org>
Ugbot pushed a commit to Ugbot/fluss that referenced this pull request Apr 26, 2026
* [helm] scheduling primitives

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

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

[helm] Add pod scheduling primitives (affinity, nodeSelector, tolerations, topologySpreadConstraints)

4 participants