Repository navigation
llm-slo-decision-gen: schedule onto tainted nodes; slo-api 0.2.1 - #92
Closed
zhanghaohit wants to merge 1 commit into
Closed
zhanghaohit wants to merge 1 commit into
zhanghaohit wants to merge 1 commit into
Conversation
The three Deployments (decision-gen, its unstable twin and slo-api) had no tolerations, nodeSelector or affinity, so on a cluster whose every node is tainted -- an NPU cluster with huawei.com/npu:NoSchedule -- none of them could schedule. Add the three as top-level values applied to every pod of the release; unset renders nothing, as before. Also default slo-api to 0.2.1, which syncs on a cluster with no LLMSLORequirement (0.2.0 crash-looped there).
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
Scheduling. decision-gen, decision-gen-unstable and slo-api had no
tolerations,nodeSelectororaffinity. On a cluster where every node is tainted -- an NPU cluster carryinghuawei.com/npu:NoScheduleon all nodes -- none of them could schedule. Adds the three as top-level values, applied to every pod of the release (they are small CPU-only services; one setting for all is what such a cluster needs). Unset renders nothing, as before.slo-api 0.2.1. The default image moves to 0.2.1, which marks its CR index synced on a cluster with no
LLMSLORequirement(modelsphere/slo-api#6); 0.2.0 crash-looped there.Chart 0.3.5 → 0.3.6.
Checked with
helm lint --strictandhelm template(unstable enabled so all three render): without the values no Deployment has tolerations or a nodeSelector; with them all three carry both.