[Mason0920]

Hi,

I need topologySpreadConstraints, so I create PR on this repo

about topologySpreadConstraints, you can visit here

You can use topology spread constraints to control how Pods are spread across your cluster among failure-domains such as regions, zones, nodes, and other user-defined topology domains. This can help to achieve high availability as well as efficient resource utilization.

I would be useful for other users.

[ranyhb]

Thanks for this — the feature is a good addition and it's now in #216, which consolidates it with #195 and #200 into a 4.2.0 release. You're credited via Co-authored-by.

I had to reimplement the templates rather than apply them, because the patch as written stops the chart rendering at all:

Error: parse error at (redash/templates/worker-deployment.yaml:83): bad character U+0024 '$'

.Values.$workerConfig.topologySpreadConstraints isn't valid — a variable can't be used as a field name there. The worker template already has the merged per-worker config in $workerConfig, so it just needs $workerConfig.topologySpreadConstraints (same shape as the nodeSelector and affinity blocks right next to it). Three smaller things: indent 8 should be nindent 8 or the YAML comes out malformed, topologySpreadConstraints is a list so the value wants [] rather than {}, and the chart has since gained a Knative Service template that needed the same block.

With that fixed, per-worker overrides work as you'd expect — workers.adhoc.topologySpreadConstraints beats the shared worker. default. Worth knowing for future PRs: helm template charts/redash catches this class of error immediately.

This will ship in the next chart version (4.2.0). Closing in favour of #216.