fix: change maxUnavailable to 0 for zero-downtime deployments

[ChiragAgg5k]

Summary

Fixes rolling deployment to ensure zero downtime by changing maxUnavailable from 1 to 0.

Problem

With minReplicas: 1 and maxReplicas: 1, the current configuration allows the old pod to be terminated before the new pod is ready, causing downtime during deployments.

Solution

Changed maxUnavailable to 0 in the deployment strategy, ensuring the old pod remains running until the new pod passes readiness checks.

Changes

• deploy/mcp-for-docs/templates/mcp-for-docs.yaml: Changed maxUnavailable: 1 to maxUnavailable: 0

Test Plan

• Helm template validates
• Deploy to staging and verify rolling update behavior

Summary by CodeRabbit

Release Notes

• Chores
  • Enhanced deployment strategy to ensure zero downtime during service updates by maintaining full pod availability throughout rollouts.

[coderabbitai]

Caution

Review failed

The pull request is closed.

Walkthrough

This change modifies a Kubernetes Deployment configuration template by adjusting the rolling update strategy parameter. Specifically, the maxUnavailable value in the RollingUpdate strategy is changed from 1 to 0. This parameter controls the number of pods that can be unavailable during a rolling update. The change affects one line in the deployment manifest with no other functional modifications.

Estimated code review effort

🎯 1 (Trivial) | ⏱️ ~2 minutes

✨ Finishing touches

🧪 Generate unit tests (beta)

• Create PR with unit tests
• Post copyable unit tests in a comment
• Commit unit tests in branch fix/rolling-deployment-zero-downtime

---

Thanks for using CodeRabbit! It's free for OSS, and your support helps us grow. If you like it, consider giving us a shout-out.

❤️ Share

• X
• Mastodon
• Reddit
• LinkedIn

_{Comment @coderabbitai help to get the list of available commands and usage tips.}