Skip to content

Avoid terminal classification of transient Admin failures - #3

Merged
igalshilman merged 2 commits into
mainfrom
pavel/avoid-transient-admin-failures
Sep 4, 2026
Merged

Avoid terminal classification of transient Admin failures#3
igalshilman merged 2 commits into
mainfrom
pavel/avoid-transient-admin-failures

Conversation

@pcholakov

@pcholakov pcholakov commented Sep 3, 2026

Copy link
Copy Markdown
Contributor

A few small additions, mainly CD safety cleanups.

-P.

Summary

Prevent the service delivery checks from treating every AdminCallRejected as
a terminal failure, and require the Terraform readiness script to match the
current generation exactly.

AdminCallRejected includes transient non-2xx Admin API responses such as
5xx errors. Argo CD and Flux now leave that state in progress until it recovers
or reaches their bounded timeout. Conditions that represent a controller error
or a foreign-deployment conflict remain terminal.

Validation

  • terraform -chdir=terraform/01-foundation validate
  • terraform -chdir=terraform/02-restate validate
  • terraform -chdir=terraform/03-services validate
  • terraform fmt -check -recursive terraform
  • bash -n scripts/wait-restatedeployment.sh
  • ruby scripts/check_repository.rb

The full validator was not run because OpenTofu is unavailable in the local
environment.

References

@pcholakov
pcholakov marked this pull request as ready for review September 3, 2026 19:47
@igalshilman
igalshilman merged commit 4271697 into main Sep 4, 2026
2 checks passed
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.

2 participants