Skip to content

fix(deploy): retry transient MIE manager reads#305

Merged
Taleef7 merged 1 commit into
mainfrom
fix/mieweb-manager-retry
Jul 16, 2026
Merged

fix(deploy): retry transient MIE manager reads#305
Taleef7 merged 1 commit into
mainfrom
fix/mieweb-manager-retry

Conversation

@Taleef7

@Taleef7 Taleef7 commented Jul 16, 2026

Copy link
Copy Markdown
Owner

Root cause

The final WebChart-stack deploy failed before authentication or container creation because manager.os.mieweb.org:443 was unreachable. The deploy helper inherited curl's long platform timeout and treated a transient GET transport failure as permanent; state-changing requests and later job polling were not the failing boundary.

Fix

  • apply 10-second connect and 30-second overall limits to every manager request
  • retry safe GET transport failures up to six times with 20-second spacing
  • keep POST and DELETE single-attempt because a lost response may already have changed manager state
  • run a no-network regression test in CI
  • document recovery and the 2026-07-16 incident

Verification

  • bash .github/scripts/mieweb-api-request.test.sh
  • bash -n on the new helper/test and CRLF-normalized existing deploy script
  • git diff --check

No runtime, dependency, schema, PHI, or Outcome Status behavior changes.

@Taleef7
Taleef7 merged commit fe06e5b into main Jul 16, 2026
8 checks passed
@Taleef7
Taleef7 deleted the fix/mieweb-manager-retry branch July 16, 2026 20:31
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.

1 participant