Skip to content

test: add long-running runtime reliability tests #69

Description

@RentnerKev

Summary

Add long-running reliability coverage beyond short production smoke
tests.

The existing production smokes remain the fast merge gate.

This issue adds longer runtime/soak testing.

Scenarios

Repeatedly exercise:

  • create/update/delete Proxy Hosts
  • create/update Redirect Hosts
  • certificate changes
  • Access Policy changes
  • repeated reconciliation
  • Caddy restart
  • Controller restart
  • Web restart
  • transient runtime failures
  • recovery
  • backup/restore where appropriate

Test Duration

Do not run a multi-hour test on every PR.

Prefer:

nightly/scheduled
or explicit release-gate workflow

while keeping representative shorter reliability checks in normal CI.

Detect

  • memory/resource leaks where reasonably observable
  • stuck reconciliation
  • duplicate apply
  • stale revision activation
  • runaway retries
  • process crashes
  • state divergence

Acceptance Criteria

  • deterministic long-running test harness
  • bounded runtime
  • useful failure diagnostics
  • no production secrets
  • clean resource cleanup
  • scheduled/release execution

Priority

P0

Dependencies and sequencing

Extends completed reconciliation #12 and fast production-smoke coverage #13, rather than replacing either. Include representative #64/#65 policy behavior when those features land. Long-running release evidence is required by #75.

Activity

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Metadata

Metadata

Assignees

No one assigned

    Labels

    area: ciContinuous integration and GitHub automation.area: runtimeOpenResty and privileged controller runtime behavior.relatedRelated work that is not a confirmed duplicate.

    Projects

    No projects

      Relationships

      None yet

      Development

      No branches or pull requests

      Issue actions