blumeops/docs/changelog.d
Erich Blume e39b601353 Fix minikube role skipping start when kubelet/apiserver are stopped
After a power loss, minikube's Docker container (host) comes back up but
kubelet and apiserver remain stopped. The status check used
`--format='{{.Host}}'` which only examined the host state, causing the
role to skip `minikube start` since it saw "Running". Remove the format
flag so `minikube status` returns non-zero (exit 7) when any component
is unhealthy, and simplify all conditions to use the exit code alone.

Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
2026-02-09 22:49:02 -08:00
..
.gitkeep Add towncrier changelog system (#86) 2026-02-03 11:48:13 -08:00
feature-fly-proxy-error-page.feature.md Serve friendly error page when Fly.io proxy upstreams are unreachable (#133) 2026-02-09 12:01:24 -08:00
feature-op-backup.feature.md Add op-backup mise task for encrypted 1Password disaster recovery (#136) 2026-02-09 20:37:39 -08:00
feature-sifaka-ops-observability.feature.md Operations and observability for sifaka NAS (#135) 2026-02-09 17:44:05 -08:00
fix-deploy-healthcheck-race.bugfix.md Fix 502 errors during Fly.io proxy deploys (#131) 2026-02-09 11:07:36 -08:00
fix-minikube-status-check.bugfix.md Fix minikube role skipping start when kubelet/apiserver are stopped 2026-02-09 22:49:02 -08:00
fix-real-client-ip-logging.bugfix.md Log real client IPs via Fly-Client-IP header (#130) 2026-02-09 11:02:06 -08:00
fix-zero-downtime-deploy.infra.md Zero-downtime Fly.io deploys (#132) 2026-02-09 11:34:19 -08:00