The reconcile job runs without a git checkout, so gh pr merge --delete-branch
fails with 'not a git repository' before enabling auto-merge. Auto-merge
itself needs no local git state; branch cleanup is already handled by the
PR-creating workflows.
* ci: auto-approve and auto-merge trusted bot pull requests
Replace the approve-only automatic-approval workflow with a combined
approve + enable-auto-merge workflow that also re-runs on every push,
plus a scheduled reconcile job that re-enables auto-merge for prepared
bot PRs if GitHub ever disables it. Bot PRs are gated to the known bot
actors, same-repo head, and base branch dev.
* ci: match bot authors by app slug in reconcile filter
gh pr list reports pull request authors of GitHub Apps as app/<slug>;
adjust the allowlist so the reconcile job actually matches bot PRs.
The Uptime-Kuma status-page API (toPublicJSON) does not emit a status
field on monitor objects, so the added status schema field and paused
early-return were dead code and did not resolve#6002. Paused monitors
stop producing heartbeats and the two consumed endpoints expose no
paused signal, so revert the change.
Unraid CPU: divide by cpus.length (the array actually being averaged)
instead of info.cpu.threads, which is more robust if the two ever
disagree, while keeping the SMT over-reporting fix.
- **Unraid CPU calculation** (#6082): Use thread count instead of core count
when normalizing CPU utilization. Each cpus[] entry is one thread, so
dividing by physical cores (cores) over-reports on SMT systems.
- **OIDC single-group sync** (#5716): Normalize the groups claim to always
be an array. Some providers (e.g. Jumpcloud) send a single group as a
plain string instead of an array, which caused the Array.isArray check
to skip group synchronization entirely.
- **Uptime-Kuma paused status** (#6002): Add the status field to the
monitor schema and check it before falling back to heartbeat inference.
When a monitor is paused, heartbeats stop arriving, so the latest
heartbeat may still show 'up' from before pausing. The status page
returns status=2 for paused monitors, which this change now respects.
Remove incompatible persisted widget query data before resetting a crashed widget so it can fetch fresh data instead of re-rendering the same stale payload.