Repository navigation
Statup works with the monitoring you already run, and a false outage no longer stays on the page.
What's new
- HTTP API. Keys made in Settings, API, shown once and kept only as a hash. With a key, a script does what an editor does: read the services and the events, add a service and set its state, publish an incident, a maintenance or an announcement and post its updates, with the same alerts as the page, signed with the key's name.
- Monitoring tools. Prometheus Alertmanager, Grafana, Zabbix and Uptime Kuma, each tried for real, and any other tool that can call a web address or run a script, report a service down or back up at one address, and Statup shows it as a detected outage, with its alerts and an incident ready to publish.
- Import from Uptime Kuma. Under New on the Services page: paste the URL of a Kuma status page, tick its groups, and Statup creates a folder per group and a service per monitor, under the same names. Importing again adds only the new monitors.
- Detected outages explained and erasable. A click on a coloured day shows that day's outages, with the reason for members. It was not an outage lets an editor erase an ended one, with those of the services that fell at the same moment.
- Clearer Alerts and Team pages. A channel's name field reads Name in Statup, an ntfy channel only needs its topic, and the channel and member lists are one line each, with a member's role and actions under Manage.
- Clearer wording across the app, in every language.
Fixes
- When Statup itself loses the network, its checks pause instead of declaring every service down after five minutes; the Journal records when the network went and came back.
Upgrade
Back up your data first (see Backups), then, next to your docker-compose.yml:
docker compose pull && docker compose up -dThe database is upgraded on start. Every change is listed in the changelog.