Observability, built in#
Every service now comes with its logs, metrics and HTTP requests, collected with nothing to install.
- Log history for every service, searchable by text and level.
- Traces: add an OpenTelemetry SDK and each request shows as a waterfall of spans across your services, with a requests overview.
- Alerts on CPU, memory, error rate and latency.
- Metrics in vCPU, with rates and time windows.
How long data is kept depends on your plan. See Observability.
Volumes#
Volumes are disks attached to one service, for data that must survive a deploy. They are backed up every 15 minutes and restore into a new volume, so a restore never overwrites what is there. See Volumes.
Point-in-time recovery#
PostgreSQL, MySQL and MariaDB databases can be restored to any second within your plan's history window, where your plan includes it. See Backups.
Improvements and fixes#
- Public networking for a database is now a switch, and Hangar assigns the port.
- A service's settings were redesigned: one form for the source, runtime limits in plain units and a new Scaling section.
- GitHub connects through Hangar's own app, with no app of your own to set up. See Git providers.
- The pricing page is public, and prices show in your market's currency.
