Exit codes
Scripts, cron and CI can rely on these.
pgbackup-r2 (the image)
Section titled “pgbackup-r2 (the image)”| Code | Meaning |
|---|---|
0 |
Success. For drill, the drill passed. For doctor, nothing failed (⚠ and - lines don’t count). |
1 |
Something failed. That includes: a drill that ran and found problems (FAIL), any ✗ from doctor, a configuration error, an S3 or Postgres error, and an unexpected crash. The error is printed with a hint, and backup and drill also POST it to HEARTBEAT_FAIL_URL. |
2 |
Bad command-line usage: an unknown command or flag. |
130 |
Interrupted (Ctrl+C, or SIGTERM when it’s the container’s main process). schedule exits 0 instead, because stopping is how it’s meant to end. |
pgb passes the image’s exit code through, plus:
| Code | Meaning |
|---|---|
1 |
pgb itself stopped before running anything. For example: no .env where one is needed, a bad PG_MAJOR, an unknown pgb init option, or doctor finding Docker not running. pgb nightly status also exits 1 when the container exists but isn’t running. |
3 |
pgb nightly status: there’s no nightly container. |
130 |
pgb drill was interrupted with Ctrl+C. The throwaway database and network are still removed. |
143 |
pgb drill was stopped with SIGTERM. The throwaway database and network are still removed. |
125–127 |
Docker’s own errors: 125 means the daemon or docker run failed (for example, a PGB_NETWORK that doesn’t exist), and 126 and 127 mean the command couldn’t run in the container. |
In PowerShell, pgb drill interrupted with Ctrl+C still cleans up, but the exit code isn’t set.
pgbackup-r2 is an open-source tool from Iotabyt. MIT licensed.