mr job retry
Re-queue a failed or cancelled download job for another attempt.
Retry only works against jobs in the failed or cancelled state;
the server rejects retry on jobs that are still active, paused, or
already completed. A retry is also refused with HTTP 409 while any queued
or running job is already fetching the same URL, so one URL is never
transferred twice. The existing job's ID is reused: progress, error
message, and completion times are cleared, then the worker re-runs the
original URL fetch.
Useful when a transient network error blew up the first attempt.
Persistent failures need an updated URL, which means calling
job submit fresh rather than job retry.
Usage
mr job retry <id>
Positional arguments:
<id>
Examples
Retry a specific failed job
mr job retry a1b2c3d4
Retry every failed job in the queue
mr jobs list --json | jq -r '.jobs[] | select(.status == "failed") | .id' | xargs -I {} mr job retry {}
Flags
This command has no local flags.
Inherited global flags
| Flag | Type | Default | Description |
|---|---|---|---|
--json | bool | false | Output raw JSON |
--no-header | bool | false | Omit table headers |
--page | int | 1 | Page number for list commands (default page size: 50) |
--quiet | bool | false | Only output IDs |
--server | string | http://localhost:8181 | mahresources server URL (env: MAHRESOURCES_URL) |
Output
Object with status set to "retrying"
Exit Codes
0 on success; 1 on any error