a month ago
Hello Railway Support,
We are reviewing recovery options for a Dockerfile-built cron service. A successful historical deployment is stopped, and its canRollback and canRedeploy flags are both false. We have recorded a build-output digest and a deployment imageDigest, but have not established their relationship.
Could you clarify:
- Can the image associated with a stopped cron deployment still be retrieved or referenced immutably? What retention or access limitations apply?
- Can a build-output digest and deployment
imageDigestidentify different image representations—for example, an image index, platform manifest, or configuration object? - What supported method identifies the authoritative registry reference and links the build output to the image actually deployed?
- Under what conditions can both eligibility flags be false for a successful historical deployment?
- What supported recovery options remain for a Dockerfile-built cron service when neither rollback nor redeploy is available?
- Could a service-specific replacement deployment execute immediately, overlap an existing execution, or affect the next scheduled cron run? What sequence preserves scheduler continuity?
This is an informational inquiry only. Please do not initiate deployments, executions, or configuration changes.
Thank you.
1 Reply
a month ago
We don't hand out the images behind past deployments, and there's no self-serve way to pull one by digest. Rollback is how a past image gets reused, and it only works inside your plan's image retention window.
Both flags can be false on a successful past deployment for a few reasons. Cron services don't offer rollback or redeploy of historical deployments at all. Separately, rollback stops being available once that deployment's image is outside the retention window.
The two digests can refer to different things. The build output can point to a multi-platform image reference, while the deployment's imageDigest is the digest of the platform-specific image that actually ran. We don't offer registry access that would let you link the two yourself.
For a Dockerfile-built cron service, the supported way to recover is a new deployment built from your source, for example a new commit or railway up.
A new deployment on a cron service doesn't run right away. It waits for the next scheduled time, and it doesn't stop a run that is already going. If a run is still going when the next one is due, that scheduled run is skipped (cron docs). To keep the schedule intact, deploy while nothing is running, which you can check with railway cron list, and let the next scheduled run pick up the new deployment.
Status changed to Awaiting User Response Railway • about 1 month ago
24 days ago
This thread has been marked as solved automatically due to a lack of recent activity. Please re-open this thread or create a new one if you require further assistance. Thank you!
Status changed to Solved Railway • 24 days ago