services are killed if registry is unavailable during update about shepherd HOT 3 CLOSED

Thanks, this should probably be done.

we had an issue with our private registry (unvailable because of network maintenance) and during that period shepherd stopped/kill a few services.

Were the affected services really stopped? I would have expected the old services to stay running it that case. (Okay, maybe it depends on the restart configuration.)

As a side effect, service update would become faster (for large images or slow registries)

Well, overall the time is the same. Also, the speed-up only works if the service is running on the same machine as the shepherd container.

from shepherd.

we had an issue with our private registry (unvailable because of network maintenance) and during that period shepherd stopped/kill a few services.

Were the affected services really stopped? I would have expected the old services to stay running it that case. (Okay, maybe it depends on the restart configuration.)

You are right, the service (in docker terms) stayed there, but all tasks (containers) were stopped. I suspect that dockerd tried to restart the tasks until each hit the restart limit.

from shepherd.

This issue has been automatically marked as stale because it has not had recent activity. It will be closed if no further activity occurs. Thank you for your contributions.

from shepherd.