Is there a way to manually trigger an agent scan to get a list of containers?

[begunfx]

There is a scan to check for updates but I would like to know if ther is a way to trigger a manual scan for available containers? If not, please add it on the agent page. If I have to delete a container (it was in a stopped state and needs to be removed) then I'd like a way to force a scan to get the container back on the list. it seems like I have to wait for scheduled scan.s

[scttbnsn]

Hey @begunfx — great question! There are actually a couple of ways to force a container refresh right now, though they're not on the Agents page specifically.

What's available today

From the Containers page:

• The "Recheck" button at the top of the Containers view triggers a full re-watch of all containers (POST /api/v1/containers/watch). This re-lists containers from Docker, discovers new ones, and prunes stale ones — so a container you deleted and re-created will reappear immediately.
• Individual containers also have a refresh action (POST /api/v1/containers/:id/watch) available from the container detail view.

From the API directly:

• POST /api/v1/containers/watch — re-watches all containers across all watchers (including agents)
• POST /api/v1/containers/:id/watch — re-watches a specific container

Docker events auto-detection:
If DD_WATCHER_{name}_WATCHEVENTS is enabled (it is by default), Drydock listens for Docker events and automatically triggers a debounced re-watch within ~5 seconds of container start/stop/remove events. So in most cases, starting a new container should be picked up quickly without waiting for the next cron cycle.

What's missing

You're right that there's no scan/refresh button on the Agents page itself. That's a reasonable feature request — having a per-agent "Rescan" button would make it easier to manage when you're working with the distributed agent architecture.

For now, the Containers page Recheck button will cover all watchers including agents, so that should unblock you. But I'll track the idea of adding a per-agent rescan action.

Docs on watcher configuration: https://getdrydock.com/docs/configuration

[begunfx]

I guess my question - is the rescan on the container page just scanning the local docker installs or does it do a re-grab of everything from the registries or both? If it just does a local scan then that's all I'm looking for, but if it also grabs from the registries, I just want to be able to manually update my local list of containers should one need to be re-watched etc.

[scttbnsn]

Good question! The Recheck button (POST /api/v1/containers/watch) does both — it re-lists your containers from Docker and checks each one against the registries for updates. There's no separate "just refresh the container list" mode.

That said, it's fast — the registry check for each container is a lightweight tag/manifest lookup, not a full image pull. So hitting Recheck after you've deleted and re-created a container will:

1. Re-discover the container from Docker
2. Check its registry for any available updates
3. Update the UI with the current state

If DD_WATCHER_{name}_WATCHEVENTS is enabled (default), Docker events will also auto-trigger a debounced re-watch within ~5 seconds of container start/stop/remove — so in most cases you won't even need to click Recheck manually.

For your use case (container was stopped, removed, and needs to come back on the list), Recheck will handle it. The registry check is a side effect but shouldn't cause any issues — it's the same thing that happens on every scheduled cron cycle.