SHS in docker on Ugreen, adv flows issues and errors

Homey Self-Hosted Server
1

Hi,

My errors and weird behavior started with 12.11.0 update.
I saw error on startup from the flow manager.
Homey [err][ManagerFlowLocal] HomeyErrorBadRequest: Invalid Token: previous_action
Z at FlowToken.validateValue (file:///app/packages/homey-core/lib/FlowToken.mjs:84:13)
E at file:///app/packages/homey-core/lib/AdvancedFlow.mjs:1378:27
at Array.map (<anonymous>)
h at AdvancedFlowLocal.validateTokens (file:///app/packages/homey-core/lib/AdvancedFlow.mjs:1376:29)
d at AdvancedFlowLocal.runTriggers (file:///app/packages/homey-core/lib/AdvancedFlow.mjs:704:18)
T at process.processTicksAndRejections (node:internal/process/task_queues:104:5)
b at async AdvancedFlowLocal.run (file:///app/packages/homey-core/lib/AdvancedFlow.mjs:398:61)
H at async file:///app/packages/homey-core/lib/ManagerFlow.mjs:866:9
$ at async Promise.all (index 0)
h at async ManagerFlowLocal.triggerFlows (file:///app/packages/homey-core/lib/ManagerFlow.mjs:858:5)
F at async file:///app/packages/homey-core/lib/Device.mjs:1463:9 {
statusCode: 400
} (edited)

I asked about this on Slack but got no response.
I thought, it might be fixed in 12.13 so updated to that version. but same errors and behavior with the flows.

The flows are shown, but without the lines between the items. You see the dots highlighted, and after about 10 seconds the lines start to appear.
When I test a flow, the flow stops at the first next item and takes a long time to complete.
Looks like the whole flow part is very very slow.

So 13.0.0 comes out, updated to that version. Still the same errors and behavior.

The errors are from the SHS itself and not from the environment it runs in, so I must be in the container itself, so Athom might need to solve this.
Cannot file a support ticket , they point to community, and on slack no response.

So that is disappointing.

Hopefully I am not alone and other have the same behavior and errors?
So somewhere on the forum that other users experienced something similar, with slow flows and editing of flows.

Before 12.11.0 I had no issues.

2

Weird that they point so easily to the community. You pay for selfhosted and it’s not like you can fix everything by yourself.

I’m still on 12.10.1 (on Proxmox, but shouldn’t matter), happy I disabled auto-update from the start, but that doesn’t help you.

I don’t have much ideas in this case, there seems to be an issue with the token defined somewhere, it feels like something Athom should investigate. But maybe someone else has a great idea?

With the docker image you could revert to a previous version, but the previous docker image might conflict with the current userdata / database. If you don’t have backups of your user data that could be a problem.

In case you want to try something make sure to backup the current userdata first so you can always go back to the current situation, without creating bigger issues.

3

Not sure where this is coming from, but given that you believe it is something in the container, I would try to remove and recreate the container.

With remove I don’t mean a simple rebuild, but actually deleting the container. I don’t have a Ugreen NAS, but it looks very much like Synology, where deleting the project deletes the container.

If you’ve done a proper mapping of the homey/user volume to a volume on your host (as explained in the how to), you should be able to remove your container and rebuild without losing any configuration.

PS: For easier container management you could also have a look at Dockhand which allows you to delete your container without having to delete your stack (project)