Network watchdog help

Node-RED
, ,
1

Is there a way in node-red to know if network has lost connection to vrm?
I know the cerbo has a watchdog ,but i’m trying to reboot the router should this happen .
I found this path but i cant find any info about it .

Screenshot 2026-07-09 112832
Screenshot 2026-07-09 112832522×746 21.3 KB

2

I have a Node-Red flow to do this. I use a custom input node to read the time of the last contact with VRM from the Cerbo. This is in milliseconds since a reference time. I then convert this to seconds and subtract it from the current time to give the elapsed time since last contact. If this exceeds a trigger value a reboot is forced. Of course this may also cause a reboot if VRM servers go down.

IMG_20260709_204116
IMG_20260709_2041161280×549 74.5 KB

3

yes I found vrm connection error
0 is no error 153 is no connection , not sure if there are other errors.

Screenshot_20260709_162254
Screenshot_20260709_162254883×852 72 KB

4

I had this working previously but now it’s not (v3.81). The com.victronenergy.logger shows as disconnected and no longer available.

image
image496×500 11.1 KB

image
image710×110 11.6 KB

I’m not sure which OS version killed it and didn’t notice when it stopped working.

Phil

5

I can confirm mine is no longer working and com.victronenergy.logger also shows as disconnected and no longer available. V3.80.

@dfaber bug or deliberate change.

6

I deleted the flow and imported a copy of it that I have on my PC and the time since last contact from com.victronenergy.logger started working, but after a reboot it stopped working again. I also no longer connect to the reboot Node from Venus (0).

image

image
image509×419 13.2 KB

image
image244×114 1.73 KB

image
image514×416 12.8 KB

7

I’m still on 3.79 and it’s working.

8

What if you redo the custom input node from scratch?
I had virtual switches I had to delete and redo.

9

Good idea, I checked it out with a new custom Node and it did not list com.victronenergy.logger or Venus (0) as available services.

10

I spent ages trolling through nodes and cannot find anything that does what the com.victronenergy.logger did.

Any clues that I missed?

If it is indeed a bug, can we please get it restored to its previous (working) state?

Phil

11

Rolling back to the previous version might be a short term solution.

12

Yes - rolled back to v3.75 and all working. But that’s only a band-aid not a fix.

Victron?

Phil

13

@dfaber
Can we get an update on this?
I’d like to update cerbo to latest version for security update.

14

Thanks for the reports, and sorry for the slow reply.

There are two separate things going on:

  1. com.victronenergy.logger missing: this is a bug in node-red-contrib-victron. Since a change in the way services are discovered at startup, the logger service was skipped, so after a reboot it no longer shows up (it only appeared again by chance if the logger restarted while Node-RED was already running). A fix is ready and will, once approved, land in the next release of node-red-contrib-victron, then in the beta 3.90, before it can get backported to 3.8x.

  2. “Venus (0)” missing in the reboot/platform node: this was a deliberate fix. Services like com.victronenergy.platform and com.victronenergy.system have no device instance, and the “(0)” suffix caused duplicate and broken entries. They are now listed without the “(0)”. Open the node, select the service again and deploy, and it will work again.

For next time: if something in the Victron Node-RED nodes stops working, please open an issue at Issues · victronenergy/node-red-contrib-victron · GitHub. Reports there get picked up much sooner than on the forum, where it is easy to miss a thread.