question

mindaugas avatar image
mindaugas asked

Node Red on Cerbo stops many times a day

I have a programed logic on Node-red adjusting grid feed in depending on SOC, solar etc.

The prevoius version worked fine for quite long time.


What I have done recently:

Upgraded Cerbo to latest version

Improve Node-Red flow adding some extra checks


Since then, not sure what was reason I have found that my logic stop working. After more investigation I found that restart of Node-Red or restart of Cerbo fix all.


I also found that after this happen I can not access neither Dasboard or Node -Red

Node-RED unreachable

Normally Node-RED can be found on this location. It either is starting and will show after a refresh, or it is not enabled in menu. More information on Node-RED on Venus can be found here.


All the rest except Node-Red works on Cerbo after crash


Full reinstall of node red using venus-data-20-node-red-reset.tgz didnt helped.

Memmory status

attc413xiruvagfpohoiwn3ysqtgwwrgv3uavswqskua7m.png


Deletion of previuos versions files did not made any impact

cerbo gxNode-RED
2 |3000

Up to 8 attachments (including images) can be used with a maximum of 190.8 MiB each and 286.6 MiB total.

6 Answers
matt1309 avatar image
matt1309 answered ·

Hi @Mindaugas


Are you able to share the changes you made in node red recently. Maybe someone can spot an issue that's causing the crash.


It may also be worth checking the node red logs. /var/log/node-red-venus

Might give you some clues as to what's causing it.

1 comment
2 |3000

Up to 8 attachments (including images) can be used with a maximum of 190.8 MiB each and 286.6 MiB total.

mindaugas avatar image mindaugas commented ·

It works for aprox 6-8 hours and then stops any time. Looks like duration from restart matters not some logic on node red. Same logic used on another cerbo by my friend who was who helped me to write it and it works without any stop.


I havent tried to disable that logic. I will try during weekend just to see if this helps to find a issue


Thanks I will check logs

0 Likes 0 ·
Fideri avatar image
Fideri answered ·

@Mindaugas

I was having unexplained issues with Node Red but starting it in "safe mode" helped. I expected it to revert to normal mode automatically on deploying my nodes. It remained in "safe mode" even after restarting my Cerbo but I could still use any function I needed. So I left it there. It's now more than 2 wks and it's running flawlessly.

F.

2 comments
2 |3000

Up to 8 attachments (including images) can be used with a maximum of 190.8 MiB each and 286.6 MiB total.

mindaugas avatar image mindaugas commented ·
Thanks

Let me try safe mode

I am about to reinstall Cerbo

Mindaugas

0 Likes 0 ·
mindaugas avatar image mindaugas commented ·

1716191292198.png

I got this warning when started in safe mode. But after Deploy all works, let me see if this will help

0 Likes 0 ·
1716191292198.png (8.3 KiB)
Fideri avatar image
Fideri answered ·

The warning is normal.

2 |3000

Up to 8 attachments (including images) can be used with a maximum of 190.8 MiB each and 286.6 MiB total.

mindaugas avatar image
mindaugas answered ·

after resize2fs.sh - got extra space

1716229441137.png


1716229441137.png (22.9 KiB)
2 |3000

Up to 8 attachments (including images) can be used with a maximum of 190.8 MiB each and 286.6 MiB total.

Anders avatar image
Anders answered ·

It does look like you have messed up your root - I experienced something similar some weeks ago - a factory rest according to Victron procedure solved this, but I also experience that Node-red lock up frequently during the day. A Cerbo GX restart solves it but is annoying.

It does not look like Node-Red actually stops running - It could be that it's only the Dashbord and the access to Node Red thats stops working -

I have a rather heavy Node red installation consuming 50% of CPU.

1 comment
2 |3000

Up to 8 attachments (including images) can be used with a maximum of 190.8 MiB each and 286.6 MiB total.

50% cpu means that one of your two cpus is totally maxed out.


-> not good for single threaded applications, like I think Node-RED is.

0 Likes 0 ·
Dirk-Jan Faber (Victron Energy) avatar image
Dirk-Jan Faber (Victron Energy) answered ·

Node red only writes to the /data partition, so the disk space can't be causing it.
If Node-RED crashes, it usually writes a log line into /var/log/node-red-venus/current.

You can check that file by typing `cat /var/log/node-red-venus/current` or `tail -f /var/log/node-red-venus/current` (and ctrl-c ing when done).

2 |3000

Up to 8 attachments (including images) can be used with a maximum of 190.8 MiB each and 286.6 MiB total.