question

ur12vman avatar image
ur12vman asked

CCGX Color control continually rebooting

CCGX in system with 3-quattros, 2-bluesolar chagers, BMV712, VE Bus BMS and Smart LiFePo4 batteries. the system has been stable for several years. Recently added in the BMS, LiFePo4 batteries and the VE Bus. All firmware up to date. All has been working for over a week. Today the CCGX began to reboot every 5-15 minutes, everything else works fine. I changed the power tap for the CCGX which made no difference then disconnected the NMEA 2000 feed which stopped the rebooting. What reasons would having a N2k connection cause the CCGX to reboot continuously? Any suggestions would be appreciated.

CCGX Color ControlNMEA 2000 - N2K
2 |3000

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

2 Answers
JohnC avatar image
JohnC answered ·

Hi @Ur12VMan

This may be related to processing power, which I believe can cause reboots if 'overchallenged'. So it may not be the NMEA kit alone, but it just happens to be part of the total.

I'm just guessing, but maybe something you can experiment with.

You've a fair bit of kit connected there, and if that's the issue, a Cerbo GX upgrade might be the only real way to overcome it if you can't be without any of the stuff you want.


2 |3000

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

ur12vman avatar image
ur12vman answered ·

@JohnC Thanks for the suggestion, I have not seen any way to see the network load on the system, I can see N2k load using Maretron N2kView, but nothing on the VE side. Do you know of a way to measure the network load? Seems like the CCGX would be able to handle it, this is a pretty small system compared to what VE suggests is possible. More timeline: After disconnecting the N2k feed, the CCGX was stable for most of a day, then it rebooted a couple more times, but not anywhere near the frequency before. So, problem still not solved... On the upgrade issue, I have a cerbo on order, it is supposed to be here in March. However we are starting a 20 day job for a gov agency monday, so it would be nice to have this working better.

Thanks for your suggestion.

6 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.

Hi @Ur12VMan,

I also think JohnC is on the right track. When there is too much information required for processing, from too many connected devices, the GX device CPU can restart.

The limits for VE.Direct devices is specified in the GX product family overview - footnote 1. But less so for all the other possible configuration variables.

We are working on improving the system performance monitoring tools inside Venus (the software on GX devices) to better help those with larger systems better manage the GX resource limitations.

This will be MUCH less of a problem with the introduction of the more powerful Cerbo GX, but Victron systems are still getting bigger and bigger every year.

The MaxiGX was the other product introduced for these bigger systems. Designed for some systems that had 100kW of DC PV through Victron MPPTs.

0 Likes 0 ·
ur12vman avatar image ur12vman Guy Stewart (Victron Community Manager) ♦♦ commented ·

Thanks Guy, good to know the cerbo will have some more cpu muscle. This would be useful to have in the manual, I had no impression that the system was getting too large. I did see the limitation on VE Direct devices, I only have one (the bmv), a 3ph quattro and a couple solar chargers didn't seem like a massive processing issue. My distributor also didn't twig to this, nor was it covered in the training. How are those cerbo's coming?

0 Likes 0 ·
Stefanie avatar image Stefanie ♦♦ commented ·

If you have ssh root access through a terminal and have a bit experience with linux command line, you can get a real time indicator for system load and possible cause for lag when using the command "top".

0 Likes 0 ·
ur12vman avatar image ur12vman Stefanie ♦♦ commented ·

Thanks for the suggestion, I have no idea how to implement any of what you wrote.


0 Likes 0 ·
wkirby avatar image wkirby ♦♦ commented ·

Another suggestion I can think of is to check the "Services" menu.
You can disable any services which you are probably not using like MQTT and Modbus TCP. This should free up some CPU cycles.

0 Likes 0 ·
1581857132025.png (23.0 KiB)
ur12vman avatar image ur12vman wkirby ♦♦ commented ·

Excellent thought, I do have those already off since they are not being used. Thanks

0 Likes 0 ·