Repeatedly getting "No-data” alarms

I have this solar system working for the last 4 years.

A solar collector of about 10kWp
Two Victron Multiplus II 48/5000/70-48 inverters (master+slave)
Two Victron 450/100 MPPT
A Cerbo GX
Dyness battery bank with 21.6kWh total capacity, 6 modules

Some months ago I started to get these messages:
ALERT - No-data alarm

And after 15-20 min:
CLEAR - No-data alarm

They appear always in pairs, and the time between the ALERT and the CLEAR is always between 15 and 20 min, this is quite precise.
During this period of 15-20 min the cerbo effectively loses connectivity and I am not able to communicate with it.

I did not change any hardware or firmware, these warnings started appearing apparently without any reason.

It looks to me that some software process that starts several times a day is causing this loss of connectivity. The cerbo is connected by both ETH cable and wifi.
It happens between 8 and 10 times a day, so I receive around 16 warnings a day, which is quite annoying.

Any help?

Take a screenshot of the alarm entry on vrm and post it here please.

This is the VRM list of alarms:

These are the alert emails that I receive. First the one when alarm fires, second when it is cleared:

Sounds like a connectivity issue preventing log uploads. Try set the GX to auto reboot on VRM connection loss (GX setting → VRM section).
Is your GX up to date?
You can also open the remote console and go to: settings → general → support
Check there is sufficient free space.

When I first noticed the issue I updated the Cerbo fw to v3.71, today I will update to v3.75 and will see if there is any difference.

The option “reboot device when no contact” was already active but the time was set to 1h, will adjust it to 5 min.

Settings/VRM online portal/Free space says 1152.25M bytes → is this OK?

When you say to open the remote console and go to: “settings → general → support” I don’t understand: what should I look at there?

5 minutes is fine for a test but too short for an issue as it can end up with rolling reboots.
That is plenty of space, I didn’t expect it to be an issue but it is worth a check.
The support tab, see below, shows free space and customisations that can sometimes cause issues.

“Settings → general → support status” seems OK to me:

support status = OK
custom startup scripts = not installed
file system status = clean

In some days I will revert the reset to something more than 5 min, as you say it is only a temporary test. Let’s see if this reset, or update to v3.75 fixes something.

Great let us know how it goes.

Hello! Update to 3.75 does not fix it.

When I change “no contact reset delay” to 15 min the alarms disappear. This would be a workaround. But this means that the system is resetting several times a day, around 8 or 10 times, every time that it loses connection, and this should not happen. So I understand that something is not working well. What could I do to solve this? Should I contact Victron technical service to investigate the origin of this phenomenon?

It might worth to make some checks on your home network.

  • Is the ethernet port on your switch/router show disconnected while GX is non-responsive ?
  • Does the GX stays connected on Wi-Fi, router or access point during alarm ? Look in Logs by searching the MAC address.
  • Internet connection is working during GX alarm ?