SpaceLynk - BACnet server not working or accessable

[Richard_M]

We have a site here about to install 100 SpaceLynk type products. They have come across an issue where when they enabled the BACnet server or change the Device ID, the BACnet server no longer works and they can't even access the BACnet menus in the System page

Has anyone seen anything like this before? They were going through final config of the devices when it happened. So far it has affected almost 20.

The units have been powered, firmware updated, standard project uploaded and BACnet setting changes.

We have seen a unit on BACnet, then after chaing the settings it is gone.

Factory reset, default database, power cycle, etc won't bring back the BACnet server.

Has anyone seen something like this before?

[Erwin-vd-Zwart]

Hi,

i assume you have the Cbus version correct?

With the KNX version i have never seen this behavior but i can ask development, what HW and FW versions do you have and can you send a snapshot of your IP and BACnet settings?

[PetrMares]

Hello,

I've got yesterday same description of behaviour from our contact in Australia referring to SHAC (W4K for Cbus). As you said Erwin, for W4K, sL such a case was not reported. I will double check with L3 support.

[Richard_M]

Hi @Erwin-vd-Zwart  Yes, this is for the C-Bus SHAC (the residential version of the NAC) in Aust. I posted the question in the KNX section because the Australian market is small and we know the issue hasn't been seen here before. The hope was that it has been experienced in similar products overseas.

NAC/SHAC firmware is v1.6.0 and HW is 1.0

@PetrMares  - this is the same issue you were asked about. We pushed it to the local engineering team and they said they reached out overseas - which I guess means you 🙂

Thanks for looking into this for us. Like I said above, the customer has a mixture of 100 NAC'sSHAC's they need to get to site in about a week. So far every time they enable BACnet and change the Device ID they lose BACnet altogether.

We are still gathering facts about the issues. There was some further info this afternoon that perhaps they got their NAC's back online by resetting them. We should be receiving some faulty units from the customer early next week so hopefully that will shed some light on whether it is a hardware or user error.

[Richard_M · Accepted solution]

Hi All. I think we have found the cause if the issue - well I didn't find it myself but some other people smarter than me did 🙂

It appears that the NAC had a user library which should have been emailing a project backup once a day. This backup was "temporarily" being written to the root partition. The issue was that the backups weren't being deleted - resulting in the root eventually filling (they were sitting at 96-98% full).

The action of changing BACnet settings (which apparently gets written to the root partition) was too much for the device and it locked up.

It appears they didn't realise this user library was there. It was probably a leftover from their earlier testing. Since it was in the main project template, it then got duplicated throughout all of the NAC/SHAC's in the project. We are now working through a process to remove this user library and clear the used memory.

Thanks all for your advice.