Ethernet and Slave Comms issues after thunderstorm

[caravanboy]

We had some brief thunderstorms overnight, at one point there was a very loud crash and then we had a 20 second power cut and Comfort started beeping and complained of Slave Communication Failure.

I received an SMS with the alarm code for "Slave Tamper"

My ETH03 also lost network presence from that point (normally I have a Home Assistant system that receives live Comfort activity via a python script so I know it was working up until that moment).

Some odd things are now happening. Here's what I've done so far in order:

1. Signed in to keypad (connected to Comfort Master unit) and did engineer reset 7.4.2 to reboot the system.
   
2. Keypad said WRONG TEXT PRESS *1 TOGETHER (and also mentioned a download failure)
   
3. Then it alarmed and said Slave comms failure. Also received an SMS (the backup one that can be configured for use in event of comms failure)
   
4. Left it until morning - noticed keypad connected to Slave also had the WRONG TEXT message. Acknowledged that.
   
5. Powered off Master (mains and battery) and back on.
   
6. It alarmed and said Slave comms failure. Acknowledged on keypad
   
7. Powered off Slave (mains and battery) - main unit alarmed and complained about slave comms failure. Acknowledged that
   
8. Restarted Slave.
   
9. Keypad (connected to slave) said WRONG TEXT PRESS *1 TOGETHER
   

This has NOT resolved my issues. The ETH03 is lit up and seems normal but I cannot ping it. Everything else on my network is functioning normally.
Comfort is sending SMS messages as normal (e.g. when I open a gate) 
Comfort is NOT playing voice messages on the keypad - e.g. "MAIN GATE OPENED"

Slave and Master are located in the same building (a normal 4-bed house), one at front, one at rear of building, connected by CAT5E cable.
ETH03 is in the Master unit.

Point 7 above implies the connection between Master and Slave panels IS okay?

Any suggestions how to diagnose or go forwards given that the ethernet connection is no longer working?

[slychiu]

The LCD "wrong text: means the Comfort has lost its voice, you can try to reprogram the vocabulary by doing Transfer Vocabulary Upgrade and choosing teh kate or myma bin file. But I am afruad the voice dsp may be damaged

Check which LEDS are on or blinkoing on the slave. If there are no leds on it means the slave is damaged
Do you have a UCM/ETH03.are the red D9 and green D10 leds blining? if only the green led is blinking it could mean the ucm is damaged. are you able to remove the ETH03 and plug into the CM9001 (if your model has the 10 way conector)

Severe lighting can damage much of the electronm\ics ewen though they have extensive surge protection. the surge may travel via the telephone line or zone sensors on the externla windows and doors

[caravanboy]

Hi, thanks for your speedy reply. Annoyingly the forum didn't send me an email even though I'm subscribed, so I didn't spot it until now.

I think the slave is probably okay because:

1. There is a green LED (possibly labelled D2) in the middle that is lit solid. The red (D1) LED is not lit.
   
2. The zones directly connected the the Slave seem to work (their names also show on the keypad when triggered)
   

Do you agree?  That doesn't explain why it says it has a comms issue with the slave after a restart though?

The UCM ETH03 is mounted in the main panel (along with the UCM GSM). The LEDs are as follows:

• Main UCM board:
  
• Green RDY LED lit
  
• Three red LEDs not lit
  
• D10 Green LED looks solid but is probably flickering very quickly
  
• D9 Red LED is flickering/flashing quickly 
  
• D10 and D9 seem to mirror the flashing pattern on the UCM GSM (which is definitely working as it is still sending SMS texts when some zones are triggered)
  
• ETH03 Daughterboard:
  
• Red power LED on solid
  
• Green status LED mostly solid, but approx every 15 seconds it goes out for a few seconds and the green link light on the cat5 socket flashes 7 times. When it does that the link light on switch port also flashes in the same way
  

I've tried a different CAT5 cable and a different switch port but I still cannot talk to the ETH03. It doesn't seem to be sending a DHCP request for an IP address.

Should I try pressing any of the reset buttons on the ETH03 - I guess that is unlikely to make a difference when the power-cycle made no difference?

I'm not sure what the 10 way connector you mentioned means. Do you mean connect the ETH03 daughterboard direct to the main Comfort board? I can see a PROG connector, but not any other obvious connector that I could use.

Any further suggestions on how to diagnose the issue? I assume I have to get the ETH03 working first or I won't be able to fix the voice issue (or do anything!).  My system did once lose its voice before in the past and a redownload of the voice files fixed it.  Without the voice the system can't be programmed from the keypads either really.

[slychiu]

youir comfort may nbot have the 10 way connnector that the ETH-03 can plug into because it is an older model. there is a reset button on ET03 but if you have powered off tyhat may not do much good. can you run comfort server manager? this is by Comfigrator > options settings. select Network and click on Comfort Servert Manager. You should see the ETH03 IP address in the Comfort server manager window that pops up

[caravanboy]

I ran Comfort Server Manager but it does not show any ETH03 on the network.
I assume this is because the ETH03 is not actually connecting to my network switch at the link level - the switch is a managed switch and it says nothing is connected to that port. I've tried a different port and a different cable but it still doesn't work.

Could the ETH03 be partially damaged even though its status lights are all okay. Is there anything else I can try to get it working, is there a way to force reset the ETH03 from the on-board buttons?

I'm still puzzled as to why Comfort complains about a comms failure on reboot if only the ETH03 is broken (since I think the Slave is working as per my previous message). Could there still be some wider issue at play.

Perhaps unrelated to this whole issue, this morning I also suddenly received a Slave battery alarm (plus an SMS message) and then 18 seconds later a slave battery restore message. Then about 90 seconds after that later I got third SMS message saying Power Failure Slave Restore. I could understand a battery issue (my battery is several years old), but why would it then mention a power failure too (just for the slave) when as far as I can see from everything else in the house there has been no power failure?
This also highlighted the fact that when Comfort rang me it has also lost its voice for the phone calls too, not just the keypad menus   I just kept pressing # in the hope that it cleanly ended the call.

[slychiu]

Locate the switches: Find the SW1 (Reset) and SW2 (Default) push-buttons on the ETH03 module.Hold SW2: Press and hold down the SW2 switch.Press SW1: While continuing to hold SW2, press and release the SW1 (Reset) switch once.Observe the D2 LED: The Green D2 LED on the ETH03 will turn off for about 5 seconds and then turn back on.Continue holding: Keep holding down the SW2 switch until the D2 LED goes through another cycle of turning off and turning back on (this will take about 10 more seconds).Discover the module: Once the reset is complete, you can run the CS Manager to discover and reconfigure the ETH03 module on your network.

[caravanboy]

Hi,

I've tried the ETH03 reset as suggested above. The LEDs behaved exactly as described but it makes no difference to the functionality.  The Green LED still goes off for a few seconds every 15 seconds and the CAT5 port's green LED flashes 7 times.

I've tried a third CAT5 cable and even a different network switch but the ETH03 module just doesn't seem to want to connect at the link level 

I think I have to assume the ETH03 is somehow damaged.   Given that the other UCM inside the panel is functioning normally I presume it can't possibly be the main board (or any other component) that could be causing the ETH03 to not connect given that everything else (other than the lost voice) seems to be okay?

I've also emailed Laser in the UK and they have an ETH03 in stock and they'll accept a return for a small fee if it turns out not to fix my issue. So I think I'll get one ordered.

[caravanboy]

Laser sorted me with a new UCM/ETH03 very quickly. I've installed that today and now have network connectivity again. Phew.

I was then able to do the Vocabulary Upgrade as you suggested and eventually I have all the voice words back and keypads now have their text too.  I did get "Comms failure" SMS messages during this process, but I'm sure that happened to me back in 2020 when I last lost the voices after some electrical brown-outs.

Although almost everything seemed to be working I then re-uploaded my last config to Comfort anyway.

However, I still have some unexpected issues. In order of importance:

1. Whenever the system restarts I immediately receive an SMS from the GSM module saying there is a comms failure, and then once Comfort starts it alarms with Slave 1 Communication Failure. I sign-in and then the slave failure goes away after a few seconds.  This has been happening since the night of the thunderstorm. It never seems to lose comms again, only on restart.
   
2. I reenabled voicemail by recording a greeting. Dialling the landline allows me to leave a voicemail as normal. But when dialling the GSM line and leaving a voicemail (something I might not have done for years as I only ever usually sign-in with that)  gives a horrible howling/feedback sound when it starts recording.
   
3. I did a Scan for UCMs as the new UCM ETH03 has newer firmware than my old one. It says the type is "General" and the Submodule is "unknown". Is that normal?
   

Actually, there's a possibility that issue number 2 might possibly be related to me trying to tune some of the parameters last year to get DTMF sign-in working on the GSM module. So if that doesn't ring any bells I'll have to look at my old notes to see what I changed.

Number 1 is the worrying issue - why would it now repeatedly just have comms failure for ~10 seconds after every restart?

[slychiu]

Good to hear about the eth03 working again.
The initial slave comms failure may be due to timing. it takes some time for the comfort to poll the slave and geta a reply si there is a comms failure in the meantime. There may also be an isue with the Slave if this did not happen before the lightning strike - even though its zones reported ok, the RS485 may be partly damaged so the comms in intermittent

UCM type is general and submodule type is  unknown  ie no special type - that is normal

the voice recordsing isue could be due to parameters as you suspect

[caravanboy]

The initial slave comms failure may be due to timing. it takes some time for the comfort to poll the slave and geta a reply si there is a comms failure in the meantime. There may also be an isue with the Slave if this did not happen before the lightning strike - even though its zones reported ok, the RS485 may be partly damaged so the comms in intermittent

This comms issue definitely never used to happen this before the recent problems. The lightning strike wasn't even direct on our property, it was within a mile though, but I guess it caused some sort of surge on the local electricity grid.

Given that after the restart it all appears okay - no more alarms or warning messages, is there anything else I can do to check the slave's functionality?

I have a python script monitoring the Comfort messages (to integrate with my Home Assistant system) so if there is a specific message the Slave would output every so often I could extend it to log those and verify there are no gaps etc?

If the RS485 part of the Slave is damaged as you suggest, can that be replaced or would I have to buy a new slave panel?