"Serious problem detected"

[SDeath 68]

Yes this is definitely still happening, had 3 of these in the last 6 days.
All logs send to the developers.

[Neo Andersson 230]

On 5/24/2022 at 5:15 PM, m.roszak said:

Forwarded the logs to the developers, they are checking them now.

Hello Micahel, sorry for bothering you. May I ask, will this error solved in next updates? It happened again at our new cusomters house..Really bad. Right one week after a big installation...Can you please give us some updated info about this?

[petergebruers 1,268]

Just noticed in myj log, "The system recovered from an unexpected error" 7 days ago, running beta 5.112 since it came out.

Edited August 29, 2022 by petergebruers

[ChristianSogaard 71]

My test HC3 with no scenes and only one QA

Please login or register to see this attachment.

[SDeath 68]

Had it twice yesterday on the same day. ??

[petergebruers 1,268]

Thank you for reporting back. Three users with this issue is not a lot, but maybe some HC3 owners have not yet "seen" it happen. It is unfortunate in any case...

[henri-allauch-13190 10]

I too reported it and opened several tickets with dump.
Answer: we are on the problem
Very random bug sometimes every day sometimes 10 days without problem without modification of codes and devices??

[Alex 32]

I have the same problem since 19 days ago and since than it did occur 5 times (time window between differs from one day up to 8 days)

 

[SDeath 68]

Twice on the same day now, 6 hours apart. ?

[ChristianSogaard 71]

this is the error that keeps on giving

Please login or register to see this attachment.

[SDeath 68]

9 times in 30 days. ?

Causing my alarm to go of now two nights in a row.

[marcwau 30]

Update after update, this problem of "The system recovered from an unexpected error" apparently does not get resolved. I started this thread when we were at version 5.092; I am now on 5.120 and it still happens, many betas and releases later.

 

This serious bug is never mentioned in the release notes under the "Known problems" although Fibaro knows about it for a year and has promised several times that this issue "has the highest priority".... After a year, that sounds less and less credible.

And Fibaro even said that they found the root cause. If that is true, why is it still not fixed?

 

Or is it unreasonable to ask for a system that runs your whole house to be stable and reliable? Instead of crashing several times a month?

I would really like to see some status update from Fibaro on this matter. 

Edited October 12, 2022 by marcwau

[SDeath 68]

Yes the bug is a bit of a mess, luckily the system comes back after about 6 minutes. Nonetheless it is a very annoying bug if you want to disable the alarm right at that moment. For me it seems to happen more often now even on the last versions. Happens a few times a week now. 
and lasts for about 6 minutes every time. 
 

I’ve had PM that they where trying different things to get it resolved but they where not sure if the solutions would work or not. 
They can trigger the issue in their own devices but this has yet to be solved, hopefully sooner than later. 
 

Maybe @m.roszak can provide an update for us?

[henri-allauch-13190 10]

Personally I had stayed in version 5.110.15 because I had 24 days without a crash. I did not change anything while I was planting 1 to 2 times a week.

The only thing different is like no code modification, no backup...

Last night I had a general power outage, I put the hc3 to sleep while it was still powered by UPS, but when I restarted, the lights were OK, but impossible to access the hc3 from the dashboard . API calls were also OUT.

I restarted after removing the power supply and putting it back on. I will update to 5.120.10 this weekend.

 

 

Please login or register to see this attachment.

 

 

Please login or register to see this attachment.

 

Edited October 14, 2022 by henri-allauch-13190

[henri-allauch-13190 10]

With 5.120.10 installed : Already 4 reboots 

Still not resolved

How to explain 24 days without reboot followed by 1 reboot per day Without modification?

 

[SDeath 68]

5.130.64 still not solved.

Edited November 30, 2022 by SDeath

[juwi2 2]

I am new to HC3 and started with Firmware 5.120.10 on 07.11.22. I had no error for 3 weeks of use (~80 Zwave devices, ~40 LUA scenes).

On 25.11.22 I updated to Firmware 5.130.64 and got this error today 2 times (4 h between the errors) on 07.12.22 after 1 1/2 week of use.

How do I create full logs from HC3 interface?
Thank you in advance.

 

Please login or register to see this attachment.

Edited December 7, 2022 by juwi2

[SDeath 68]

Logs are not needed, they know what causes the issue and can reproduce it on their devices.

Somehow it's taking them ages to fix this very annoying and frustrating error.
The system is basically down for at least 5-10 minutes, causing al kinds of safety issue, because during that time none of my alarms work, only the device itself will sound if it has a siren built-in.
Still baffles me how long it takes to get this fixed, and seems to be a very low priority for Fibaro.

 

It's been months since the last update on this.

But for the future you can fetch the logs by entering this in a browser and login with the credentials:
http://<FIBARO_HC3_IP>/api/service/logsDump

 

11 minutes ago, juwi2 said:

I am new to HC3 and started with Firmware 5.120.10 on 07.11.22. I had no error for 3 weeks of use (~80 Zwave devices, ~40 LUA scenes).

On 25.11.22 I updated to Firmware 5.130.64 and got this error today 2 times (4 h between the errors) on 07.12.22 after 1 1/2 week of use.

How do I create full logs from HC3 interface?
Thank you in advance.

 

Please login or register to see this attachment.

 

[juwi2 2]

@ 

Please login or register to see this link.

:

I could imagine to eliminate the problem with the following workaround. Make an automated restart of the HC3 every day (at night).

But this has to be tested and it is still a workaround to prevent the spontaneous restarts of the Fibaro HC3.

Create a lua scene:

Condition:

Every Night @ 02.00 am.
{
    type = "date",
    property = "cron",
    operator = "match",
    value = {"00", "02", "*", "*", "*", "*"},
    isTrigger = true
}

 

LUA Scene itself:
 

-- Reboot system

local currentTime = os.date("%d.%m.%Y %H:%M Uhr");

local phoneID = {tonumber(fibaro.getGlobalVariable("Smartphone_XYZ"))} -- Handy-IDs

local ventile = {x,y,z} -- IDs from stuff to be shut off before the system reboot (in my case irrigation valves)

 

if 

 ( tonumber(fibaro.getGlobalVariable("Urlaub")) == tonumber("0") ) // if you use an present simulation this check is needed. You can remove the if then passage end the else bracket if not needed.

 then

  

    --[[ To prevent a irrigation cycle from being started before the reboot, the valves are switched off.--]]

    for i = 1,#ventile do  

        if 

        ( (fibaro.getValue(ventile[i], "value")) == true )

            then

            fibaro.call(ventile[i], "turnOff");

        end     

    end

    --[[ All valves switched off --]] 

 

    --[[ Notification to Smart Phones @ phoneID]]--

    print(phoneID);

    fibaro.alert('push', phoneID, "Fibaro HC3 was rebootet on "..currentTime.." by this LUA scene.");

 

    fibaro.sleep(20*1000);

    fibaro.debug("Restart...");

    fibaro.homeCenter.systemService.reboot()

 

 else

  fibaro.debug("House is in secured vacation mode (Present Simulation) and is not restarted because of the presence simulation!");

end

Edited December 8, 2022 by juwi2

[SDeath 68]

That doesn't seem to matter, I've had one of these restarts only a few hours after a reboot.
So it doesn't really help in any way IMHO.
And also still causes a downtime of 5 minutes while the HC3 reboots and causes the same safety issue, but then every night, while we are sleeping.
Not the best time for a less protected environment.

 

Fibaro just needs to get their act together and fix this issue, they know what's causing it, just fix it.