HC2 remote access died and 2 days later the whole system stopped. Support gave not acceptable answers

[Voordeh 1]

Hi,

I have bought and installed the HC2 to to keep major things running in my house when I'm absent for longer periods.

Things like Heating, Security, Some lights and some monitoring (water usage, electricity usage etc)

You would expect that an Security system would be so solid that it would run always.

I have been 3 months in the house, and the system was running without any problem. (actually the system had been running for over a year without problems)

After I left the house for a stay of 3 months in an other country the system continued to work for a week sending push messages etc.

After this I all of a sudden couldn't get remote access to the system. However by my outside cameras I could see that timers etc. (Outside lights) were still working

Two days later the whome system stopped working. Lights stay on day and night.

 

[SmartHomeEddy 1,005]

1 hour ago, Voordeh said:

I have bought and installed the HC2 to to keep major things running in my house when I'm absent for longer periods.

Things like Heating, Security, Some lights and some monitoring (water usage, electricity usage etc)

You would expect that an Security system would be so solid that it would run always.

 

I understand what you want it to do, but the Fibaro Homecenter isn't a Security System in terms of reliability. I use "classic" security systems for alarm, video, etcetera. and use output from those systems in my Homecenter. 

 

And your HC2 also depends on a reliable network and internet connection. 

 

Maybe other members could advise how te setup a Homecenter (in your case HC2) as reliable as can be, to get through months of absence. 

 

 

 

[Voordeh 1]

When I reported this to the custommer support the solution the gave me was:

Please reboot the HC, additionally delete it from the Fibaro ID and add again, 

However how do you reboot a system from 4000 km away ???? Need a second system to reboot the first system????

When I wrote this to them I only got the answer:

Sorry I can't do more.

 

So my question is, Is this normal for the Fibaro systems, In that case I have to trow it out of the window and see that I get a robust system from a company who care about it's customers.

 

 

[Voordeh 1]

12 minutes ago, SmartHomeEddy said:

 

I understand what you want it to do, but the Fibaro Homecenter isn't a Security System in terms of reliability. I use "classic" security systems for alarm, video, etcetera. and use output from those systems in my Homecenter. 

 

And your HC2 also depends on a reliable network and internet connection. 

 

Maybe other members could advise how te setup a Homecenter (in your case HC2) as reliable as can be, to get through months of absence. 

 

 

 

The network in this case is not the problem, I use a double connection with fallback. And I can see that both lines are up. Hope that someone comes with a solution of auto re-booting after detecting malfunction. Thanks for your help.

[SmartHomeEddy 1,005]

12 minutes ago, Voordeh said:

The network in this case is not the problem, I use a double connection with fallback. And I can see that both lines are up. 

???

 

For example I saw that @tinman did a review of a logfile from an other user and saw some "faults" that for example could cause memory leaks or increasing CPU usage. You don't want that when you are months away. Maybe a refresh of the system by an installer like tinman could help you through the months. 

 

 

 

Edited December 9, 2020 by SmartHomeEddy

[Tim__ 126]

Support's suggestion to reboot the hung device sounds perfectly fine to me.

 

Another user posted on here recently that they use a remote power switch, that they can use to power cycle the home centre remotely via IP. Might be useful in your case?

[tinman 1,070]

 

8 hours ago, SmartHomeEddy said:

I understand what you want it to do, but the Fibaro Homecenter isn't a Security System in terms of reliability.

 

HC2 was once certified EN50130/1 grad III or IV, that was in 2012 however (and of course specific setup). Generally when using Fibaro HC2 or HC3, i speak about alarming system - but not about security (alarm) system. There are other solutions which works togeher with Fibaro Home Centers (e.g. Satel).

 

41 minutes ago, Tim__ said:

Support's suggestion to reboot the hung device sounds perfectly fine to me.

 

generally HC2 run stable enought to stay online for long time, year and more without reboot is not problem on healthy HC. However, there might be always combination of side effects making the HC unstable (enought) to not execute watchdog scripts (they running middle of the night on HCL/HC2 and HC3 - which additionally is using different system healty check).

 

You can setup reboot on the HC with VD (or QA on HC3) or LUA scene at every week/month/day - what so ever fit your setup. I know some customers set reboot each every night (HCx), honestly this is definitely not good idea because it might confict with cron jobs. Good time for reboot is 1AM till 1:30AM, 5AM till 6AM, definitely no go to reboot at midnight (database cleanup cron job).

[michal85pl 15]

15 godzin temu, tinman napisał:

You can setup reboot on the HC with VD (or QA on HC3) or LUA scene at every week/month/day - what so ever fit your setup. I know some customers set reboot each every night (HCx), honestly this is definitely not good idea because it might confict with cron jobs. Good time for reboot is 1AM till 1:30AM, 5AM till 6AM, definitely no go to reboot at midnight (database cleanup cron job).

Has anyone created such QA for HC3 and would like to share?

[Voordeh 1]

On 12/9/2020 at 9:53 PM, tinman said:

 

 

HC2 was once certified EN50130/1 grad III or IV, that was in 2012 however (and of course specific setup). Generally when using Fibaro HC2 or HC3, i speak about alarming system - but not about security (alarm) system. There are other solutions which works togeher with Fibaro Home Centers (e.g. Satel).

 

 

generally HC2 run stable enought to stay online for long time, year and more without reboot is not problem on healthy HC. However, there might be always combination of side effects making the HC unstable (enought) to not execute watchdog scripts (they running middle of the night on HCL/HC2 and HC3 - which additionally is using different system healty check).

 

You can setup reboot on the HC with VD (or QA on HC3) or LUA scene at every week/month/day - what so ever fit your setup. I know some customers set reboot each every night (HCx), honestly this is definitely not good idea because it might confict with cron jobs. Good time for reboot is 1AM till 1:30AM, 5AM till 6AM, definitely no go to reboot at midnight (database cleanup cron job).

Thank you Tinman (and others)

I ask myself why the support people don't come up with this type of solutions. I will certainly make a program to reboot (on a regualar basis). Do you know if debug statements in lua code could course a memory problem if I don't comment them out?

[tinman 1,070]

On 12/13/2020 at 11:36 AM, Voordeh said:

Do you know if debug statements in lua code could course a memory problem if I don't comment them out?

 

debug commands itself are - like the name says - for debug, not for "daily visualization".  There is huge chance to slow down HC2 with debug messages, and definitely having lot of them, coloured, html formated - what so ever not really necessary things for daily use - eats much more resources and they might (due to the fact, that they not anymore LUA at that point) run out of control and memory. 

[Voordeh 1]

19 hours ago, tinman said:

 

debug commands itself are - like the name says - for debug, not for "daily visualization".  There is huge chance to slow down HC2 with debug messages, and definitely having lot of them, coloured, html formated - what so ever not really necessary things for daily use - eats much more resources and they might (due to the fact, that they not anymore LUA at that point) run out of control and memory. 

Thanks Tinman,

 

I think I left a couple of them (which i was using for test purpose and forgot to take them out) I hope that after a power down and power up everything will be back and I can take them out.

[druqs 1]

On 12/9/2020 at 8:53 PM, tinman said:

HC2 was once certified EN50130/1 grad III or IV

This seems an odd exception to certification requirements as it has no native battery backup? One of the reasons we hated certain aspects of EN50130 so much here in the UK was what we saw as the over-specification of battery back up on the CIE, for durations of power failure which are rare here in the UK. My Eaton (Scantronic) i-On 160 En50130 G3 alarm system, first designed in the early pre-PD6662 days but updated many times since has space in the cabinet for TWO 17aH batteries as a result.

 

Also this would have been the first all-wireless Grade 3 panel I'm aware of and the first Grade 4 panel full-stop I've ever seen marketed especially to Joe Public.

 

Are you sure about this claim - it certainly is interesting - is there anything written down anywhere? Now "Grade 1" I can believe - ungraded seems more likely!

 

I'm with @SmartHomeEddy on this, keep the alarm separate from a well known specialist manufacturer, with and just use basic 'hooks' into HC, e.g. a scene to run when the alarm goes off,

auto part-set at night etc...