Jump to content

Welcome to Smart Home Forum by FIBARO

Dear Guest,

 

as you can notice parts of Smart Home Forum by FIBARO is not available for you. You have to register in order to view all content and post in our community. Don't worry! Registration is a simple free process that requires minimal information for you to sign up. Become a part of of Smart Home Forum by FIBARO by creating an account.

 

As a member you can:

  •     Start new topics and reply to others
  •     Follow topics and users to get email updates
  •     Get your own profile page and make new friends
  •     Send personal messages
  •     ... and learn a lot about our system!

 

Regards,

Smart Home Forum by FIBARO Team


  • 0

dead nodes


Question

Posted

I have a problem with dead nodes. They follow me through every software version. Now I run on 4.058 Beta.

I tried the mesh network reconfiguration. I get tihs result:

 

[09:00:50] 476: Requesting neighbor list from the node is in progress
[09:01:10] 476: Getting new neighbor list failed

 

So what does it tell me? What can I do to solve this prob?

Thanks

MIB

Recommended Posts

  • 0
Posted

If it's a powered node, it's not good. If it's a battery powered node, it's normal (in a sense...) because the device is probably sleeping. You can wait until the wake-up interval passes or manually wake up the device. It should then log "New neighbor list received" after 10-20 seconds. I think you did a "full" mesh reconfiguration so I think it is now best to finish that (wake up all battery devices). But if that doesn't help... do it more systematically, one by one. A small drawing or an explanation of your network + placement of the dead nodes would be helpful.

  • 0
Posted

The dead node issues are all gone on my system.

Once you have implemented Peters suggestion I would look at the following :

1. Change the direction of the antenna and shift the positron of the box around

2. Upgrade the antenna

3.Buy a second controller

Depending on how bad your setup is ie distance between powered and battery devices you should have varying luck with the above.

With two boxes I have not had a single rogue dead node since implementation which is now a good 6 months ago.

  • 0
  • Inquirer
  • Posted

    Thanks for the quick reply.

    Here are some more Infos.

    Z-Wave Version: 3.67

    Number of devices: 93.

    Recommended devices polling time interval: 440s.

    I get for every reconfiguration attempt the failed answer.

    I wake up all the nodes every 30 minutes with a LUA script.

    About 10 devices are battery devices the rest are FIBEFGS-222, FIB_FGWPF-101, FIB_FGRM-222.

    Basement, ground floor, upper floor. Every light switch is equipped with FIBEFGS-222. Every window is equipped with FIB_FGRM-222. HC2 is placed on ground floor.

    Dead nodes change. Nothing is regular. Sometimes you get "transfer ok" then "transfer failed" and it turns into a dead node. I wake it up and after some attempts it works.

    I restart the HC2 every night with LUA script.

    I tried an AEON Labs DSD37 range extender. It made it worse. It is not in use anymore.

    I forced re-configure all non battery operated devices and worked two days to find out which device was where in the house.

    Please login or register to see this image.

    /emoticons/default_sad.png" alt=":(" srcset="https://forum.fibaro.com/uploads/emoticons/[email protected] 2x" width="20" height="20" />

    Any idea what I can do?

    MIB

    • 0
    Posted

    Have you tried walking around the house with a fibaro motion sensor to test the strength of your wireless network?

    • 0
  • Inquirer
  • Posted

    no that I have not done yet.

    Will give it a try and let you know the result.

    What makes me crazy is the change ...

    • 0
    Posted

    Ouch! Big that's a big network, but not so much different from mine. We definitely don't live in the same house

    Please login or register to see this image.

    /emoticons/default_icon_smile.gif" alt=":-)" /> so different environment.

    Is OK if I think aloud?

    If it affects a large part of the network, it might be:

    a) Reception / Transmission of the HC2

    b) Too much traffic on the Wireless

    c) Rogue devices

    d) Something wrong with the HC2, hardware or software

    For a) best practice keep HC2 0,5 meter away from other devices, walls, metal, wifi, cordless phone, ... Maybe you can find another place in the house, then reconfigure the mesh.

    b) Traffic can be cause by too much polling, devices at the edge of the network having to retransmit, devices that are set to report frequently. Examples of the last category are AEON labs smart switches or Wall plugs, dimmers reporting power. Also check RGBW set as "input". Scripts running in a tight loop sending the same command might also go unnoticed.

    c) It has been reported a few times but I don't think it's the most obvious cause. From personal experience: I had a Fibaro Door/Window sensor that would have an insanely low wake-up interval, causing delays on my network. I think a MultiSensor running with poor batteries might do odd things too (this device was outdoor).

    d) Something wrong with HC2 hardware... never read a post firmly stating that the radio failed, but theoretically possible. Software? Did the problem of the "mesh reconfiguration" start after a software update?

    The trouble with dead nodes, if you wake them up, is that the HC2 will attempt all possibilities to connect to the node. While it is doing that, the network is busy (if my knowledge of Z-Wave is correct, at least for 5 seconds if it's really dead). So a script that "revives" them may improve the impression of "availability" on the one hand, but causing "delays" on the other hand.

    I'm still thinking aloud... If it's possible, acceptable, I'd let the dead devices go dead. Then evaluate what's left and where they are.

    On the other hand... Pick one mains powered device, less that 5 meter from the HC2. One that doesn't go dead. Then try a "Mesh network reconfiguration" on that one only. If it fails, you would be the first one to report that. Also, like aleks suggested, you might be able to gain some insight by excluding and including a wall plug, a motion sensor or a smoke sensor then use it as a range detector.

    This explanation is in the documentation of the wall plug. But this "range tester" also exists in the smoke sensor and motion sensor. You can use it to probe the network:

    To test the range:

    1) Insert Plug into a mains socket,

    2) Press and hold the B button for 10-15 seconds until the LED ring glows violet,

    3) Release the B button,

    4) Press the B button briefly.

    5) LED ring signals Z-Wave network range - see below for signaling modes description.

    6) To exit range testing mode press the B button briefly. Plugs relay will not change its status.

    Z-Wave network range signaling modes:

    LED ring pulsing green - Wall Plug tries to establish direct connection with the main controller. If direct connection cannot be established, Plug will try routing connection with the main controller which will be signaled by LED ring pulsing yellow.

    LED ring glowing green - Wall Plug managed to establish a direct connection with the main controller.

    LED ring pulsing yellow - Wall Plug is trying to establish routed connection with the main cotroller, via other Z-Wave devices acting as signal repeaters.

    LED ring glowing yellow - Wall Plug managed to establish routed connection with the main controller. After 2 seconds the Plug will try again to establish a direct connection with the main controller, which will be signaled by LED ring pulsing green.

    LED ring pulsing violet - Wall Plug is located outside the Z-Wave network's range or the network is busy. Ultimately, failure to connect with the main controller will be signalled by LED ring pulsing red. After 2 seconds the Plug will again try to establish a direct connection with the main controller which will be signaled by LED ring blinking green.

    • 0
    Posted

    Where did you put your HC2 controller?

    I had similar problems but most of them were gone after I moved the controller from the cellar to the ground floor.

     

    Also update to version 4.056 helped a lot (before I had 3.59 or something like this).

    • 0
    Posted

    (...)I had similar problems but most of them were gone after I moved the controller from the cellar to the ground floor. (...)

    Yes, there is some science behind that. In 2008, MERTEN published a document with a table, giving estimates for signal attenuation, for different materials. For iron-reinforced concrete they say 30% to 90% of the signal can be lost... That would mean you would rely a lot on the closest modules to relay the messages, instead of having lots of direct connections. Also, walls appear thicker when you are not looking at them at a straight angle. Imagine you wanted to drill a hole at a 45 degrees angle. You would need a longer drill. Of course, signals can penetrate doors and bounce around.

    Out of curiosity... Were you able to confirm that by using a Fibaro device as a network probe?

    • 0
    Posted

    Well I have a 3 storey house like Mib, albeit with a larger installation, and until I installed a second HC2 I was always getting dead nodes, so many of them that it was hard to understand what was causing it.

     

    Moving the box improved the situation and installing a longer antenna also did some, but the second controller eliminated every single dead node. 

     

    Perhaps since then the software has improved and it may be possible for my network to function on a single HC2, but I'm not going to try it out any time soon

    Please login or register to see this image.

    /emoticons/default_smile.png" alt=":)" srcset="https://forum.fibaro.com/uploads/emoticons/[email protected] 2x" width="20" height="20" />

     

    That said, I'd still try all of the tips suggested on this thread before you go and fork out more cash on a 2nd controller.

     

    PS: Thanks for providing the details on the zwave range signalling modes Peter. I was replying from my phone before so couldn't get to the info.

    • 0
  • Inquirer
  • Posted

    OK, I plugged a brand new FIB_FGWPF-101 just beneath my HC2 into a main socket and made the range test.

    Result: LED ring is pulsing violet then red then green again.

     

    So is my mesh network down?

    • 0
    Posted

    I think the order is: pulsing green, pulsing red, pulsing violet, repeat. Meaning, try direct connection and fail, try nearest neighbor and fail, repeat.

    Can you control this plug from your HC2?

    • 0
    Posted

    I just realized that the plug might behave like this when it is not included. The probe uses something similar to a ping (echo) packet, so it has to be part of yuor network.

    • 0
    Posted

    I just realized that the plug might behave like this when it is not included. The probe uses something similar to a ping (echo) packet, so it has to be part of your network.

    • 0
    Posted

    Note to self: don't tap twice on that 'post' button

    Please login or register to see this image.

    /emoticons/default_icon_smile.gif" alt=":-)" />

    • 0
  • Inquirer
  • Posted

    OK I will include it

    • 0
    Posted

    I see. If you look at the instructions, it is not immediately obvious, isn't it? I have to remember that if I post those instructions again! Sorry, my bad!

    • 0
  • Inquirer
  • Posted

    I can turn it on and off via browser.

    I started the test.

    green pulsing (9 seconds) - purple pulsing (7 seconds) - permanent red (2 seconds)

    then it starts again with green ....

    • 0
    Posted

    If I do that, at 5 or 10 m distance (without walls) - the indicator is solid green, as expected.

    I'm... speechless. I need to think...

    • 0
    Posted

    The fact that you can't reconfigure your network agrees with the fact that probing fails. Yet you can switch on/off that wall plug. You get dead nodes, consistent with network failures... But you can "revive" them and use them. I can't think of a defect that explains all this.

    Is it possible that you have another device using the same frequency? That is: 868,42 MHz in Europe. I have seen cordless headphones and baby monitors using this...

    Do you have a smoke sensor or a motion sensor, to confirm the behavior of the wall plug?

    Would it be feasible and not too much trouble to put your HC2 several meters away from its current location? I don't think you need a network connection for this test. Then probe the network?

    • 0
  • Inquirer
  • Posted

    Me either. I'm really without any idea what is wrong. If I would not have spent such a lot of money on that project belive me I would have thrown out.

    My wife is mad at the stuff because sometimes lights go on at night in the bedroom. That is real fun.

    Well I use iPhone, iPad but no wireless mouse, headphone, baby monitors.

    Smoke Sensor is wired one on burglar alarm.

    Motion Sensors I have some but their beauvoir is weird anyway.

    The HC2 is now located near my desktop in my office on ground floor. No prob to move it around. I've got enough LAN connection in the new house.

    Join the conversation

    You can post now and register later. If you have an account, sign in now to post with your account.

    Guest
    Answer this question...

    ×   Pasted as rich text.   Paste as plain text instead

      Only 75 emoji are allowed.

    ×   Your link has been automatically embedded.   Display as a link instead

    ×   Your previous content has been restored.   Clear editor

    ×   You cannot paste images directly. Upload or insert images from URL.

    ×
    ×
    • Create New...