Yet Another lastWorkingRoute visualiser

[10der 830]

Please stop forcing the use this app!

Edited November 20, 2019 by 10der

[RH_Dreambox 179]

No, the new load.php with load.settings does not work for me. The rooms and devices files are not created, but I can still create them manually.
What I note when I create them manually is that I have about 3-4 seconds delay before the login dialog pops up. Can this delay cause the files to not be created?
I begin to think I have problems with the internal network

 

load.settings

Please login or register to see this code.

 

Please login or register to see this attachment.

[10der 830]

2 minutes ago, RH_Dreambox said:

No, the new load.php with load.settings does not work for me. The rooms and devices files are not created, but I can still create them manually.
What I note when I create them manually is that I have about 3-4 seconds delay before the login dialog pops up. Can this delay cause the files to not be created?
I begin to think I have problems with the internal network

 

load.settings

Please login or register to see this code.

 

Please login or register to see this attachment.

 

please call directly load.php and read screen error!

 

for example, i am a simulated error with accessing to HC2

 

[RH_Dreambox 179]

I have also tried to pause Kaspersky Internet Security, but it will be the same.

 

 

Please login or register to see this attachment.

[10der 830]

@RH_Dreambox PM

[AndyY 11]

10 hours ago, 10der said:

about color - yes, propose me color  cuz deciding what color best it’s pain for me, lol

Hmm, difficult decision . How about yellow?

What comes to my mind, if we use a color for older LWR datetime, we loose other colors. Whould it be possible to use present color with a bright hue? NORMAL -> BRIGHT

If that gets too complicated to implement we can take yellow. Thanks for your great efford!

[10der 830]

15 minutes ago, AndyY said:

Hmm, difficult decision . How about yellow?

What comes to my mind, if we use a color for older LWR datetime, we loose other colors. Whould it be possible to use present color with a bright hue? NORMAL -> BRIGHT

If that gets too complicated to implement we can take yellow. Thanks for your great efford!

 

check box like  [X] Show LWR 

I think it can solve our problems :)

[AndyY 11]

On 9/22/2018 at 8:26 PM, 10der said:

check box like  [X] Show LWR 

I think it can solve our problems :)

Hello 10der, did you do already something? No problem at all . I don't know much about js but I did some copy&paste and try&error in your index.html . Result:

Maybe not perfect but it works for me. If anyone is interested try attached file index2.html. I only added 3 lines. Hope this is ok @10der

Please login or register to see this attachment.

[Momos 101]

Since 3 days ago my network stopped doing the mesh reconfiguration every night.  I am not awake at that time to actually look at the status box, but judging after the result of this script,  all were last updated on the 25th.

 

Anyone else experienced this ?

 

[10der 830]

45 minutes ago, Momos said:

my network stopped doing the mesh reconfiguration every night

Fibaro Skynet is dead, you won!  

Many users on this forum dreaming about turning off this feature....

But if be honestly idk what happened.

[barend121 13]

Mine updated this night again.. and have some very weird results.... Non-logical routes and a few nodes never got an update at all (purple)  6 of 64 devices are purple..

[hansib 2]

Just FYI, it works on mac as well on chrome and safari

I downloaded the 2 files on page 1, updated with my credentials/IP

saved the 2 files, devices.json and rooms.json from the apis: 

http://<IP>/api/rooms

http://<IP>/api/devices

Started a simple HTTP server in the downloads-directory with: python -m SimpleHTTPServer 8000

Then, by going to 0.0.0.0:8000 you will get the index.html-file.

 

Excellent!

Edited October 3, 2018 by hansib

[Tony270570 92]

thank you  @10der thanks to you i've got a nice spaghetti soup  

Less funny if I loose 2 of my devices I'm "loosing" half of the network. I don't understand why so much devices are using those 2 nodes to build the routes.

I've got 50 extra devices splitter between  2 other machines, both are slaves of the machine on the graph displayed. Didn't had a look to those networks yet ... 

Any idea doctor ?

 

Please login or register to see this spoiler.

 

Edited October 4, 2018 by Tony270570

[10der 830]

@Tony270570 wow. You're right, this is pretty magical.

but I am not a doctor  But I know who is doc here  @petergebruers

thoughts aloud: turn off power on one of.. and you lose :) look like crazzzy!

[Tony270570 92]

Don't know if magical is the appropriate word  Not want to bother @petergebruers he already spent lot of time for the community but if somebody from the @FIBARO team wants to have a look inside the beast, i'll let them in. 

This weekend I'll physically disconnect the 2 overloaded devices and force to rebuild the entier network , then read the result, I'll then reconnect the 2 devices and rebuild the neighborhood only for these 2. I'll then check if it remains more "logical" after the daily automatic network rebuild.

To be followed.

Thank you @10der as usual nice job

 

Edited October 3, 2018 by Tony270570

[petergebruers 1,261]

 

No no no no.

 

1) This is not TCP/IP routing. It is Z-Wave routing. If it *was* IP routing then loss of those devices *might* be a catastrophe (unless you have special routing protocols and a backup router).

2) This is only the last working route. Please check text output of my script, the devices have plenty of alternatives

 

For example:

 

Please login or register to see this code.

 

As you can see, LWR of device 86 indeed contains "4" as a hop.

 

Device 4 is acting as a router for 14 other devices as well. (the data is slightly different, because this report says "2018-09-04").

 

But this part "# 32 [1,4,1090..." means device 86 has 32 (thirty two) neighbors! It is not an orphaned node.

 

Also, device 4 is *not a bottleneck* because in Z-Wave only one device can talk at once... All communication happens sequentially. When the radio channel is busy, no other device can talk.

 

If you power down those (magic) "hub" nodes before 03:00 and turn them back on after 04:00 the HC will send new routes to all devices. Interesting experiment to do...

 

In fact, you can turn them of right now. All impacted devices will then select a different route, but you cannot see that on this visualizer, unless you run neighbor update on an impacted node (to update neighbors, remove node with device id 4, get new routes and update the LWR data on the HC). You might notice a "glitch" but it should not break down.

 

EDIT:

 

I'll keep repeating this until it sinks in... The ROUTER is the SENDING device. The devices in between, the hops, are "repeaters". They make no routing decisions at all. Z-Wave is source routed. This means: Node 86 asks node 4 to send data to the controller. The decision to use node 4 is made by the sending node.  Node 4 dutifully accepts the packet from the sending devices (86) and forwards (repeats) it to the controller.

 

Even if you have 4 hops... It is the sending node that decides which nodes to use.

 

 

Edited October 3, 2018 by petergebruers

[Tony270570 92]

no no no no & no ! you said that you slow down !! 

1 hour ago, petergebruers said:

1) This is not TCP/IP routing. It is Z-Wave routing. If it *was* IP routing then loss of those devices *might* be a catastrophe (unless you have special routing protocols and a backup router).

This is why I use " " when I wrote "loosing"

 

1 hour ago, petergebruers said:

Also, device 4 is *not a bottleneck* because in Z-Wave only one device can talk at once... All communication happens sequentially. When the radio channel is busy, no other device can talk.

So is it correct if I say that having one device as a repeater for 32 other devices might cause some delay if several devices starts taking at the same time? could we consider such situation might become in some cases a bottleneck ?

 

EDIT  : OK sorry understood it is exactly the same in direct communication with the HC. Extra question : is a simple device has the same capacity as the main router to handle a stack of messages ? if I understand you well I'd say yes, but do I understand you well ?

 

 

1 hour ago, petergebruers said:

If you power down those (magic) "hub" nodes before 03:00 and turn them back on after 04:00 the HC will send new routes to all devices. Interesting experiment to do...

This is exactly what I'm expecting, I'll check that.

1 hour ago, petergebruers said:

but you cannot see that on this visualizer, unless you run neighbor update on an impacted node

Correct ! I said i'll force rebuild the network for those 2 devices and I was wrong I need to rebuild the entire network instead. Thank you Peter for the clarification !!

1 hour ago, petergebruers said:

I'll keep repeating this until it sinks in... The ROUTER is the SENDING device. The devices in between, the hops, are "repeaters". They make no routing decisions at all. Z-Wave is source routed. This means: Node 86 asks node 4 to send data to the controller. The decision to use node 4 is made by the sending node.  Node 4 dutifully accepts the packet from the sending devices (86) and forwards (repeats) it to the controller.

 

Even if you have 4 hops... It is the sending node that decides which nodes to use.

Well registered and understood  but why node 4 is the center of the world in this house ?? it's really place less then 2 meter from the HC2 does it mean that I need to move the HC2 to get a better communication with all those device ? I'll also try this, I've got a stronger antenna and I'll connect it tonight !!! 

 

Peter thank you X 1000 for your lights ...

ORA-03113 

 

 

Edited October 3, 2018 by Tony270570

[10der 830]

7 hours ago, Tony270570 said:

ORA-03113 

End of file communication ? I do not remember now, cuz more than 10y now I am MS SQL DBA  but I clearly remember Oracle SQL

 

And ha ha if I right guess about your nick ... you are one month older than me

Totally off topic here

Edited October 4, 2018 by 10der

[Tony270570 92]

You're seeing right ! Let me guess you're from Jun  the EOF was addressed to @petergebruers, he's spending so much time helping people and I don't want giving him extra work, but I know I has understood the message  

off topic closed

 

[petergebruers 1,261]

ORA EOF

 

 

Give me some time to formulate some answers. And while you're waiting, you can play Another exciting game of Clusenix

 

Please login or register to see this spoiler.

 

[Tony270570 92]