-
-
Notifications
You must be signed in to change notification settings - Fork 1.6k
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
All stops working after (seemingly) heavy control #1094
Comments
@glentakahashi and @sylarevan hit the same problem. With also some possible solution: #764 (comment) Let's use this issue to tackle this problem.
I checked with my own network. It seems to get unreliable when more than 6 group commands are send. This seems to be the same as what @glentakahashi experiences |
I've made some changes and got good results. Here is a log where I send 40 To test this:
|
Very cool! I will test this this weekend as well. @Koenkk did you also find that your bulbs were being responsive to the group commands as well? Even when I got to the point where i could send 2 requests a second without the CC2531 hitting limits, my bulbs didn't respond to requests anymore. |
@glentakahashi they indeed stop responding after a few commands, however I think, apart from adding extra delays for group controls, there is not much to do about this. |
Yea that makes sense. It seems some bulbs don’t have the power to support
lots of group requests which is unfortunate.
…On Fri, Feb 15, 2019 at 14:45 Koen Kanters ***@***.***> wrote:
@glentakahashi <https://github.com/glentakahashi> they indeed stop
responding after a few commands, however I think, apart from adding extra
delays for group controls, there is not much to do about this.
—
You are receiving this because you were mentioned.
Reply to this email directly, view it on GitHub
<#1094 (comment)>,
or mute the thread
<https://github.com/notifications/unsubscribe-auth/AA-bL5_JWtRkiuUPCPcA1wpWl9oXllTLks5vNw5CgaJpZM4a8vKb>
.
|
I have great results too! This is my log where I sent 40 messages in one second with just hue bulbs in a group: https://hastebin.com/anenoleleb.cs And this is 40 messages in one second with hue bulbs and a tradfri bulb: https://hastebin.com/jobamunewu.cs @glentakahashi See this comparison for how a hue bulb and a tradfri bulb handle these messages: I'm very happy with these results as my dimmer is behaving perfectly with the hue bulbs: It indeed looks like the Tradfri bulbs can't handle the same speed of messages, because during dimming mine freezes frequently. This behavior is not limited to groups; it has trouble with fast (4 per second) direct messages as well, which the hues are fine with. Before the Hue's would be fine when addressed directly, but not in groups. This issue with hues in groups has been resolved for me! @Koenkk It seems like my bottlenecks on the zigbee2mqtt side are fixed, thanks a lot! |
I've verified that the TRADFRI bulbs can indeed not handle many commands per second. I've made a setup were I paired a TRADFRI bulb directly to the remote (so there is no hub/zigbee2mqtt between) and started spamming the bulb with commands. Around 00:10 you see that the TRADFRI bulb does not respond anymore, after a few seconds, it starts responding again. Video: https://www.youtube.com/watch?v=LK9Llve8Xl8 As @EmilFlach mentions, I think the problems on the zigbee2mqtt side are indeed solved. @glentakahashi please let me know your results. |
@Koenkk @EmilFlach Just tested and got the same results as you two. Zigbee2mqtt supports about 5 commands a second, but it's the bulbs that can't keep up. The Sengled bulbs actually support even fewer commands than the TRADFRI bulbs in your video. I can only hit about 10 commands before it stops responding. Hopefully bulbs in the future better support group commands! |
Good, then I think this issue can be closed? |
I think so, not much else to do from zigbee2mqtt side
…On Sun, Feb 17, 2019 at 15:18 Koen Kanters ***@***.***> wrote:
Good, then I think this issue can be closed?
—
You are receiving this because you were mentioned.
Reply to this email directly, view it on GitHub
<#1094 (comment)>,
or mute the thread
<https://github.com/notifications/unsubscribe-auth/AA-bL-6oQFQCV5NMMiTdgm9ReQoWhjJ2ks5vObkbgaJpZM4a8vKb>
.
|
I have recently setup zigbee2mqtt in Hassio and after some testing I find it a very promising project.
I do have one repeating issue:
seemingly after some heavy control, like turning on/off multiple lights at the same time (using light groups in Homeassistant), all control stops working. If I check the zigbee2mqtt log i get a lot of
zigbee2mqtt:error 2019-2-15 01:42:47 Zigbee publish to device '0x000b57fffe97bbb4', genOnOff - on - {} - {"manufSpec":0,"disDefaultRsp":0} - null failed with error Error: request timeout
When I do a network scan at this point (i use zigbee2mqtt_networkmap) the log only has
where usually this gives out a lot more information. When I check the graphical map, I still see al my devices, but all the connecting lines between the blocks are gone.
After replugging the cc2531 USB dongle and restarting zigbee2mqtt through Hass.io, everything works again.
Normally (when all is working) the network map looks like this:
![zigbee2mqtt_networkmap_okay_full](https://user-images.githubusercontent.com/30867746/52827856-c481a780-30c6-11e9-8809-119f1d3c8357.JPG)
My zigbee2mqtt configuration.yaml is pretty standard i think:
My database.db:
What I use:
Any idea what causes this? Is this a known issue?
Anyway, keep up the good work, this is really great stuff.
The text was updated successfully, but these errors were encountered: