Daikin integration fills logs with 'took longer than the scheduled update interval' #82972
Comments
|
Hey there @fredrike, mind taking a look at this issue as it has been labeled with an integration ( Code owner commandsCode owners of
(message by CodeOwnersMention) daikin documentation |
|
Same issue here Running Home Assistant 2022.12.7 in Docker. |
|
same problem. the controller is answering but very slowly even id the wifi signal is good 2023-03-09 10:15:18.483 WARNING (MainThread) [homeassistant.components.daikin] Connection failed for 192.168.1.162 Home Assistant 2023.4.1 |
|
@fredrike would you mind changing the update interval to, for instance, 60 seconds instead of the standard 30 seconds? |
|
Decreasing the update interval is only a band-aid workaround. Who knows if it wouldn't still take over 60 seconds. |
Any suggestions on how we can try to identify what is causing the delay? I refuse to blame the Wifi. |
|
I don't know. Maybe enabling more logging levels would help? @fredrike would have to answer this. |
|
It's being addressed upstream in pydaikin: https://bitbucket.org/mustang51/pydaikin/pull-requests/55 |
|
Thanks @Djelibeybi, I'll create a new new version from bitbucket and ping you when it is ready. |
|
@Djelibeybi @mover85 #94214 has been closed. Will there be a new PR to get that fix in HA? |
|
Yes there will be a new PR, just waiting for a new version of pydaikin. I'm testing it on the forum if anyone wants to test it. https://community.home-assistant.io/t/daikin-airbase-testing-new-version-of-pydaikin/584104/41 |
|
Should this be fixed by #95656? |
|
Ok, no, it isn't fixed. I still get the warnings in the log with HA 2023.7.2: |
|
If you ping the controller is the latency consistent? |
|
If I ping directly from Home Assistant to the AC then the latency is mostly at around 1 ms. There are some outliers around 10 ms and also some around 100 ms. But never more than about 150 ms. Here are the ping statistics for 10000 pings: |
|
That seems fine, it might be a timeout not being triggered for some reason. |
|
Seems like this also affects HA startup time. I got the following in my log after the last startup of HA 2023.7.3: It was by far the slowest integration to start: |
|
Just a me too.... |
@Djelibeybi I'm now using HA 2023.9.2 which contains pydaikin 2.11.1 and I still get those warnings in the log. So it seems like that PR didn't fix the issue after all. |
|
I also have some unifi AP. By default, when power is set to « auto », it’s more or less equal to max power. When I turned my 2.4 ghz network power from auto to low, I get far less of those timeouts. Just found out. Will confirm (or not) after more days but sounds promising. Can someone with a similar setup test that too? |
Just made the same changes. |
|
Does not fix anything over here. |
|
Yeah after some days, not really better over here either. Sorry |
|
Does anyone have any further suggestions/comments regarding this ongoing issue? |
|
Same issue for me |
|
I have a weird question: do we all have a Unifi Wifi network here? Here is a typical log at my place (I filtered out only the Daikin part): |
|
I have 3 Unify APs, filled my log also. Turned off the mesh setting on the APs, improved a bit. Now aprox 2-3 entrys in the log per day instead of 30-40 |
No |
|
I have 4 Daikin units using this integration, and a Unifi network. My HA instance has been up for 5 days, and there are 37,835 “longer than 10 seconds” in the log! They cover all the Daikin units and multiple entities in each. The Daikin Mobile Controller app also takes a long time to refresh any status, so I figured slow response is “normal”. I’d really love for this timeout to be extended to stop all these messages! |
|
Hi, |
|
Anyone has got any solution to this? This is the only regular error / warning message left that I can't resolve in my HA setup. |
|
I’d really love for something to be done to address this. Maybe a configurable timeout interval? But 10 seconds is clearly too short for many systems. The message has occurred in my current log 27,518 times. And yet all my air conditioners work flawlessly in HA. If not configurable, maybe a one minute timeout would suffice. I can’t see a downside really. |
|
Many decide to use Faikin to remedy this situation if I understood
properly. Otherwise I removed all cloud dependancies by replacing the std
wifi chip by an older, lan-only, version.
I’ve few timeouts. Like on 8 splits, overall maybe 12 per day.
I isolated the splits on a wifi net too, a relaxed, 802.11b old stuff iot
compatible and got my decent devices on another network.
You can also put manually delays in your automations. As well, I avoid
triggering automation at fix hours since I suspect this the moment where
stats are polled by the system already.
…On Wed 27 Mar 2024 at 02:05, sddgit ***@***.***> wrote:
I’d really love for something to be done to address this. Maybe a
configurable timeout interval? But 10 seconds is clearly too short for many
systems. The message has occurred in my current log 27,518 times. And yet
all my air conditioners work flawlessly in HA. If not configurable, maybe a
one minute timeout would suffice. I can’t see a downside really.
—
Reply to this email directly, view it on GitHub
<#82972 (comment)>,
or unsubscribe
<https://github.com/notifications/unsubscribe-auth/AHPKKRDSFET7Z3CK4MWTF2LY2ILOJAVCNFSM6AAAAAASPL4VL6VHI2DSMVQWIX3LMV43OSLTON2WKQ3PNVWWK3TUHMZDAMRRG4ZTSMJXGI>
.
You are receiving this because you commented.Message ID:
***@***.***>
|
|
I have 4 splits (in Australia). They all behave exactly the same way wrt timeouts. Each individual “device” (climate device itself, temperature sensors etc.) on each unit suffers many, many timeouts. But any attempt to control the device is instantaneous. Even the official Daikin app is painfully slow to refresh status. I truly believe this is “normal” behaviour, at least for my systems. Different models cold well behave differently. I have a rock solid Unifi wifi system. I don’t think it has anything to do with wifi, more just the units themselves. A simple change to increase the timeout would be most welcome! It looks to me like the Air Touch 4 Daikin integration has a timeout of 30 seconds. |
|
can anyone advise how to change the 10 seconds timeout setting? |
|
I’m not home but I’ll try to have a look when back. I think the package
maintainers may have their reason not to have done it yet.
One last tip, I also disabled a lot of useless sensors (like whatever pump
frequency and the likes) don’t known if it helps or not. I suppose the
systems polls less data.
…On Mon 1 Apr 2024 at 01:15, robinhood-code ***@***.***> wrote:
can anyone advise how to change the 10 seconds timeout setting?
—
Reply to this email directly, view it on GitHub
<#82972 (comment)>,
or unsubscribe
<https://github.com/notifications/unsubscribe-auth/AHPKKRBHF5VZ2FMHQDSG22LY3CKIRAVCNFSM6AAAAAASPL4VL6VHI2DSMVQWIX3LMV43OSLTON2WKQ3PNVWWK3TUHMZDAMRYHE2DGMRTGY>
.
You are receiving this because you commented.Message ID:
***@***.***>
|
For me, the warning message is always: Logger: homeassistant.helpers.entity Update of sensor.guest_ac_inside_temperature is taking over 10 seconds so the timeout is for these 2 specific sensors but not the device itself. As I'm using the inside_temp in my automation I can't disable these 2 sensors. |
|
I get mostly inside_temperature warnings, and they happen every minute for each of my units. But I also get climate device and power measurement timeouts. I’m not sure if disabling the sensors will stop the integration polling for those sensors. Does it definitely stop some timeout errors for you @kameo4242 ? |
|
Hi everyone,
Same here, mostly temps failing to be fetched.
My setup is 2 units, 8 splits, unifi Wifi network (with 4 AP, dedicated
subnet on 2.4ghz, allowing 2mbps as lowest rate), all local with no cloud
dependency:
- 4 splits couldn't report their climate state at 2:33:03 (which include
temperature, so it could be related to this too)
- 2 splits couldn't report their climate state at 3:30:03 and 5:52:14
- 23 times the Update of sensor.XXX_inside_temperature is taking over 10
seconds with all splits more or less usually by packs (like 4 at 12:42)
I have to add that my onecta app is also reporting very weird temps
readings (like 26°C in parents room instead of 20°C in reality).
I've a few suspicions which would need confirmation here:
1/ The system automatically polls the climate units at regular intervals
but probably all at once, which could be part of the issue. Maybe spreading
those autopolls over 2 minutes would help.
2/ The Daikin temp probes are not really neither accurate or fast enough.
Their data bus may take more time than expected to poll the temperature,
leading the motherboard to wait and time out.
They are already not very accurate and cheap, so this could be part of the
problem
3/ Their BRP069B45 series may have poor wifi integration/performance
Le mar. 2 avr. 2024 à 01:31, sddgit ***@***.***> a écrit :
… I get mostly inside_temperature warnings, and they happen every minute for
each of my units. But I also get climate device and power measurement
timeouts. I’m not sure if disabling the sensors will stop the integration
polling for those sensors. Does it definitely stop some timeout errors for
you @kameo4242 <https://github.com/kameo4242> ?
—
Reply to this email directly, view it on GitHub
<#82972 (comment)>,
or unsubscribe
<https://github.com/notifications/unsubscribe-auth/AHPKKRG3ISN4POSGV4KUHQLY3H36TAVCNFSM6AAAAAASPL4VL6VHI2DSMVQWIX3LMV43OSLTON2WKQ3PNVWWK3TUHMZDAMZQG44DQNZZGE>
.
You are receiving this because you were mentioned.Message ID:
***@***.***>
--
__
Philippe Humeau
|
|
Whatever the issue, the fix should be simple - either change the timeout value or, better still, make it configurable. Hoping someone will have the time to look into doing this. |
|
Possible cause: #114345 |
The problem
Since quite a long time now, all my daikin devices cause warnings in the logs:
This has absolutely nothing to do with wifi speeds or disconnects.
Compareable issues were reported earlier, but not fixed.
#75383
#52739
What version of Home Assistant Core has the issue?
core-2022.11.4
What was the last working version of Home Assistant Core?
No response
What type of installation are you running?
Home Assistant OS
Integration causing the issue
daikin
Link to integration documentation on our website
https://www.home-assistant.io/integrations/daikin/
Diagnostics information
config_entry-unifi-7fa49060278dbf40c8e2d50cb7aa4ee1.json (1).txt
Example YAML snippet
No response
Anything in the logs that might be useful for us?
No response
Additional information
No response
The text was updated successfully, but these errors were encountered: