OH 4 Webconsole extremly slow

Setup, Configuration and Use Installation
1

Hi,

since I “reinstalled” my OH 4.1.0 yesterday, I copied my .things, .items, .map and just begann from the beginning.
I’ve installed knx, zwave, network, atro and ntp Bindings.

For Background Info, my productive OH-Automation is still running on another Pi with OH1.8!!! Now I try to migrate automation-workload one after the other.
During the last Version of migration every thing was fine. But since I reinstalled Openhabian yesterday the webconsole is terrible slow! I takes roundabout 30 sec. to switch from “Things” to “Items” and back as well.

I guess there is a simple failture in any config file (thing, item, etc.) but I cannot find the source of this.

It turns me nuts. I connot work like this.

What can I do? Is there any idea??

Thank you in advance…
Bernd

##   Release = Raspbian GNU/Linux 11 (bullseye)
##    Kernel = Linux 6.1.21-v7+
##  Platform = Raspberry Pi 3 Model B Rev 1.2
##    Uptime = 0 day(s). 00:05:06
## CPU Usage = 17.1% avg over 4 cpu(s) (4 core(s) x 1 socket(s))
##  CPU Load = 1m: 0.19, 5m: 0.58, 15m: 0.32
##    Memory = Free: 0.21GB (22%), Used: 0.73GB (78%), Total: 0.94GB
##      Swap = Free: 2.28GB (100%), Used: 0.00GB (0%), Total: 2.28GB
##      Root = Free: 22.97GB (82%), Used: 4.78GB (18%), Total: 28.98GB
##   Updates = 57 apt updates available.
##  Sessions = 1 session(s)
## Processes = 133 running processes of 32768 maximum processes
...
==> /var/log/openhab/openhab.log <==
2023-12-31 14:59:03.605 [WARN ] [nx.internal.client.AbstractKNXClient] - Giving up reading datapoint 2/0/23, the number of maximum retries (3) is reached.
2023-12-31 14:59:03.731 [WARN ] [calimero.link.192.168.28.253:3671   ] - negative confirmation of 2/0/11: 2e 00 9d e0 11 fa 10 0b 01 00 00
2023-12-31 14:59:13.734 [WARN ] [nx.internal.client.AbstractKNXClient] - Giving up reading datapoint 2/0/11, the number of maximum retries (3) is reached.
2023-12-31 14:59:13.860 [WARN ] [calimero.link.192.168.28.253:3671   ] - negative confirmation of 2/0/22: 2e 00 9d e0 11 fa 10 16 01 00 00
2023-12-31 14:59:23.863 [WARN ] [nx.internal.client.AbstractKNXClient] - Giving up reading datapoint 2/0/22, the number of maximum retries (3) is reached.
2023-12-31 14:59:33.936 [WARN ] [nx.internal.client.AbstractKNXClient] - Giving up reading datapoint 2/0/21, the number of maximum retries (3) is reached.
2023-12-31 14:59:44.010 [WARN ] [nx.internal.client.AbstractKNXClient] - Giving up reading datapoint 2/0/16, the number of maximum retries (3) is reached.
2023-12-31 14:59:54.084 [WARN ] [nx.internal.client.AbstractKNXClient] - Giving up reading datapoint 2/0/18, the number of maximum retries (3) is reached.
2023-12-31 15:00:04.158 [WARN ] [nx.internal.client.AbstractKNXClient] - Giving up reading datapoint 2/0/6, the number of maximum retries (3) is reached.
2023-12-31 15:05:32.218 [INFO ] [el.core.internal.ModelRepositoryImpl] - Loading model 'Network.things'
2

It’s hard to say. There usual culprit is lack of RAM or high load but assuming the info posted above is while there is a problem that’s not the issue.

It’s impossible to say whether the KNX warnings are true cause it a symptom of the problem. But in general it’s really hard to cause a performance proven like this through configs. Even if you have a rule run amok that would only consume one CPU.

I guess to start create a clean install and migrate stuff little by little until the problem reoccurs.

3

Those errors are because you make group adresses readable but they are actually not. Check your ets config. I also had errors like that because Iof that.

4

I re-configured the knx.things und items… Now the log is almost without some knx WARNINGS.
Im on the rest.

But die Web-Site is still horrible slow…

5

I suspect I have found or solved the issue…
Since I corrected my Astro-Binding geolocation to a useable value, the performance is back again.