Zwave not reliable

Canedje · February 27, 2020, 9:08pm

I’m runnig Home Assitant now for 2 weeks.
Running on a Raspberry Pi4B 4Gb.
I do have a conbee II Zigbee Dongle and a Zwave.me zwave dongle.
After sometime at least once a day non of the zwave devices are reacting anymore or very slow (reacting after 20 seconds or so)
I do need to do a reset of HA to get the system started again running for some time.
I do not see very busy network on OZW.
If I look to the PIRS in Overview they stil recieving signals going on and off.
If I do try to switching by hand on overview in Lovelace. The devices don’t react.
I do use about 30 devices: the most Fibaro and Neo Coolcam
Any ideas what to do?

Zigbee is giving no problem.

If I do decide to install an other type Zwave dongle do I need to pair al devices again? Or is the zwcf file takes care of the pairing?

Tinkerer · February 28, 2020, 7:45am

Probably it’s a mesh communication issue - check OZW_Log.txt and what your mesh looks like (the mesh is just a rough guide if you’re using Z-Wave Plus, but it’s still useful).

Other things worth checking are the RTT attributes, and sentFailed

Pairings are held in the USB stick, so you’d have to remove all the devices from the current mesh, then replace the controller, then include with the new one.

dgaust · February 28, 2020, 8:18am

There’s a known issue with the aeotec z-wave stick and the Raspberry Pi 4. Not sure if that’s the z-wave device you have, but try putting it on a USB2 hub.

Tinkerer · February 28, 2020, 8:18am

I’d say they’re not using an Aeotec dongle

dgaust · February 28, 2020, 8:19am

I missed that

Canedje · February 28, 2020, 10:40am

Thanks for the answers.
I checked the OZW_Log.txt file just a minute agoa. no strange items.
I will check again if I do recognize problems in reliability.

A pitty that a do need pairing by a new dongle .

If I want a new zwave plus dongle. Any advice?

I do see an unknown node in my mesh zwave network.
I do try to get rith of it by switching off the zwave network and modifying the zwcf file by removing the node.
But after restart the unknown node is popping up again.
Any idea how to remove unknown nodes?

Canedje · February 28, 2020, 11:02am

I do see this types of errors in my logfile:

Developer Tools


INFO

Error while getting Updates: Conflict: terminated by other getUpdates request; make sure that only one bot instance is running (409)

11:59:52 AM – /usr/local/lib/python3.7/site-packages/telegram/ext/updater.py (ERROR) - message first occurred at 11:58:45 AM and shows up 3 times

Error saving event: <Event state_changed[L]: entity_id=sensor.fibaro_dubbele_schakelaar_gang_energie, old_state=<state sensor.fibaro_dubbele_schakelaar_gang_energie=5.94; node_id=21, value_index=0, value_instance=2, value_id=144115188433190914, power_consumption=0.0, unit_of_measurement=kWh, friendly_name=Fibaro dubbele schakelaar gang energie @ 2020-02-28T11:37:25.235214+01:00>, new_state=<state sensor.fibaro_dubbele_schakelaar_gang_energie=5.94; node_id=21, value_index=0, value_instance=2, value_id=144115188433190914, power_consumption=0.0, unit_of_measurement=kWh, friendly_name=Fibaro dubbele schakelaar gang energie @ 2020-02-28T11:59:30.090385+01:00>>

11:59:30 AM – recorder (ERROR)

Error executing query: (sqlite3.DatabaseError) database disk image is malformed [SQL: INSERT INTO events (event_type, event_data, origin, time_fired, created, context_id, context_user_id) VALUES (?, ?, ?, ?, ?, ?, ?)] [parameters: ('state_changed', '{"entity_id": "sensor.fibaro_dubbele_schakelaar_gang_energie", "old_state": {"entity_id": "sensor.fibaro_dubbele_schakelaar_gang_energie", "state": " ... (780 characters truncated) ... :00", "last_updated": "2020-02-28T10:59:30.090385+00:00", "context": {"id": "b9aadc8caad34cb59ae4e9024399e707", "parent_id": null, "user_id": null}}}', 'LOCAL', '2020-02-28 10:59:30.090461', '2020-02-28 10:59:30.128684', 'b9aadc8caad34cb59ae4e9024399e707', None)] (Background on this error at: http://sqlalche.me/e/4xp6)

11:59:30 AM – recorder (ERROR)

Error saving event: <Event state_changed[L]: entity_id=sensor.fibaro_system_fgs223_double_relay_exporting_12, old_state=<state sensor.fibaro_system_fgs223_double_relay_exporting_12=False; node_id=21, value_index=32, value_instance=2, value_id=144115188433191424, power_consumption=0.0, unit_of_measurement=, friendly_name=Fibaro dubbele schakelaar gang @ 2020-02-28T11:37:25.177540+01:00>, new_state=<state sensor.fibaro_system_fgs223_double_relay_exporting_12=False; node_id=21, value_index=32, value_instance=2, value_id=144115188433191424, power_consumption=0.0, unit_of_measurement=, friendly_name=Fibaro dubbele schakelaar gang @ 2020-02-28T11:59:30.066734+01:00>>

11:59:30 AM – recorder (ERROR)

Error executing query: (sqlite3.DatabaseError) database disk image is malformed [SQL: INSERT INTO events (event_type, event_data, origin, time_fired, created, context_id, context_user_id) VALUES (?, ?, ?, ?, ?, ?, ?)] [parameters: ('state_changed', '{"entity_id": "sensor.fibaro_system_fgs223_double_relay_exporting_12", "old_state": {"entity_id": "sensor.fibaro_system_fgs223_double_relay_exporting ... (786 characters truncated) ... :00", "last_updated": "2020-02-28T10:59:30.066734+00:00", "context": {"id": "25c6e4a72ddc4f18b139fec108db76e7", "parent_id": null, "user_id": null}}}', 'LOCAL', '2020-02-28 10:59:30.066811', '2020-02-28 10:59:30.112950', '25c6e4a72ddc4f18b139fec108db76e7', None)] (Background on this error at: http://sqlalche.me/e/4xp6)

11:59:30 AM – recorder (ERROR)

Error saving event: <Event state_changed[L]: entity_id=zwave.fibaro_dubbele_schakelaar_gang, old_state=<state zwave.fibaro_dubbele_schakelaar_gang=ready; node_id=21, node_name=FIBARO System FGS223 Double Relay, manufacturer_name=FIBARO System, product_name=FGS223 Double Relay, query_stage=Complete, is_awake=True, is_ready=True, is_failed=False, is_info_received=True, max_baud_rate=40000, is_zwave_plus=True, capabilities={'beaming', 'zwave_plus', 'listening', 'routing'}, neighbors={1, 2, 3, 4, 7, 8, 12, 16, 18, 19, 22, 23, 26, 27, 29, 30, 31, 32, 33, 34, 39, 42, 43, 44}, sentCnt=22, sentFailed=0, retries=0, receivedCnt=17, receivedDups=0, receivedUnsolicited=1, sentTS=2020-02-28 11:37:25:221 , receivedTS=2020-02-28 11:59:29:766 , lastRequestRTT=28, averageRequestRTT=45, lastResponseRTT=41, averageResponseRTT=69, application_version=Unknown, friendly_name=Fibaro dubbele schakelaar gang @ 2020-02-28T11:37:49.610913+01:00>, new_state=<state zwave.fibaro_dubbele_schakelaar_gang=ready; node_id=21, node_name=FIBARO System FGS223 Double Relay, manufacturer_name=FIBARO System, product_name=FGS223 Double Relay, query_stage=Complete, is_awake=True, is_ready=True, is_failed=False, is_info_received=True, max_baud_rate=40000, is_zwave_plus=True, capabilities={'beaming', 'zwave_plus', 'listening', 'routing'}, neighbors={1, 2, 3, 4, 7, 8, 12, 16, 18, 19, 22, 23, 26, 27, 29, 30, 31, 32, 33, 34, 39, 42, 43, 44}, sentCnt=22, sentFailed=0, retries=0, receivedCnt=18, receivedDups=0, receivedUnsolicited=2, sentTS=2020-02-28 11:37:25:221 , receivedTS=2020-02-28 11:59:29:952 , lastRequestRTT=28, averageRequestRTT=45, lastResponseRTT=41, averageResponseRTT=69, application_version=Unknown, friendly_name=Fibaro dubbele schakelaar gang @ 2020-02-28T11:37:49.610913+01:00>>

11:59:30 AM – recorder (ERROR)

Error executing query: (sqlite3.DatabaseError) database disk image is malformed [SQL: INSERT INTO events (event_type, event_data, origin, time_fired, created, context_id, context_user_id) VALUES (?, ?, ?, ?, ?, ?, ?)] [parameters: ('state_changed', '{"entity_id": "zwave.fibaro_dubbele_schakelaar_gang", "old_state": {"entity_id": "zwave.fibaro_dubbele_schakelaar_gang", "state": "ready", "attribute ... (2031 characters truncated) ... :00", "last_updated": "2020-02-28T10:59:30.065016+00:00", "context": {"id": "51d2257ad9fa426eba19cf7522d853a6", "parent_id": null, "user_id": null}}}', 'LOCAL', '2020-02-28 10:59:30.065133', '2020-02-28 10:59:30.086240', '51d2257ad9fa426eba19cf7522d853a6', None)] (Background on this error at: http://sqlalche.me/e/4xp6)

11:59:30 AM – recorder (ERROR)

Error saving event: <Event state_changed[L]: entity_id=sensor.fibaro_dubbele_schakelaar_buitengan_energie, old_state=<state sensor.fibaro_dubbele_schakelaar_buitengan_energie=13.87; node_id=21, value_index=0, value_instance=1, value_id=72057594395262978, power_consumption=0.0, unit_of_measurement=kWh, friendly_name=Fibaro dubbele schakelaar buitengang energie @ 2020-02-28T11:37:24.786527+01:00>, new_state=<state sensor.fibaro_dubbele_schakelaar_buitengan_energie=13.87; node_id=21, value_index=0, value_instance=1, value_id=72057594395262978, power_consumption=0.0, unit_of_measurement=kWh, friendly_name=Fibaro dubbele schakelaar buitengang energie @ 2020-02-28T11:59:29.936567+01:00>>

11:59:30 AM – recorder (ERROR)

Can that be the problem?
If yes what is causing this and how to fix?
It’s a corrupt database problem I do read.
But don’t know how to fix this.

raymonvdm · February 28, 2020, 3:46pm

You can transfer the Z-Wave network to a new stick of the stick supports it. I`m using 3 UZB sticks for this proces since i bricked on using the firmware tool. (Two working sticks left on firmware 5.36)

I am also experiencing Z-Wave issues and i`m not using a Pi but a normal desktop. I think the issue is more related to Open-Zwave than to HA

Tinkerer · February 28, 2020, 4:15pm

That’s a different log than the one I was directing you to, but … that can cause other issues

Delete /config/home-assistant_v2.db (or wherever it is if you didn’t use Hass.io/Docker)
Restart Home Assistant

Canedje · February 28, 2020, 4:40pm

So there is a method to update the zwave.me firmware?
where to find? or how to do?

Canedje · February 28, 2020, 4:42pm

Thanks again
I know you mentioned an other file, but I did see this happening at the log file

raymonvdm · February 28, 2020, 4:51pm

The only way to do this is installing Debian 8 on a system and installing Z-Way. I did mine using Virtualbox on Windows 10 and connecting the UZB stick to the VM

wget https://storage.z-wave.me/z-way-server/z-way-server-Debian-v3.0.0.tgz
tar xvfz z-way-server-Debian-v3.0.0.tgz
cd z-way-server
chmod 666 /dev/ttyACM0

apt-get install libcurl3
apt-get install libarchive13
apt-get install libavahi-compat-libdnssd1

LD_LIBRARY_PATH=./libs ./z-way-server

And then connect to the ip adres of you VM (might want to set networking to bridge for this VM)

http://192.168.xxx.xx:8083/

Then go to the Z-Wave node in the expert webinterface and check for firmwares, if it states that the firmware is up2date you need to use the code “all” to see the correct firmwares (also take note on the bootloaders)

http://192.168.xx.xx:8083/expert/

5.27 can give you an upgrade, 5.xx higer needs to be downgrade before you can upgrade to 5.36

Canedje · February 28, 2020, 5:11pm

Thanks again
sounds complex to me.
Is it possible to do it straigh on my Raspberyy Pi using a SSH?

raymonvdm · February 28, 2020, 5:14pm

Sorry, forgot you`re using a Pi. Yes it is possible but not on all version of Raspbian

https://z-wave.me/z-way/download-z-way/

Canedje · February 28, 2020, 8:45pm

The network is not responding again now.
No special message at the OZW_Log.txt.No sent-failed or so.
The traffic self is also not to much in my opnion

I recognize a pattern.
It looks like it is happening from 21:00 hour.
Im start running at that time an transition in 15 minutes for a RGBW dimmer in steps of a 1%.
I do use a node called lullaby.
I do have the feeling that this node is using to much system or so that stops the Zwave mesh to work.

I tried to switch by hand at the lovelace overview several switches. Nothing happening.
Suddenly after about 5 minutes or so the lights are switching on and off the ones i switched by hand.
Zwave is keeping in mind what I asked for, but is only able to execute after 5 minutes or so?

I realy don’t understand what is happening.

Tinkerer · February 29, 2020, 8:26am

It sounds like that “node” (I’m guessing Node Red) is overloading the capabilities of your Pi.

One thing to do is look at the light transition setting since with many integrations you can set a transition of many minutes.

Canedje · February 29, 2020, 8:56am

Thanks again. I did have an transition of 15 minutes.
I stopt that que as experiment
and do put a simple switch on and off que instead

Any ideas how to make a less intensive transition que?

Tinkerer · February 29, 2020, 3:03pm

Within Node Red? No idea, don’t use it. You’d want to open a new thread in the right sub-section here.

Canedje · February 29, 2020, 6:57pm

Thanks.

It is not the solution. Zwave still not reliable.
Special in the evening when light is on.
I do see a lot of traffic on one node. (node 2)
It is a neo coolcam wallplug doing nothing but be on.
I show a part of the OZW file:

2020-02-29 19:53:00.749 Detail, 
2020-02-29 19:53:00.749 Detail, Node002, Refreshed Value: old value=false, new value=false, type=bool
2020-02-29 19:53:00.749 Detail, Node002, Changes to this value are not verified
2020-02-29 19:53:00.749 Info, Node002, Received Meter report from node 2: Power=15.97W
2020-02-29 19:53:00.749 Detail, Node002, Refreshed Value: old value=15.98, new value=15.97, type=decimal
2020-02-29 19:53:00.749 Detail, Node002, Changes to this value are not verified
2020-02-29 19:53:00.749 Info, Node002,     Previous value was 15.98W, received 7 seconds ago.
2020-02-29 19:53:00.749 Detail, Node002, Refreshed Value: old value=15.96, new value=15.98, type=decimal
2020-02-29 19:53:00.749 Detail, Node002, Changes to this value are not verified
2020-02-29 19:53:00.749 Detail, Node002, Refreshed Value: old value=13, new value=7, type=int
2020-02-29 19:53:00.749 Detail, Node002, Changes to this value are not verified
2020-02-29 19:53:00.749 Detail, Node002, Notification: ValueChanged
2020-02-29 19:53:00.774 Detail, Node002, Notification: ValueChanged
2020-02-29 19:53:00.794 Detail, Node002, Notification: ValueChanged
2020-02-29 19:53:00.811 Detail, Node002, Notification: ValueChanged
2020-02-29 19:53:01.810 Detail, Node002,   Received: 0x01, 0x12, 0x00, 0x04, 0x00, 0x02, 0x0a, 0x32, 0x02, 0xa1, 0x42, 0x59, 0x40, 0x00, 0x07, 0x59, 0x4a, 0xbf, 0x00, 0x80
2020-02-29 19:53:01.811 Detail, 
2020-02-29 19:53:01.811 Detail, Node002, Refreshed Value: old value=false, new value=false, type=bool
2020-02-29 19:53:01.811 Detail, Node002, Changes to this value are not verified
2020-02-29 19:53:01.811 Info, Node002, Received Meter report from node 2: Voltage=228.48V
2020-02-29 19:53:01.811 Detail, Node002, Refreshed Value: old value=228.58, new value=228.48, type=decimal
2020-02-29 19:53:01.811 Detail, Node002, Changes to this value are not verified
2020-02-29 19:53:01.811 Info, Node002,     Previous value was 228.58V, received 7 seconds ago.
2020-02-29 19:53:01.811 Detail, Node002, Refreshed Value: old value=228.49, new value=228.58, type=decimal
2020-02-29 19:53:01.812 Detail, Node002, Changes to this value are not verified
2020-02-29 19:53:01.812 Detail, Node002, Refreshed Value: old value=13, new value=7, type=int
2020-02-29 19:53:01.812 Detail, Node002, Changes to this value are not verified
2020-02-29 19:53:01.812 Detail, Node002, Notification: ValueChanged
2020-02-29 19:53:01.830 Detail, Node002, Notification: ValueChanged
2020-02-29 19:53:01.847 Detail, Node002, Notification: ValueChanged
2020-02-29 19:53:01.864 Detail, Node002, Notification: ValueChanged
2020-02-29 19:53:02.747 Detail, Node002,   Received: 0x01, 0x12, 0x00, 0x04, 0x00, 0x02, 0x0a, 0x32, 0x02, 0xa1, 0x4a, 0x00, 0x06, 0x00, 0x07, 0x00, 0x07, 0xbe, 0x00, 0x82
2020-02-29 19:53:02.747 Detail, 
2020-02-29 19:53:02.748 Detail, Node002, Refreshed Value: old value=false, new value=false, type=bool
2020-02-29 19:53:02.748 Detail, Node002, Changes to this value are not verified
2020-02-29 19:53:02.748 Info, Node002, Received Meter report from node 2: Current=0.06A
2020-02-29 19:53:02.748 Detail, Node002, Refreshed Value: old value=0.07, new value=0.06, type=decimal
2020-02-29 19:53:02.748 Detail, Node002, Changes to this value are not verified
2020-02-29 19:53:02.748 Info, Node002,     Previous value was 0.07A, received 7 seconds ago.
2020-02-29 19:53:02.748 Detail, Node002, Refreshed Value: old value=0.06, new value=0.07, type=decimal
2020-02-29 19:53:02.748 Detail, Node002, Changes to this value are not verified
2020-02-29 19:53:02.748 Detail, Node002, Refreshed Value: old value=13, new value=7, type=int
2020-02-29 19:53:02.748 Detail, Node002, Changes to this value are not verified
2020-02-29 19:53:02.748 Detail, Node002, Notification: ValueChanged
2020-02-29 19:53:02.771 Detail, Node002, Notification: ValueChanged
2020-02-29 19:53:02.787 Detail, Node002, Notification: ValueChanged
2020-02-29 19:53:02.803 Detail, Node002, Notification: ValueChanged

This is going on for a long time.
Do you recognize something strange?
Why is there this amount of communiaction?

If I do a heal network. Everything reacting oke again for a (short) while
In day time I do not see this delays

Do you know how to check how busy my RPI is somehow?

Tinkerer · February 29, 2020, 8:34pm

Don’t do Heal network you’re breaking things.

However, that log shows that something is sending notifications over the mesh every second. Z-Wave is low bandwidth, having something that chatty may be causing you issues.