Closed kbx81 closed 4 years ago
I have the same issue with the same firmware update and Roomba version. As with you, commenting out the Roomba section in configuration.yaml allows HA to start. For me, I believe there was an ERROR 104
When first connecting to the Roomba and preceding the cascading failures.
Hey there @pschmitt, mind taking a look at this issue as its been labeled with a integration (roomba
) you are listed as a codeowner for? Thanks!
The MQTT port (1883) is showing up as "filtered" when probed with nmap, so presumably that's why it's failing.
I'm having a similar problem with 0.98.5. Commenting out the Roomba section was the only recourse. Interestingly, HA continued to talk to my Roomba 980. Apparently there wasn't an update for it.
As a side note, in my case I'm running the Roomba980-Python library rather than the Roomba support built into HA (that doesn't work for me). I just checked the status and nobody has reported issues with the library as yet.
@kbx81, are you running the stock component? I was under the impression it was not compatible with the 980 nor the i7.
I’m running the stock Roomba component and it’s been working fine with my i7+ since I got it last year. It also worked with my 980.
@ratsputin I will assure you that the stock component has been working flawlessly with the i7(+) I have for months now -- this is not a new setup that randomly doesn't work (or that broke after an HA upgrade/reconfig). :) I'm familiar with Roomba980-Python as well as dorita but haven't tried to pull either of them into HA. I actually used dorita to determine (and verify the other day) my robot's credentials for its config in HA. It was still able to communicate with the robot although I didn't try to use it for control.
I just ran into the same issue today as my Roomba i7 updated to the same firmware. Been using it with the 980 python code since it came out.
Same issue with my S9+, same error. commenting it out of configuration.yaml lets everything start up normally.
So i found https://github.com/NickWaterton/Roomba980-Python/issues/39#issuecomment-538665656 Which lead me to https://github.com/zhouF96/Roomba980-Python/commit/ede7e95e5b10c7e1d52154815f96098859287365
And that does seem to be the issue. I manually edited the following in HA
/usr/local/lib/python3.7/site-packages/roomba/roomba.py
/usr/local/lib/python3.7/site-packages/roomba/password.py
and changed TLSv1
to TLS
, and when i started back up, everything was happy again.
You can run the following to fix it. The first line makes a backup of all .py files, the second uses sed to make a global inline replacement.
for file in /usr/local/lib/python3.7/site-packages/roomba/*.py; do cp "$file" "${file}.bak"; done
sed -i 's/PROTOCOL_TLSv1/PROTOCOL_TLS/g' /usr/local/lib/python3.7/site-packages/roomba/*.py
Looks like @NickWaterton patched it, so we just need https://pypi.org/project/roombapy/ updated, and https://github.com/home-assistant/home-assistant/blob/dev/homeassistant/components/roomba/manifest.json updated to that version.
@namachieli Thanks so much for tracking this down. I made the suggested changes and my i7+ is once again working.
I opened a PR on the roombapy repository to try to get the package authors attention.
Bump. 😄
FWIW I made the above fixes/patches to password.py
and roomba.py
in my (now) 0.102 HA container and the robot is back and working in HA.
Is anyone able to act on the PR and get the fix merged in so it can get released? 😄
Home Assistant release with the issue: 0.99.3
Last working Home Assistant release (if known): 0.99.3
Operating environment (Hass.io/Docker/Windows/etc.): Hass.io 189 on RPi 3 B+
Component/platform: https://www.home-assistant.io/components/roomba/
Description of problem: My iRobot Roomba i7+ received a firmware update to version 3.0.11 on September 27th. It appears that, since this update, the component fails (or causes the cloud component to fail?) at startup.
Problem-relevant
configuration.yaml
entries and (fill out even if it seems unimportant):Removing this configuration results in HA starting normally.
Traceback (if applicable):
After restoring onto a new SD card, the error is slightly different:
Additional information: Things I've tried in an effort to fix:
custom_components
(excepthacs
)It is absolutely possible that the firmware update breaking the component is somehow a coincidence, however I can say with absolute certainty it has was working the day prior to the firmware update. No HA updates or configuration changes were made between when it was last working and when it stopped working so I'm not sure what else to attribute this new error to.