core: Vera Hub Sensor States Not Updating in Home-Assistant
Home Assistant release with the issue:
0.95.4
Last working Home Assistant release (if known): 0.94.4
Operating environment (Hass.io/Docker/Windows/etc.):
Docker Component/platform:
https://www.home-assistant.io/components/vera/
Description of problem: States not updated in Home-Assistant entities from Vera Hub since updating to 0.95. Possibly related to pyvera update #24289
Problem-relevant configuration.yaml
entries and (fill out even if it seems unimportant):
vera:
vera_controller_url: http://192.168.1.XX:3480/
Traceback (if applicable):
Additional information:
People seeing similar behaviour.
https://community.home-assistant.io/t/0-95-adguard-life360-plaato-airlock/123665/137?u=j_io_b
About this issue
- Original URL
- State: closed
- Created 5 years ago
- Reactions: 2
- Comments: 70 (13 by maintainers)
Commits related to this issue
- Update pyvera to 0.3.3 Fixes #24987 — committed to brandond/home-assistant by brandond 5 years ago
- Update pyvera to 0.3.3 Fixes #24987 — committed to brandond/home-assistant by brandond 5 years ago
- Update pyvera to 0.3.3 (#25820) Fixes #24987 — committed to home-assistant/core by brandond 5 years ago
- Update pyvera to 0.3.3 (#25820) Fixes #24987 — committed to home-assistant/core by brandond 5 years ago
- Fix for: https://github.com/home-assistant/home-assistant/issues/24987 because device_id is a string and the keys in _self.devices are integers... Maybe some other code should be changed so the ty... — committed to MarkJenniskens/pyvera by MarkJenniskens 5 years ago
I’m waiting on https://github.com/pavoni/pyvera/pull/113 to be accepted and pushed to PyPi so that I can reopen my hass PR. Anyone using Docker is welcome to use docker.io/brandond/home-assistant:pyvera-debug in the mean time.
Thanks @lielle1 - it appears that some alerts are missing the
PK_Device
attribute that ties the alert to a device ID. The fact that the traceback wasn’t going into the log you were looking at made this harder to diagnose. I’ll put in a PR to fix this, but I’m curious what is generating these deviceless alerts.0.96.0 was recently released that appears to include the 0.3.2 fix - everyone should be able to update and see if the issue is now resolved. I’ve updated and so far it appears to be working fine. I’ll report back if I see any issues in the next couple days as that’s how long it usually took for it to fail on 0.95.
I’ll be happy to test Docker image, I’m use HA docker on Synology NAS
@basharfaraj The version of pyvera that was included with HA 94.4 (last known working) was 0.2.45 The change for HA 95.x bumped pyvera to 0.3.1. Based on the commit that @InkblotAdmirer references, from what the pyvera project updates mention, seems there is a bug in 0.3.1 which is resolved in 0.3.2 (presumably it is able to handle null/missing values better). The change to 0.3.2 looks like it will be included in HA 96.x whenever that is available.
pyvera 0.3.2 is included in my docker image so I can just make the changes in the two files referenced in the commit @InkblotAdmirer linked and it should be good to go. You may or may not have 0.3.2 installed if you are not running this in docker?
To confirm you have pyvera 0.3.2 installed, run the following: pip install pyvera This should update/install 0.3.2 if you don’t have it - no idea if this would break anything else though so try this at your own risk.
Changes to make: Edit the two files referenced and change the pyvera version number from 0.3.1 to 0.3.2 homeassistant/components/vera/manifest.json requirements_all.txt (this is in the parent folder next with the home assistant directory in my Docker image)
Or alternatively wait for HA 96.x to be released. No guarantee the changes above will actually work but seems highly likely.
+1 Same issue 0.95.4
After re-starting Hassio, a couple of hours later motion sensors (Philio, Neo Coolcam, D-Link) stop updating event in Hassio. All automations based on them stop working. They are still reporting events at Vera Lite unit though.
I set the Vera component in debug logging and it shows normal polling activity until eventually it stops reporting and polling. It just stops reporting polling activity 😦 Last lines in the logs are
2019-07-07 11:15:00 DEBUG (Vera Poll Thread) [pyvera] get_changed_devices() requesting payload {‘timeout’: 30, ‘minimumdelay’: 200, ‘id’: ‘lu_sdata’, ‘loadtime’: 1562385868, ‘dataversion’: 385901882} 2019-07-07 11:15:30 DEBUG (Vera Poll Thread) [pyvera.subscribe] Poll returned
Look as if HA gets a timeout when waiting for Vera to reply, it won’t continue polling. Is this the case? Is this the expected behaviour? If so, could it be tweaked by adjusting timeout, or not stopping polling even if a timeout is eventually got?