Fix ZHA force polled entities.

[Adminiuga]

Description:

Fix "force_polled" ZHA entities.

Checklist:

• The code change is tested and works locally.
• Local tests pass with tox. Your PR cannot be merged unless tests pass
• There is no commented out code in this PR.

[Adminiuga]

Ha, this is a nice PR#

[dmulcahey]

can merge when everything passes

[dmulcahey]

Ha, this is a nice PR#

Go play powerball

[tbrock47]

Not sure where to post this, but my ZHA devices were reporting better in 90.0 than in 90.1. can someone explain why that would have been? I'd like my devices to respond how they did in 90.0.

[robbiet480]

@tbrock47 Can you describe exactly the behaviors you are seeing and with what devices?

[tbrock47]

I'm running a raspberry pi with that nortek security and control stick that everyone else uses. I have about six sengled bulbs. In version 90.0 when zha would start failing the zha device would start reporting unavailable. With 90.1 they just report online even though they can't be reached. I'm sorry I don't recall exactly what the log entry was when they start stating that they can't be contacted anymore. Reason being is I had an automation built that would restart home assistant when all of my zha devices become unavailable which now doesn't work because everything stays set as online. I'm a git newbie so my apologies if there was a more appropriate way to report my issues.

…

On Sun, Mar 24, 2019, 7:30 PM Robbie Trencheny ***@***.***> wrote: @tbrock47 <https://github.com/tbrock47> Can you describe exactly the behaviors you are seeing and with what devices? — You are receiving this because you were mentioned. Reply to this email directly, view it on GitHub <#22222 (comment)>, or mute the thread <https://github.com/notifications/unsubscribe-auth/AmG8cBw4xdUxk7DNZJD3AnvVY2KlVdHTks5vaCbNgaJpZM4cAD9M> .

[robbiet480]

@tbrock47 This was the right place.

@dmulcahey pulling you in on this, especially interesting since you have Sengled bulbs too and haven't experienced these issues...

[tbrock47]

Some more details which would probably be helpful...

key	value
arch	armv7l
dev	false
docker	false
hassio	false
os_name	Linux
python_version	3.5.3
timezone	America/Denver
version	0.90.1
virtualenv	true

All the problem bulbs are Sengled E11-G13.
I have a few zwave devices that work flawlessly.

This is the automation I spoke of that executes a restart when no ZHA devices report online anymore. But as I said, now that their state no longer changes from online, the automation never fires.

alias: Home Assistant ZHA Failure Restart
initial_state: "on"
trigger:
  platform: event
  event_type: state_changed
condition:
  - condition: template
    value_template: "{{ trigger.event.data.new_state.domain == 'zha' }}"
  - condition: template
    value_template: >
      {% set ns = namespace(found=false) %}
      {%- for entity in states.zha -%}
      {% if entity.state == 'online' %}
      {%- set ns.found = true -%}
      {% endif %}
      {%- endfor -%}
      {{ not ns.found }}
action:
  - service: notify.notify
    data_template:
      message: "ZHA Failure Detected. Home Assistant Restarting."
  - service: homeassistant.restart

[MartinHjelmare]

@robbiet480 we want issues reported in the issue tracker. Not in merged PRs.

@tbrock47 Please open a new issue and copy the very good description already posted here. Make sure to fill out the issue template completely. And link to this PR. Thanks!

Here's the link to our issue tracker:
https://github.com/home-assistant/home-assistant/issues

[robbiet480]

@MartinHjelmare you're right, thought this was the issue :(

@tbrock47 Let us know when that new issue is opened, thanks!

[Adminiuga]

@tbrock47 Thank you for a good description of the problem. Let me know when you open the issue, but from the description the problem sounds very similar to zigpy/bellows#124 issue. We'll need bellows, zigpy, homeassistant.components.zha and homeassistant.core debug logs to confirm.