Skip to content

0.27.15, Recovery Times and Dashboard Filters

Pre-release
Pre-release

Choose a tag to compare

@TheThinkingHome TheThinkingHome released this 30 Oct 22:55
· 13 commits to main since this release
d45a1fc

Recovery Times

Corrects a small bug where the integration would incorrectly report the duration of an outage. A device that sent several messages as it came back was reported down from the moment it was called frozen, not from its last valid report: a window sensor out for 66 minutes was announced back "after 8 minutes". The down_for on the device_sentinel_recovered event, and the Daily Brief, count from the device's last report again.

Press a Type or an Area to Filter

You can press a device's type to show only devices of that type. It works on the Devices, Battery Trends and Classification tabs, in Simple and Advanced, the same as choosing the type in the Type filter. "Show all" brings the whole list back.

You can press a device's area to show only devices from that area. On the Devices tab in Advanced, in the list of watched devices and in the excluded list.

Clearer Words on the Dashboard and in the Daily Brief

Classification shows "Not Watched" for a device Device Sentinel does not watch. Each one read "Not Set", a link that saved a type nothing used.

Choosing "No area" on a device's page asks first. It reads "Remove Window Dining Room Left from Dining Room in Home Assistant?", with "Remove from area" and "Keep its area", because taking a device out of its area also changes Home Assistant.

A label BUGFIX. A label named with a comma, such as "Garden, Front", was not lit in the Labels row.

A device that fails on one day and recovers on another states the full duration of the outage. In Short reads "stopped reporting at Oct 29, 1:04 AM and recovered at Oct 30, 1:48 AM". Thanks to Tim Plas, whose brief showed it.

A restart in Last 24 Hours reads "restarted". The system's row no longer says "system".

Known Limits

Going back to 0.27.14 and forward again while a device has a problem can make that one recovery's down time too long. The next problem on that device is timed correctly.

Upgrading

Update and restart. Your learned rhythms, battery history, and signal history are safe, and there is nothing to change. Down times recorded before this release keep the values they were given, and a problem standing when you update is timed from when it was reported.