Configurable behavior when Frigate itself fails requests

[borgqueenx]

Apologise if i missed anything. Normally, when frigate card loses connection, it retries to connect. Although i also have quite some times where i see some code of frigate being unavailable, and then it does not retry. even though when i restart the home assistant app, it immediately works again. Is there perhaps something to be done about this?

Thanks!

[borgqueenx]

here is one of the errors. again this goes away if i restart the home assistant companion app:

[borgqueenx]

is it possible to send a personal message somehow @dermotduffy ? i captured a video, also about the card causing a crash of home assistant pretty reliably. but because it contains some data and imaging about our home i prefer to keep it private.

[dermotduffy]

@borgqueenx I am dermotduffy on discord, feel free to message me there.

The photo shows a failed request to Frigate, but not a failed connection to Home Assistant. Failures of the latter kind (to Home Assistant) should always retry. Failures from Frigate currently do not retry.

If you are having regular Frigate failures, it might be worth understanding why that is happening (e.g. either Frigate logs, or Home Assistant logs may show something). If the card is sending too much traffic to the backend, this can also be throttled (see https://card.camera/#/configuration/performance).

If may be possible to add retry support or perhaps "ignore" support for some or all of these failures, but the consequences are probably going to be that the card appears either "hung" (for retries) or "empty" (for ignore, e.g. no snapshots shown at all when you visit the snapshots view).

[borgqueenx]

here is another error that was given today:

I already dmmed you on discord, by the way, with a video of the card crashing.