After a long, long period without any issues at all, I’m having periodic lockups of my HA instance. This is running vanilla HassOS (up to date) from a clean install onto a new uSD card (SanDisk “max endurance”, bought new one month ago when the problem started in an attempt to diagnose it) and my imported configuration, while tweaked a little, hasn’t changed too much over the last 6 months.
When it fails, it seems to just drop off the network and stops responding to pings. It always recovers with a power cycle and, at the moment, lasts for maybe 3-4 days before it fails again. It doesn’t always fail at the same time of day and I can’t get any pattern from it.
I would appreciate some suggestions on how to diagnose this as, when it last happened, I pulled the HA logs before rebooting, but they didn’t contain anything useful that might suggest why it had failed. The Pi4 has been a pretty good platform for me, but I depend on HA too much now, so if I can’t work out why this is happening, I might just have to bin the Pi altogether
Any suggestions, tips or assistance would be very much appreciated.
If the machine does not even respond to ping, it’s probably an issue with the underlying GNU/Linux operating system.
When you experience the freeze next time, before rebooting, take a look at non-HA-related logs (dmesg, journalctl -b, stuff in /var/log/). I guess you’ll need to hook up a screen and a keyboard to the RPi if networking does not work.
Thanks. I considered this when this first started happening, but reading this post about it, it still looks a pretty enormous faff to setup on a Pi 4 (at least compared to SD boot which literally takes 5 mins to do).
In the end, I just bought a new, high-endurance SD card so I could rule out a failing SD card, but the lockups have continued at about the same frequency. I suspect that if I can’t resolve this, my next step will be to dump the Pi altogether for some more robust hardware, which will support SSD properly!
