Closed mstormi closed 3 years ago
This may have been caused by the improper sync of zram that was happening on shutdown until Sunday. I don't think it is a major issue.
until Sunday ? Are you aware that zram-config, unlike openhabian, does not get updated on every openhabian-config start ? There (still) multiple reports of reboot to take very long or "fail".
Yes I am aware. They need to update. See my comment of the other thread.
ok let's close it given we assume with your fix and my permissions PR there's nothing left to do. We can reopen if I'm wrong here.
just FYI, I ran into an extraordinarily mean error on my own production system that took me half a day to find out (with shutters all down and heating off ☹️ ):
Some .rrd file (rrd4j persistent storage) was broken and that made Java crash repeatedly. But deleting it didn't help. Persistence files are on ZRAM but seems when I deleted it ZRAM took the older version from the lowerFS and seems that was broken in the first place. Duh.
No, don't know yet what to make of that, just documenting for now.
Maybe one thing is we should extend the permissions_corrections() apply to /opt/zram/.bind/ too (the lowerFS copies of all the ZRAM'ed dirs).