On 3/4/25 21:59, Max Nikulin wrote:
In this particular case I do not think it is a drive failure. I suspect mounting /home was a mistake.
Indeed. "-ro" is not read only when a journal is in play.
Booting a live image may destroy hibernation data since live system may mount the same swap partition and reinitialize it. Eben, have you find a reliable way to avoid swap reuse?
The environment from which I do the backups is system-rescue.org . AFAICT it doesn't mount anything automatically. I don't generally mount a source partition, but this time I needed some info from it.
My impression is that some device drivers do not save complete state during hibernation, so booting other OS may cause some problems after resuming.
Yup. In fact sometimes just hibernate / resume is enough to cause failure. Thankfully all my devices use well-behaved drivers.
I considered it as a reason why Windows does not allow booting other OS when it is hibernated (at least with BIOS namely shutdown was required, I am unsure concerning UEFI).
Huh, I was not aware.
