No I never used Sanoid, since I’m completely happy with my small scripts and the zfs command line. I take snapshots once a week and I send them to two backups. The snapshots take seconds while the backup needs more actions like powering-on the backup-server and laptop, connecting the laptop to Ethernet and start the backup scripts, that will run for ~1 hour.
I don’t need more frequent backups, since most of my actions are related to Virtual Machines and I’m happy with that weekly snapshot and backup. Any issue can be solved by rolling back the VM and reapply the updates.
A last remark about point 7. Those tools will work fine for your ext4 health, but I was not talking about ext4 health. I was talking about the internal format of a file. On the ext4 level everything can be OK, but the media player might not be able to read the wma or mp3 file, because the sound part of the files is corrupted and does not follow the wma or mp3 format. The result is random noises or silence for a longer period (seconds to a minute), while the media player is trying to play the file. We are dealing with two distinct levels the ext4/partition level (free space; sectors used by each file etc) and the format inside a file say mp3, wma, doc, rpm, deb; jpeg; wmv; m4v etc. There are thousands of these formats and it is impossible for ext4, ntfs or any other program to know all those internal formats and to check and repair them. For audio files you could use Audacity to manually correct the problem as good as possible by cutting the bad part out.
Ext4 and NTFS do overwrite the file in place, leaving you after a crash with half of the new content and half of the old content and thus often an inconsistent internal format. For Ext4 and NTFS nothing changed, the same sectors are still allocated to the file. ZFS and BTRFS always create a new copy, so after the crash you have an uncorrupted copy, either the old file or the new file.
Again in your case the chance on this type of file corruption is absolutely minimal, since you use a laptop, so no power-fails and if you use Ubuntu, one of its flavors or derivatives you have a rock solid and reliable OS too. Most Linux distros are good, even some rolling release like Manjaro are relatively reliable.
This is helpful. Helps me to better understand the ‘copy on write’ concept. I’ve used Windows for a long time and, TTBOMK, only had very rare occasions of file corruption. Whenever I rebuild my Ubuntu install, I’ll research ZFS and consider using it. Thanks.