eupolicy.social is one of the many independent Mastodon servers you can use to participate in the fediverse.
This Mastodon server is a friendly and respectful discussion space for people working in areas related to EU policy. When you request to create an account, please tell us something about you.

Server stats:

226
active users

#ufs

0 posts0 participants0 posts today

Seems like, that #FreeBSD with #ZFS on #HAST with #CARP isn't reliable? I see often corrupted ZFS pools after failovers. Some minor research concluded, that ZFS is writing a lot of meta data, and that ZFS is prone to corruptions when something on block level doesn't get synchronized good enough. #UFS should be much more reliable. Maybe I will just use UFS for my HA storage and then #rsync to a #ZFS powered Backup server once daily or so. Any ideas anyone? #BSD #RunBSD

Replied in thread

@cks

The same thought occurred to me. There have been cute tricks like that played. I've even played some of them on other systems.

But as far as I know that first entry in the i-node table was just unused.

I wonder whether fsck even checked it. Modern FreeBSD fsck_ffs does, but that's a very different beast.

Replied in thread

@dvl @dch
Not 100% sure about loader.efi, but at least boot1.efi would be affected if any read-incompatible feature(s) is/are enabled and active even on data-only pools.
This is because, in boot1.efi, it sniffs the existence of /boot/loader.efi to determine whether the pool is bootable or not.
loader.efi should need to sniff at least pool name(s), whether it's bootable or not by somehow reading the pool(s).

If you're 100% sure there's not at all a plan to have bootable ZFS (means, use UFS for boot throughout the future), and once confirmed it works as wanted, using gptboot.efi would be the way to go.
It does not sniff ZFS pools (not linked against ZFS codes) and (in contrast with boot1.efi) not obsoleted.

As someone who lost data before (multiple times) I went for a overkill solution. 😎

Currently in the works:
- #ZFS Mirror 2 x 4TB SSD
- Directly attached 2TB SSD via a USB adapter (Running backup leveraging #rsync every day)
- Offsite 4TB spinning disk that get's plugged in every week (Also using #rsync here)

#FreeBSD and #ZFS keeps my data safe and a disaster strategy is also in place.

Btw, three different filesystems used (#ZFS, #EXT4, #UFS) :freebsd:

Better safe than sorry I'd say! 😉

Continued thread

Took the next step also upgrading my router/firewall VM to #FreeBSD 14.1. This was the typical multi-reboot dance

# make installkernel
# shutdown -r now
# etcupdate -p
# make BATCH_DELETE_OLD_FILES=yes installworld delete-old
# shutdown -r now
# etcupdate
# pkg upgrade -f
# pkg autoremove
# make BATCH_DELETE_OLD_FILES=yes delete-old-libs
# shutdown -r now

... because it's on #UFS. But of course, I took a #ZFS snapshot of the virtual disk first 😎