Jason Tubnor ๐ฆ๐บ
Christopher ObbardI'm setting up my fileserver again. Before I used MergerFS and 4x4tb NAS drivers, to aggregate the files on all the drives as one mountpoint. The nice thing about mergerfs is that the disks are completely separate (files are written whole to each disk - nice !). The data is not important at all to me and backed up so I don't need RAID etc. Any suggestions of an alternative approach? I hear ZFS pools are interesting, but never used zfs. #linux #mergerfs #zfs #datahoarding
@robn Reflecting on previous discussions, should the documentation for say #RHEL and compatible operating systems include the swap to an EPEL LTS kernel for #ZFS installs to ensure it fits within a supported framework of #OpenZFS development and support? This would also prevent surprise upgrades where #RedHat in their wisdom, decide to back port kernel symbols.
Dan LangilleMy current plan, to move from SSD to SATADOM (and to smaller disks, effectively reducing my zpool size) is to use `zfs add`, not `zfs replace`.
See what was done over here: https://infosec.exchange/@bob_zim/115531520404665696
I'll be going from this:
[18:13 r730-01 dvl ~] % zpool list zroot
NAME SIZE ALLOC FREE CKPOINT EXPANDSZ FRAG CAP DEDUP HEALTH ALTROOT
zroot 424G 32.8G 391G - - 14% 7% 1.00x ONLINE -
[18:13 r730-01 dvl ~] % gpart show da0
=> 40 937703008 da0 GPT (447G)
40 1024 1 freebsd-boot (512K)
1064 984 - free - (492K)
2048 41943040 2 freebsd-swap (20G)
41945088 895756288 3 freebsd-zfs (427G)
937701376 1672 - free - (836K)
to this:
dvl@r730-04:~ $ zpool list zroot
NAME SIZE ALLOC FREE CKPOINT EXPANDSZ FRAG CAP DEDUP HEALTH ALTROOT
zroot 107G 1.07G 106G - - 0% 0% 1.00x ONLINE -
errors: No known data errors
dvl@r730-04:~ $ gpart show ada0
=> 40 242255584 ada0 GPT (116G)
40 532480 1 efi (260M)
532520 2008 - free - (1.0M)
534528 16777216 2 freebsd-swap (8.0G)
17311744 224942080 3 freebsd-zfs (107G)
242253824 1800 - free - (900K)
Zimmie (@bob_zim@infosec.exchange)
@jornane@ipv6.social @ronnie_bonkers@mastodon.social โzpool replaceโ wouldnโt work in this particular case: > The size of new-device must be greater than or equal to the minimum size of all the devices in a mirror or raidz configuration. The point here was to increase swap space, which means the space remaining for ZFS to use would be smaller.
When a ZFS system starts going rogue and fills the disk, it would be nice if `sudo` still worked.
I know of quota, reservation, and refreservation.
I've tested with `zfs create -o refreservation=5GB zroot/reservation` etc, but the system still gets to the point of disk space exhaustion and sudo starts getting uppity.
So far, I think the easiest way may be quota.
`zfs set quota=15G zroot/var/log`
That is similar to creating a separate zpool of size 15G and moving /var/log over to that zpool.
Stefano MarinelliI've just used them, so time to repost this:
Enhancing FreeBSD Stability with ZFS Pool Checkpoints
https://it-notes.dragas.net/2024/07/01/enhancing-freebsd-stability-with-zfs-pool-checkpoints/
vermadenLatest ๐ฉ๐ฎ๐น๐๐ฎ๐ฏ๐น๐ฒ ๐ก๐ฒ๐๐ - ๐ฎ๐ฌ๐ฎ๐ฑ/๐ญ๐ญ/๐ญ๐ณ (Valuable News - 2025/11/17) available.
https://vermaden.wordpress.com/2025/11/17/valuable-news-2025-11-17/
Past releases: https://vermaden.wordpress.com/news/
#verblog #vernews #news #bsd #freebsd #openbsd #netbsd #linux #unix #zfs #opnsense #ghostbsd #solaris #vermadenday
vermadenLatest ๐ฉ๐ฎ๐น๐๐ฎ๐ฏ๐น๐ฒ ๐ก๐ฒ๐๐ - ๐ฎ๐ฌ๐ฎ๐ฑ/๐ญ๐ญ/๐ญ๐ณ (Valuable News - 2025/11/17) available.
https://vermaden.wordpress.com/2025/11/17/valuable-news-2025-11-17/
Past releases: https://vermaden.wordpress.com/news/
#verblog #vernews #news #bsd #freebsd #openbsd #netbsd #linux #unix #zfs #opnsense #ghostbsd #solaris #vermadenday
AnthropyReminder that SMART is not magic, in my experience, disks are dead long before SMART stops reporting them as 'PASSED'.
I actually came across the perfect example today: Two disks in my backup server are having issues, one is clearly broken, the other gave 2 checksum errors.
Despite this, neither disk is 'failed' according to SMART, one has 456 offline uncorrectable errors / pending sectors, the other one is fine.
Don't rely on broken hardware to tell you that it's broken.
Practical ZFS - DiscussionsRecording filesystem activity?
The more disk benchmarking I try to do, the more I want to more accurately simulate the actual workloads running in my system. Iโm looking for a way that I can collect data about ZFS filesystem usage over a period of time to then craft fio benchmarks closer to what actually happens in the system.
I donโt need to get overly scientific - for the output Iโm looking for something like:
- % mix of reads vs writes
- block size the application tried to write
- average queue depth (how โbusyโ the system is)
Ideally Iโd let this run for a few days to collect a sufficient sample.
Does something like this exist?
1 post - 1 participant
Recording filesystem activity?
The more disk benchmarking I try to do, the more I want to more accurately simulate the actual workloads running in my system. Iโm looking for a way that I can collect data about ZFS filesystem usage over a period of time to then craft fio benchmarks closer to what actually happens in the system. I donโt need to get overly scientific - for the output Iโm looking for something like: % mix of reads vs writes block size the application tried to write average queue depth (how โbusyโ the system is...
nepi