#openzfs — Public Fediverse posts
Live and recent posts from across the Fediverse tagged #openzfs, aggregated by home.social.
-
@pertho Some years back we (my work) licensed SPECsfs2014, the SPEC consortium's file-system benchmark. We still weren't able to get much out of it, because the SPEC workload models weren't very realistic, or at least didn't relate usefully to our users' workloads. I've discussed at a couple of #OpenZFS user summits the need for both better workload models and a mechanism to *build* models from real-life workloads rather than armchair theorizing.
-
@pertho Some years back we (my work) licensed SPECsfs2014, the SPEC consortium's file-system benchmark. We still weren't able to get much out of it, because the SPEC workload models weren't very realistic, or at least didn't relate usefully to our users' workloads. I've discussed at a couple of #OpenZFS user summits the need for both better workload models and a mechanism to *build* models from real-life workloads rather than armchair theorizing.
-
@pertho Some years back we (my work) licensed SPECsfs2014, the SPEC consortium's file-system benchmark. We still weren't able to get much out of it, because the SPEC workload models weren't very realistic, or at least didn't relate usefully to our users' workloads. I've discussed at a couple of #OpenZFS user summits the need for both better workload models and a mechanism to *build* models from real-life workloads rather than armchair theorizing.
-
@pertho Some years back we (my work) licensed SPECsfs2014, the SPEC consortium's file-system benchmark. We still weren't able to get much out of it, because the SPEC workload models weren't very realistic, or at least didn't relate usefully to our users' workloads. I've discussed at a couple of #OpenZFS user summits the need for both better workload models and a mechanism to *build* models from real-life workloads rather than armchair theorizing.
-
@pertho Some years back we (my work) licensed SPECsfs2014, the SPEC consortium's file-system benchmark. We still weren't able to get much out of it, because the SPEC workload models weren't very realistic, or at least didn't relate usefully to our users' workloads. I've discussed at a couple of #OpenZFS user summits the need for both better workload models and a mechanism to *build* models from real-life workloads rather than armchair theorizing.
-
We had THE TALK on the #OpenZFS production user call regarding pragmatic encrypted dataset management.
-
We had THE TALK on the #OpenZFS production user call regarding pragmatic encrypted dataset management.
-
We had THE TALK on the #OpenZFS production user call regarding pragmatic encrypted dataset management.
-
We had THE TALK on the #OpenZFS production user call regarding pragmatic encrypted dataset management.
-
We had THE TALK on the #OpenZFS production user call regarding pragmatic encrypted dataset management.
-
The recording of the July 8th, 2026 #OpenZFS Production User Call is up:
We discussed Rob Norris' recent commits and active projects, OpenZFS security handling, the "mountpoint" property, a long-overdue deep dive into practical encrypted dataset management, desired encryption features, and more!
"Don't forget to slam those Like and Subscribe buttons."
You can support all Call For Testing efforts via BSD Fund: https://bsdfund.org
-
The recording of the July 8th, 2026 #OpenZFS Production User Call is up:
We discussed Rob Norris' recent commits and active projects, OpenZFS security handling, the "mountpoint" property, a long-overdue deep dive into practical encrypted dataset management, desired encryption features, and more!
"Don't forget to slam those Like and Subscribe buttons."
You can support all Call For Testing efforts via BSD Fund: https://bsdfund.org
-
The recording of the July 8th, 2026 #OpenZFS Production User Call is up:
We discussed Rob Norris' recent commits and active projects, OpenZFS security handling, the "mountpoint" property, a long-overdue deep dive into practical encrypted dataset management, desired encryption features, and more!
"Don't forget to slam those Like and Subscribe buttons."
You can support all Call For Testing efforts via BSD Fund: https://bsdfund.org
-
The recording of the July 8th, 2026 #OpenZFS Production User Call is up:
We discussed Rob Norris' recent commits and active projects, OpenZFS security handling, the "mountpoint" property, a long-overdue deep dive into practical encrypted dataset management, desired encryption features, and more!
"Don't forget to slam those Like and Subscribe buttons."
You can support all Call For Testing efforts via BSD Fund: https://bsdfund.org
-
The recording of the July 8th, 2026 #OpenZFS Production User Call is up:
We discussed Rob Norris' recent commits and active projects, OpenZFS security handling, the "mountpoint" property, a long-overdue deep dive into practical encrypted dataset management, desired encryption features, and more!
"Don't forget to slam those Like and Subscribe buttons."
You can support all Call For Testing efforts via BSD Fund: https://bsdfund.org
-
🤔🤪 "How to Build a Minimal #ZFS #NAS Without Synology, QNAP, TrueNAS" is the epic saga of someone who thinks slapping together some random hardware and sprinkling in "OpenZFS" is the ultimate life hack. 🛠️😅 Because, clearly, the world needed another guide on how to be a tech hipster by not using the perfectly good tools that already exist. 🤷♂️
https://neil.computer/notes/how-to-setup-minimal-zfs-nas-without-truenas/ #OpenZFS #TechHacks #DIYStorage #Minimalism #HackerNews #ngated -
🤔🤪 "How to Build a Minimal #ZFS #NAS Without Synology, QNAP, TrueNAS" is the epic saga of someone who thinks slapping together some random hardware and sprinkling in "OpenZFS" is the ultimate life hack. 🛠️😅 Because, clearly, the world needed another guide on how to be a tech hipster by not using the perfectly good tools that already exist. 🤷♂️
https://neil.computer/notes/how-to-setup-minimal-zfs-nas-without-truenas/ #OpenZFS #TechHacks #DIYStorage #Minimalism #HackerNews #ngated -
🤔🤪 "How to Build a Minimal #ZFS #NAS Without Synology, QNAP, TrueNAS" is the epic saga of someone who thinks slapping together some random hardware and sprinkling in "OpenZFS" is the ultimate life hack. 🛠️😅 Because, clearly, the world needed another guide on how to be a tech hipster by not using the perfectly good tools that already exist. 🤷♂️
https://neil.computer/notes/how-to-setup-minimal-zfs-nas-without-truenas/ #OpenZFS #TechHacks #DIYStorage #Minimalism #HackerNews #ngated -
🤔🤪 "How to Build a Minimal #ZFS #NAS Without Synology, QNAP, TrueNAS" is the epic saga of someone who thinks slapping together some random hardware and sprinkling in "OpenZFS" is the ultimate life hack. 🛠️😅 Because, clearly, the world needed another guide on how to be a tech hipster by not using the perfectly good tools that already exist. 🤷♂️
https://neil.computer/notes/how-to-setup-minimal-zfs-nas-without-truenas/ #OpenZFS #TechHacks #DIYStorage #Minimalism #HackerNews #ngated -
🤔🤪 "How to Build a Minimal #ZFS #NAS Without Synology, QNAP, TrueNAS" is the epic saga of someone who thinks slapping together some random hardware and sprinkling in "OpenZFS" is the ultimate life hack. 🛠️😅 Because, clearly, the world needed another guide on how to be a tech hipster by not using the perfectly good tools that already exist. 🤷♂️
https://neil.computer/notes/how-to-setup-minimal-zfs-nas-without-truenas/ #OpenZFS #TechHacks #DIYStorage #Minimalism #HackerNews #ngated -
I am experiencing this with #OpenZFS
➜ ~ sudo zpool status -x
pool: ughetto
state: DEGRADED
status: One or more devices could not be used because the label is missing or
invalid. Sufficient replicas exist for the pool to continue
functioning in a degraded state.
action: Replace the device using 'zpool replace'.
see: https://openzfs.github.io/openzfs-docs/msg/ZFS-8000-4J
scan: scrub in progress since Sat Jun 27 18:44:31 2026
141G / 141G scanned, 11.8G / 141G issued at 184M/s
0B repaired, 8.39% done, 00:12:00 to go
config:
NAME STATE READ WRITE CKSUM
ughetto DEGRADED 0 0 0
mirror-0 DEGRADED 0 0 0
653332104939013944 UNAVAIL 0 0 0 was /dev/sdb1
sdc ONLINE 0 0 0
cache
sda FAULTED 0 0 0 corrupted data
errors: No known data errorsThis has already happened, but at a point where I didn’t mind losing the data, so I destroyed the pool and I recreated it from scratch, with no errors. Now, I would like to try to repair the corrupted data… Any tips? I would hate it if I have to throw away the hard disk, it’s less than one year old and I have been using it only for a couple of months… Is replacing the disk my only option?
#ZFS #systemAdministration #help #sysAd #techSupport #dataLoss #storage #partition #hardDisk #HDD #Linux #filesystem #data #storage
-
I am experiencing this with #OpenZFS
➜ ~ sudo zpool status -x
pool: ughetto
state: DEGRADED
status: One or more devices could not be used because the label is missing or
invalid. Sufficient replicas exist for the pool to continue
functioning in a degraded state.
action: Replace the device using 'zpool replace'.
see: https://openzfs.github.io/openzfs-docs/msg/ZFS-8000-4J
scan: scrub in progress since Sat Jun 27 18:44:31 2026
141G / 141G scanned, 11.8G / 141G issued at 184M/s
0B repaired, 8.39% done, 00:12:00 to go
config:
NAME STATE READ WRITE CKSUM
ughetto DEGRADED 0 0 0
mirror-0 DEGRADED 0 0 0
653332104939013944 UNAVAIL 0 0 0 was /dev/sdb1
sdc ONLINE 0 0 0
cache
sda FAULTED 0 0 0 corrupted data
errors: No known data errorsThis has already happened, but at a point where I didn’t mind losing the data, so I destroyed the pool and I recreated it from scratch, with no errors. Now, I would like to try to repair the corrupted data… Any tips? I would hate it if I have to throw away the hard disk, it’s less than one year old and I have been using it only for a couple of months… Is replacing the disk my only option?
#ZFS #systemAdministration #help #sysAd #techSupport #dataLoss #storage #partition #hardDisk #HDD #Linux #filesystem #data #storage
-
I am experiencing this with #OpenZFS
➜ ~ sudo zpool status -x
pool: ughetto
state: DEGRADED
status: One or more devices could not be used because the label is missing or
invalid. Sufficient replicas exist for the pool to continue
functioning in a degraded state.
action: Replace the device using 'zpool replace'.
see: https://openzfs.github.io/openzfs-docs/msg/ZFS-8000-4J
scan: scrub in progress since Sat Jun 27 18:44:31 2026
141G / 141G scanned, 11.8G / 141G issued at 184M/s
0B repaired, 8.39% done, 00:12:00 to go
config:
NAME STATE READ WRITE CKSUM
ughetto DEGRADED 0 0 0
mirror-0 DEGRADED 0 0 0
653332104939013944 UNAVAIL 0 0 0 was /dev/sdb1
sdc ONLINE 0 0 0
cache
sda FAULTED 0 0 0 corrupted data
errors: No known data errorsThis has already happened, but at a point where I didn’t mind losing the data, so I destroyed the pool and I recreated it from scratch, with no errors. Now, I would like to try to repair the corrupted data… Any tips? I would hate it if I have to throw away the hard disk, it’s less than one year old and I have been using it only for a couple of months… Is replacing the disk my only option?
#ZFS #systemAdministration #help #sysAd #techSupport #dataLoss #storage #partition #hardDisk #HDD #Linux #filesystem #data #storage
-
I am experiencing this with #OpenZFS
➜ ~ sudo zpool status -x
pool: ughetto
state: DEGRADED
status: One or more devices could not be used because the label is missing or
invalid. Sufficient replicas exist for the pool to continue
functioning in a degraded state.
action: Replace the device using 'zpool replace'.
see: https://openzfs.github.io/openzfs-docs/msg/ZFS-8000-4J
scan: scrub in progress since Sat Jun 27 18:44:31 2026
141G / 141G scanned, 11.8G / 141G issued at 184M/s
0B repaired, 8.39% done, 00:12:00 to go
config:
NAME STATE READ WRITE CKSUM
ughetto DEGRADED 0 0 0
mirror-0 DEGRADED 0 0 0
653332104939013944 UNAVAIL 0 0 0 was /dev/sdb1
sdc ONLINE 0 0 0
cache
sda FAULTED 0 0 0 corrupted data
errors: No known data errorsThis has already happened, but at a point where I didn’t mind losing the data, so I destroyed the pool and I recreated it from scratch, with no errors. Now, I would like to try to repair the corrupted data… Any tips? I would hate it if I have to throw away the hard disk, it’s less than one year old and I have been using it only for a couple of months… Is replacing the disk my only option?
#ZFS #systemAdministration #help #sysAd #techSupport #dataLoss #storage #partition #hardDisk #HDD #Linux #filesystem #data #storage
-
I am experiencing this with #OpenZFS
➜ ~ sudo zpool status -x
pool: ughetto
state: DEGRADED
status: One or more devices could not be used because the label is missing or
invalid. Sufficient replicas exist for the pool to continue
functioning in a degraded state.
action: Replace the device using 'zpool replace'.
see: https://openzfs.github.io/openzfs-docs/msg/ZFS-8000-4J
scan: scrub in progress since Sat Jun 27 18:44:31 2026
141G / 141G scanned, 11.8G / 141G issued at 184M/s
0B repaired, 8.39% done, 00:12:00 to go
config:
NAME STATE READ WRITE CKSUM
ughetto DEGRADED 0 0 0
mirror-0 DEGRADED 0 0 0
653332104939013944 UNAVAIL 0 0 0 was /dev/sdb1
sdc ONLINE 0 0 0
cache
sda FAULTED 0 0 0 corrupted data
errors: No known data errorsThis has already happened, but at a point where I didn’t mind losing the data, so I destroyed the pool and I recreated it from scratch, with no errors. Now, I would like to try to repair the corrupted data… Any tips? I would hate it if I have to throw away the hard disk, it’s less than one year old and I have been using it only for a couple of months… Is replacing the disk my only option?
#ZFS #systemAdministration #help #sysAd #techSupport #dataLoss #storage #partition #hardDisk #HDD #Linux #filesystem #data #storage
-
reboot -r
– no longer reboots the operating system. Not reproducible with FreeBSD 15.1-RELEASE.
In the ALT text here (auto-detected):
Trying to mount root from zfs:maximal/ROOT/sixteen []...
FreeBSD 16.0-CURRENT main-n286985-90ea8e89d9b7 GENERIC-NODEBUG amd64 1600019 1600019
<https://cgit.freebsd.org/src/log/?qt=range&q=90ea8e89d9b7&h=main>, I have not attempted to bisect.
Seeking RB_REROOT in GitHub finds two commits, the most recent (December 2025):
<https://github.com/freebsd/freebsd-src/commit/4453ec5b8716bc465ff5192986099dc75d1f2ce7>
– is that the regression, or am I way off? According to cgit, it has not been cherry-picked to releng/15.1 <https://cgit.freebsd.org/src/log/?qt=grep&q=reboot%3A+Default+to+a+clean+shutdown&h=releng%2F15.1> or stable/15.
-
reboot -r
– no longer reboots the operating system. Not reproducible with FreeBSD 15.1-RELEASE.
In the ALT text here (auto-detected):
Trying to mount root from zfs:maximal/ROOT/sixteen []...
FreeBSD 16.0-CURRENT main-n286985-90ea8e89d9b7 GENERIC-NODEBUG amd64 1600019 1600019
<https://cgit.freebsd.org/src/log/?qt=range&q=90ea8e89d9b7&h=main>, I have not attempted to bisect.
Seeking RB_REROOT in GitHub finds two commits, the most recent (December 2025):
<https://github.com/freebsd/freebsd-src/commit/4453ec5b8716bc465ff5192986099dc75d1f2ce7>
– is that the regression, or am I way off? According to cgit, it has not been cherry-picked to releng/15.1 <https://cgit.freebsd.org/src/log/?qt=grep&q=reboot%3A+Default+to+a+clean+shutdown&h=releng%2F15.1> or stable/15.
-
reboot -r
– no longer reboots the operating system. Not reproducible with FreeBSD 15.1-RELEASE.
In the ALT text here (auto-detected):
Trying to mount root from zfs:maximal/ROOT/sixteen []...
FreeBSD 16.0-CURRENT main-n286985-90ea8e89d9b7 GENERIC-NODEBUG amd64 1600019 1600019
<https://cgit.freebsd.org/src/log/?qt=range&q=90ea8e89d9b7&h=main>, I have not attempted to bisect.
Seeking RB_REROOT in GitHub finds two commits, the most recent (December 2025):
<https://github.com/freebsd/freebsd-src/commit/4453ec5b8716bc465ff5192986099dc75d1f2ce7>
– is that the regression, or am I way off? According to cgit, it has not been cherry-picked to releng/15.1 <https://cgit.freebsd.org/src/log/?qt=grep&q=reboot%3A+Default+to+a+clean+shutdown&h=releng%2F15.1> or stable/15.
-
reboot -r
– no longer reboots the operating system. Not reproducible with FreeBSD 15.1-RELEASE.
In the ALT text here (auto-detected):
Trying to mount root from zfs:maximal/ROOT/sixteen []...
FreeBSD 16.0-CURRENT main-n286985-90ea8e89d9b7 GENERIC-NODEBUG amd64 1600019 1600019
<https://cgit.freebsd.org/src/log/?qt=range&q=90ea8e89d9b7&h=main>, I have not attempted to bisect.
Seeking RB_REROOT in GitHub finds two commits, the most recent (December 2025):
<https://github.com/freebsd/freebsd-src/commit/4453ec5b8716bc465ff5192986099dc75d1f2ce7>
– is that the regression, or am I way off? According to cgit, it has not been cherry-picked to releng/15.1 <https://cgit.freebsd.org/src/log/?qt=grep&q=reboot%3A+Default+to+a+clean+shutdown&h=releng%2F15.1> or stable/15.
-
reboot -r
– no longer reboots the operating system. Not reproducible with FreeBSD 15.1-RELEASE.
In the ALT text here (auto-detected):
Trying to mount root from zfs:maximal/ROOT/sixteen []...
FreeBSD 16.0-CURRENT main-n286985-90ea8e89d9b7 GENERIC-NODEBUG amd64 1600019 1600019
<https://cgit.freebsd.org/src/log/?qt=range&q=90ea8e89d9b7&h=main>, I have not attempted to bisect.
Seeking RB_REROOT in GitHub finds two commits, the most recent (December 2025):
<https://github.com/freebsd/freebsd-src/commit/4453ec5b8716bc465ff5192986099dc75d1f2ce7>
– is that the regression, or am I way off? According to cgit, it has not been cherry-picked to releng/15.1 <https://cgit.freebsd.org/src/log/?qt=grep&q=reboot%3A+Default+to+a+clean+shutdown&h=releng%2F15.1> or stable/15.
-
Manipulating a Pool from the Rescue System - Klara Systems
https://klarasystems.com/articles/manipulating-a-pool-from-the-rescue-system/
-
Manipulating a Pool from the Rescue System - Klara Systems
https://klarasystems.com/articles/manipulating-a-pool-from-the-rescue-system/
-
Manipulating a Pool from the Rescue System - Klara Systems
https://klarasystems.com/articles/manipulating-a-pool-from-the-rescue-system/
-
Manipulating a Pool from the Rescue System - Klara Systems
https://klarasystems.com/articles/manipulating-a-pool-from-the-rescue-system/
-
Manipulating a Pool from the Rescue System - Klara Systems
https://klarasystems.com/articles/manipulating-a-pool-from-the-rescue-system/
-
@oxy I enjoyed a killing on Kubuntu 26.04 a day or so ago.
IIRC free memory and swap were nearly exhausted (unrelated to ARC) due to VirtualBox throwing a wobbly with a FreeBSD guest, following a disrupted connection to an old mobile hard disk drive on USB, which I use (with OpenZFS) to store the virtual hard disks for many of my test machines.
I feared that Firefox would be killed.
Happily: Linux made VirtualBox, which was not responding, the victim.
-
@oxy I enjoyed a killing on Kubuntu 26.04 a day or so ago.
IIRC free memory and swap were nearly exhausted (unrelated to ARC) due to VirtualBox throwing a wobbly with a FreeBSD guest, following a disrupted connection to an old mobile hard disk drive on USB, which I use (with OpenZFS) to store the virtual hard disks for many of my test machines.
I feared that Firefox would be killed.
Happily: Linux made VirtualBox, which was not responding, the victim.
-
@oxy I enjoyed a killing on Kubuntu 26.04 a day or so ago.
IIRC free memory and swap were nearly exhausted (unrelated to ARC) due to VirtualBox throwing a wobbly with a FreeBSD guest, following a disrupted connection to an old mobile hard disk drive on USB, which I use (with OpenZFS) to store the virtual hard disks for many of my test machines.
I feared that Firefox would be killed.
Happily: Linux made VirtualBox, which was not responding, the victim.
-
@oxy I enjoyed a killing on Kubuntu 26.04 a day or so ago.
IIRC free memory and swap were nearly exhausted (unrelated to ARC) due to VirtualBox throwing a wobbly with a FreeBSD guest, following a disrupted connection to an old mobile hard disk drive on USB, which I use (with OpenZFS) to store the virtual hard disks for many of my test machines.
I feared that Firefox would be killed.
Happily: Linux made VirtualBox, which was not responding, the victim.
-
@oxy I enjoyed a killing on Kubuntu 26.04 a day or so ago.
IIRC free memory and swap were nearly exhausted (unrelated to ARC) due to VirtualBox throwing a wobbly with a FreeBSD guest, following a disrupted connection to an old mobile hard disk drive on USB, which I use (with OpenZFS) to store the virtual hard disks for many of my test machines.
I feared that Firefox would be killed.
Happily: Linux made VirtualBox, which was not responding, the victim.
