So this should do it:
mdadm --create /dev/md0 --level1 0 --name raid10 -raid-devices 4 /dev/nvm1n1 /dev/nvm2n1 /dev/nvme3n1 /dev/nvme4n1
The reason for the nested raid0's was that I read that mdadm level10 wasn't strictly raid10 and works with 2 and 3 drives too. That makes me...
No more warranty I'm afraid. They are recertified HGST Ultrastars. Offtopic but I will never get recertified drives again. 1 died in days (disappeared completely), 2 leaked helium like a balloon. Those 3 got replaced under warranty. Yet another drive dropped to ~70% helium in a matter...
So, yeah you're right I just had to discover it for myself;).
I triggered these manually:
SMART Self-test log structure revision number 1
Num Test_Description Status Remaining LifeTime(hours) LBA_of_first_error
# 1 Extended offline Completed: read failure 90%...
Thanks, appreciate the reply. I would appreciate it even more if people could back their suspicions with reasoning :cool:
I actually have some experience with those kind of issues and I don't think either one of those is the case. When I ran a pool of disks on a system with an PSU that barely...
Cheers, appreciate the reply. Not to be argumentive or anything, but that one has been there for like forever, way before these zfs errors occurred.
In the past 25 minutes:
NAME STATE READ WRITE CKSUM
Moving and migrating data to make changes to my ZFS pool, one drive is acting up. ZFS showing:
# zpool status
status: One or more devices has experienced an unrecoverable error. An
attempt was made to correct the error. Applications are unaffected.
These don't take a normal/half height bracket to go straight in a pci slot? I need to look up what a chassis backplate is then, and whether I can use that.
Plus new cables ... 4x 30 euros. Its getting expensive lol
I have been thinking about these, I found some for sale but all the cables have straight plugs into the OCUlink connectors. This will take 2 perhaps even 3 slots (depending on how bendable the plugs are) and I have yet to see small or 90 degree angled OCUlink plugs. Even with angled, it depends...
Now that my new U.2 SSD's are operational I added them as a mirror to my 4-mirror spinning rust pool as a special vdev for metadata and small blocks. For the moment I set it to 64k.
Ideally, I would like to figure out the current block sizes per dataset so I can set recordsize to the best value...
Thanks, appreciate youre all thinking along here.
I don't know where you got that riser card from but from here:
A backplane is basically an extension, female/male or vice-versa. A...
Thanks @UhClem . Is there a way to find out about that legacy / sff-9401 standard?
@RolloZ170 I have 2 cables, both don't work. Both SSD's are new, came sealed and all.
I have 2 PCIe cards that can mount U.2 SSD's coming in tomorrow and 2 new different brand OCUlink cables. I will try those...
I read quite a bit of trouble with cables and OCUlink, ASrock, Supermicro ...
I hooked up 2x Intel DC P4510 to the OCUlink connectors using a SFF-8611 to SFF-8639 U.2 NVME PCIe cable ... its this one:
Drives are not...
I am weighing some options. :P .
I own a Asus Z11PA-U12/10G-2S motherboard.
Currently running a bunch of spinning disks which I might want to replace with SSD's.
I also have 2x U.2 SSD's which I am about to connect to the OCUlink connectors.
The motherboard has 3 x miniSAS SATA...
Thanks for sharing.
It could very well be that the results would be reversed if you had tested under Linux with ext4 and fio. Not that it invalidates your results because they were taken on Windows with NTFS and different tooling (on the contrary), it does however make them incomparable to the...