Hello,
before I tell you my problem, I would like to give you the background story to my problem:
I had an 8x 18TB RAIDZ2 pool, which was running fine for quite some time now. The pool was filled with 80%, so I decided to upgrade my pool with another raidz2 8x18TB VDEV, to double my existing storage and double performance.
Before my upgrade, I was running those 8 drives inside a fractal case with 8 bays, connected to a HBA card. Due to the missing of another 8 bays, I decided to run those new 8 drives inside of 2x external 4bay enclosures, each with one fan in front. I extended the SATA power cables from my PSU inside the case and connected the new 8 drives to a new LSI SAS 9200-16e with 2x SFF8088 to SATA cables.
I was lazy and didn't do a full scan/smart of my new 18TB drives, so I just imported them and started to rebalance my pool (move data between datasets).
This huge file transfer operation was running for about 2 days straight, without any problems, until this afternoon: 2 of my new drives showed many write and read errors (about 50-100), the pool status was degraded.
I stopped the file transfer operation, restarted the server and tried to do a short smart test of all drives simultaneously. During the test i noticed, that at least 1 drive was re spinning over and over again, so I aborted this test.
Then i pulled each of those new 18TB HDDs, connected them to my home pc and did a short smart test and a short read performance test. Everything looked fine, none of the drives were re spinning.
Then I switched out, all SATA cables, connected to different SATA power cables, connected to a different HBA. Nevertheless, each change didn't solve the problem with re spinning disks....
On top of all, after the change of the HBA, I thought it was not re spinning every time, so i started a scrub. At first, it was running fine, but then the re spinning process began again and 1 disk didn't appear at all, so my pool was stated unavailable.
I'm very frustrated, because i don't have a full backup of my pool and much data on this pool isn't backed up.
How would you try to rescue this data? There is a lot panic now, so before I do anything stupid, I would like to ask you guys first. Thank you in advance.
before I tell you my problem, I would like to give you the background story to my problem:
I had an 8x 18TB RAIDZ2 pool, which was running fine for quite some time now. The pool was filled with 80%, so I decided to upgrade my pool with another raidz2 8x18TB VDEV, to double my existing storage and double performance.
Before my upgrade, I was running those 8 drives inside a fractal case with 8 bays, connected to a HBA card. Due to the missing of another 8 bays, I decided to run those new 8 drives inside of 2x external 4bay enclosures, each with one fan in front. I extended the SATA power cables from my PSU inside the case and connected the new 8 drives to a new LSI SAS 9200-16e with 2x SFF8088 to SATA cables.
I was lazy and didn't do a full scan/smart of my new 18TB drives, so I just imported them and started to rebalance my pool (move data between datasets).
This huge file transfer operation was running for about 2 days straight, without any problems, until this afternoon: 2 of my new drives showed many write and read errors (about 50-100), the pool status was degraded.
I stopped the file transfer operation, restarted the server and tried to do a short smart test of all drives simultaneously. During the test i noticed, that at least 1 drive was re spinning over and over again, so I aborted this test.
Then i pulled each of those new 18TB HDDs, connected them to my home pc and did a short smart test and a short read performance test. Everything looked fine, none of the drives were re spinning.
Then I switched out, all SATA cables, connected to different SATA power cables, connected to a different HBA. Nevertheless, each change didn't solve the problem with re spinning disks....
On top of all, after the change of the HBA, I thought it was not re spinning every time, so i started a scrub. At first, it was running fine, but then the re spinning process began again and 1 disk didn't appear at all, so my pool was stated unavailable.
I'm very frustrated, because i don't have a full backup of my pool and much data on this pool isn't backed up.
How would you try to rescue this data? There is a lot panic now, so before I do anything stupid, I would like to ask you guys first. Thank you in advance.
Last edited:
