I got several 2 TB HDDs, I think they got just A FEW bad sectors.
Obviously they cant hold my data individually, for safe.
They are unsafe even as backup's.
I wanna ask, will it fit with snapraid?
I got some old/old copy of useful files,
(files generated by freefilesync's versioning),
and even finally have SOME data loss, it's acceptable.
I plan to use snapraid of 3 parity disk + 3 data disk.
in which (2 bad HDD + 1 "no bad sector but old" HDD) for parity,
and the 3 "no bad sector but old" HDD for data disk of old data.
I would like to ask comment if would this be useful for those bad hdd?
thanks.
If you would like to refer to this comment somewhere else in this project, copy and paste the following link:
My understanding is that any modern drive should reallocate bad sectors to spare areas within the next write. If bad sectors persist multiple writes it means it run out of space in the spare area. If drives reach this condition, it's really time to replace them, as they must have developed a tremendous number of bad sectors over time. You can use nwipe from the PartedMagic iso to write and verify random data. Give them a few cycles.
if it rejects more than 1 (for normal parity) or 2 (for dual parity) then your data is gone (including if this happens during the reconstruction that can take days - note even if you don't replace a drive after drobo fails a drive it shuffles a lot of data across the remaining disks)
the format is proprietary and you can't get anything back if something is wrong, apart from the fact that you can't just copy your files from individual disks like for snapraid.
Incidentally I've got a drobo pro just to store all my older and smaller drives...
Such failed disks usually are just not worth the time. Now if you insist snapraid would be probably the best place to have them - it has absolutely all the things you would want in such scenario:
all disks independent
nothing proprietary
it doesn't work at block level but instead at file level; it doesn't need to scan the whole disks, if some file is unreadable you at least know what file you're looking for and can (maybe) get it from some other place, etc
However still I wouldn't bother keeping failed disks in a working system. No matter the improvements over the last 20 (30?) years disk reads are still somehow blocking (a retry on a disk can lock up your system at least for a while) and each time you boot you'll get the "smart error" from BIOS. You can disable SMART from BIOS (bad idea) or get used to it (still bad idea...). If you feel you must use it put some checksumming filesystem on it (btrfs for linux or ReFS for windows but remember to enable integrity, it isn't on by default) and use it for offline backups, just connect it to the PC from time to time to make some backups overnight and take it out next day.
If you would like to refer to this comment somewhere else in this project, copy and paste the following link:
I got several 2 TB HDDs, I think they got just A FEW bad sectors.
Obviously they cant hold my data individually, for safe.
They are unsafe even as backup's.
I wanna ask, will it fit with snapraid?
I got some old/old copy of useful files,
(files generated by freefilesync's versioning),
and even finally have SOME data loss, it's acceptable.
I plan to use snapraid of 3 parity disk + 3 data disk.
in which (2 bad HDD + 1 "no bad sector but old" HDD) for parity,
and the 3 "no bad sector but old" HDD for data disk of old data.
I would like to ask comment if would this be useful for those bad hdd?
thanks.
My understanding is that any modern drive should reallocate bad sectors to spare areas within the next write. If bad sectors persist multiple writes it means it run out of space in the spare area. If drives reach this condition, it's really time to replace them, as they must have developed a tremendous number of bad sectors over time. You can use nwipe from the PartedMagic iso to write and verify random data. Give them a few cycles.
Last edit: a2885989 2018-01-26
I wouldn't trust them in a manual RAID system. I would trust them in one of my drobos.
Drobo is absolutely not a good fit:
Incidentally I've got a drobo pro just to store all my older and smaller drives...
Such failed disks usually are just not worth the time. Now if you insist snapraid would be probably the best place to have them - it has absolutely all the things you would want in such scenario:
However still I wouldn't bother keeping failed disks in a working system. No matter the improvements over the last 20 (30?) years disk reads are still somehow blocking (a retry on a disk can lock up your system at least for a while) and each time you boot you'll get the "smart error" from BIOS. You can disable SMART from BIOS (bad idea) or get used to it (still bad idea...). If you feel you must use it put some checksumming filesystem on it (btrfs for linux or ReFS for windows but remember to enable integrity, it isn't on by default) and use it for offline backups, just connect it to the PC from time to time to make some backups overnight and take it out next day.
Been shuffling drives with bad sectors to 8 different drobos for almost 10 years. Not a problem yet.
I wouldn't recommend writing data to a known bad drive.