Associate
If you run any kind of RAID/Unraid/TrueNAS etc, do you keep a spare disk so you can replace a failed one straight away?
Running a server with any type of RAID - do you keep a spare disk on hand?
- Thread starter nst68
- Start date
Don
Depends on how important the data is and what RAID level you are running.
My Synology at home with 3 disks in SHR (RAID5 equivalent) - no I do not have a spare, the most important data is backed up in 2 other places anyway.
At work, the majority of my servers run RAID10 arrays, the majority of which have a hot spare installed installed ready and configured to automatically rebuild, and we also have multiple cold spare drives on site.
My Synology at home with 3 disks in SHR (RAID5 equivalent) - no I do not have a spare, the most important data is backed up in 2 other places anyway.
At work, the majority of my servers run RAID10 arrays, the majority of which have a hot spare installed installed ready and configured to automatically rebuild, and we also have multiple cold spare drives on site.
Soldato
I'm not sure if you're asking about home or enterprise level, but yes we run hot spares in all of our SANs, multiple hot spares and we have 4 hour same day SLA for disk failures should we lose one.
Associate
We don't. But instead we have a daily checklist to go through every day, one of items is to check health of all SAN arrays, we get automatic alerts should anything go south and we have 24/7 support person under phone with 4h vendor support/replacement.
HDD failures nowadays are not that common, if you have RAID-10 (RAID-5 at minimum) and 3-2-1 backups, you're golden
HDD failures nowadays are not that common, if you have RAID-10 (RAID-5 at minimum) and 3-2-1 backups, you're golden
Soldato
I agree in some respect, we see a lot less disk failures, but we have a tendency to see when one disk goes, others tend to be on the way out or go at around a similar time too, hence why our hot spares are always there. We have about 60 arrays so fortunately we have a lot of replication too which helps.
Soldato
Nope, I don't keep spares on my Home NASs (2 of them). If a drive fails, I can go a buy one same day. I don't see th point of having money sitting on a shelf.
For a business environment, It all depends on what you are storing, how important it is and if it's backed up.. Generally I like to keep multiple hot spares in an array, but I'm not talking about small arrays like a Synology NAS, I'm talking large Enterprise SANs.
As for raid 10... I've never been a fan, personally I run everything RAID-6, even at home.
For a business environment, It all depends on what you are storing, how important it is and if it's backed up.. Generally I like to keep multiple hot spares in an array, but I'm not talking about small arrays like a Synology NAS, I'm talking large Enterprise SANs.
As for raid 10... I've never been a fan, personally I run everything RAID-6, even at home.
Soldato
For the home stuff I try and watch the EOL dates for drives to make sure I get an identical model before they are discontinued or hard to come by as a spare.
Ive recently gone from 1 to 2 disk redundancy in my synology NAS and also have a hot spare as well. Just for piece of mind really as much as anything else.
As mentioned above, the biggest risk it seems is just when the array is recovering from one failed disk and re-writing any spare disk , this is the most likely time for a 2nd disk to fail (or this is prevailing wisdom as far as I understand it)
but you can only go SO far at home (and of course it depends on what "value" the data has for you).
As mentioned above, the biggest risk it seems is just when the array is recovering from one failed disk and re-writing any spare disk , this is the most likely time for a 2nd disk to fail (or this is prevailing wisdom as far as I understand it)
but you can only go SO far at home (and of course it depends on what "value" the data has for you).