I have a Synology NAS with 5 2TB drives in it. One is making an
annoying clunk... clunk... clunk... noise every 5-10 seconds.
SMART stats look OK, although one drive has 9 years of run time on it!
The 5 drives are in an SHR array with 1 drive's worth of redundancy.
I want to replace the oldest drive, thinking it's the one making the
noise -- but would feel really stupid if another drive failed while I
did that! --- SBBSecho 3.20-Win32
* Origin: realitycheckbbs.org -- yesterday's tech today (46:1/115)
Roll up a piece of paper, put it against your ear then the other end against each drive and you might be able to figure which is giving the loudest clunking.
I was just going to tell him to turn the music up louder so you didn't hear it, much like you should also do when your car starts making wierd noises. :D
Accession wrote to nelgin <=-
I was just going to tell him to turn the music up louder so you didn't hear it, much like you should also do when your car starts making wierd noises. :D
I was just going to tell him to turn the music up louder so youThat reminds me, I need to install an amplifier...
didn't hear it, much like you should also do when your car starts
making wierd noises.
I usually work with headphones on, listening to ambient tracks - but
every once in a while I don't - that's when I hear it.
I have a Synology NAS with 5 2TB drives in it. One is making anRoll up a piece of paper, put it against your ear then the other end against each drive and you might be able to figure which is giving the loudest clunking.
annoying clunk... clunk... clunk... noise every 5-10 seconds.
SMART stats look OK, although one drive has 9 years of run time on
it!
The 5 drives are in an SHR array with 1 drive's worth of redundancy.
I want to replace the oldest drive, thinking it's the one making the
noise -- but would feel really stupid if another drive failed while
I did that!
telnet://bbs.roonsbbs.hu:1212 <<=-
Anywho, why would you think that another drive would fail while you replace the bad one?
Anywho, why would you think that another drive would fail while you
replace the bad one?
Murphy, combined with old drives. 82K, 27K, 3K, 64K and 78K hours.
Accession wrote to poindexter FORTRAN <=-
On Fri, Nov 01 23:42:36 -0500, you wrote:
Murphy, combined with old drives. 82K, 27K, 3K, 64K and 78K hours.
Looks like you may need 3 drives, then. Do one, cross fingers. Rinse
and repeat. :)
Anywho, why would you think that another drive would fail while
you replace the bad one?
Murphy, combined with old drives. 82K, 27K, 3K, 64K and 78K
hours.
Looks like you may need 3 drives, then. Do one, cross fingers. Rinse
and repeat. :)
My server is fairly old, and I don't even know how to look up how many hours each hard drive has been used. But there's no clunking, yet, so here's hoping we just let it ride and hope for the best!
telnet://bbs.roonsbbs.hu:1212 <<=-
Murphy, combined with old drives. 82K, 27K, 3K, 64K and 78K
hours.
Looks like you may need 3 drives, then. Do one, cross fingers.
Rinse and repeat. :)
And that begs the question - these are all 2TB drives. Do I look at
larger drives? Apparently, you can replace RAID elements with larger drives, and once they're all replaced, expand the RAID volume to use
the unused space.
Or, I could get larger drives, wipe the volume, create a volume with
fewer disks and save space for a hot-swap?
telnet://bbs.roonsbbs.hu:1212 <<=-
And that begs the question - these are all 2TB drives. Do I look at
larger drives? Apparently, you can replace RAID elements with larger drives, and once they're all replaced, expand the RAID volume to use the unused space.
Or, I could get larger drives, wipe the volume, create a volume with
fewer disks and save space for a hot-swap?
My server is fairly old, and I don't even know how to look up how many
hours each hard drive has been used. But there's no clunking, yet, so
here's hoping we just let it ride and hope for the best!
smart tells you.
on linux you can check:
# smartctl -a /dev/sda
[...]
9 Power_On_Hours 0x0032 021 021 000 Old_age
Always - 57908
[...]
57908/24 = 2412 days, approx 6 years :)
Anywho, why would you think that another drive would fail while you replace the bad one?
Or, I could get larger drives, wipe the volume, create a volume with
fewer disks and save space for a hot-swap?
jinkusu wrote to poindexter FORTRAN <=-
Or, I could get larger drives, wipe the volume, create a volume with
fewer disks and save space for a hot-swap?
2tb drives? depending on how many bays you have left... just create a
new volume, copy all your data to it, and remove the old drives TBH.
all of your data would fit on one drive at this point.
if you don't have the bays left to create a new volume i'd get an
external disk adapter, temporarily copy your data to a new disk, remove the existing array completely, and then build a new volume from new
disks.
Murphy, combined with old drives. 82K, 27K, 3K, 64K and 78K hours.
Murphy, combined with old drives. 82K, 27K, 3K, 64K and 78K hours.
Oddly, I swapped the first drive, but smartctl doesn't reflect the
correct stats for the new drive - unless I bought a drive that also had 82k hours on it!
Do you have to do some sort of resync after you swap, by chance? Or should i automatically detect all of that?
It's all GUI-based. It detects that a drive is missing, that there's
a new drive in there, and wouldn't you like to rebuild your array?
It rebuilds the drive from the parity info, took about 4 hours.
I ended up buying 3 more 2TB drives, going to swap them out as I go.
Accession wrote to poindexter FORTRAN <=-
No matter what, it wouldn't refresh the new drive's life cycle? Did you buy it new or used?
Ac> No matter what, it wouldn't refresh the new drive's life cycle? Did you
Ac> buy it new or used?
It took a reboot, now it shows 89 hours. Phew!
Sysop: | Eric Oulashin |
---|---|
Location: | Beaverton, Oregon, USA |
Users: | 91 |
Nodes: | 16 (0 / 16) |
Uptime: | 17:21:54 |
Calls: | 5,074 |
Calls today: | 6 |
Files: | 8,491 |
Messages: | 352,928 |
Posted today: | 1 |