Reallocated sector count

[SpaceInvaderOne]

Hi ,

 

How serious is having a Reallocated sector count raw value of 1. I just got this warning on the smart. Should i replace the drive asap or could i just acknowledge the error and see if it gets worse??

 

# Attribute Name Flag Value Worst Threshold Type Updated Failed Raw Value

1 Raw read error rate 0x000f 079 063 044 Pre-fail Always Never 84007277

3 Spin up time 0x0003 092 089 070 Pre-fail Always Never 0

4 Start stop count 0x0032 100 100 020 Old age Always Never 412

5 Reallocated sector count 0x0033 100 100 036 Pre-fail Always Never 1 

[itimpi]

There is nothing wrong with having a reallocated sector as long as the value remains constant.  Modern drives are designed to reallocate sectors if one fails.  If it keeps increasing then the drive needs replacing.

[SpaceInvaderOne]

Thanks for advise. I will keep an eye on it

[isvein]

I have an Samsung 750GB that has an runtime of 4y, 8m, 24d, 11h, and has 1relocated sector.

As long as it does not increase and/or Current pending sector(s) is at 0, its not much to worry about

[garycase]

As noted, there's nothing wrong with re-allocated sectors.  All modern drives are DESIGNED to reallocate any defective sectors to a spare location.    A much more serious issue is if you start getting "pending" sectors ... these are defective sectors that are in use and can't be reallocated because the drive can't successfully read them to reallocate the contents.    If you start getting these, I'd replace the drive ASAP.    But if you simply have a few reallocated sectors, the drive is fine.    As already suggested, I'd simply watch to be sure the count doesn't start getting out of control.  An occasional increase in the number is okay; but if it starts growing rapidly you likely have a major area of the disk that has become defective and I'd then replaced the drive.

 

[TUMS]

What's wrong with leaving a drive in the array until it's totally dead? Dual parity support might make this safer no?  Reading the freenas zfs forum that's what a lot of those users do.  Or is this tactic not really advisable with unraid?

[itimpi]

What's wrong with leaving a drive in the array until it's totally dead? Dual parity support might make this safer no?  Reading the freenas zfs forum that's what a lot of those users do.  Or is this tactic not really advisable with unraid?

what is the point in leaving a dead drive in the array.  Once you get any write errors to a drive unRAID will stop writing to it so no point in keeping it in the array.    You also increase the chance of encountering a multiple drive failure scenario which will lead to data loss.

[TUMS]

What's wrong with leaving a drive in the array until it's totally dead? Dual parity support might make this safer no?  Reading the freenas zfs forum that's what a lot of those users do.  Or is this tactic not really advisable with unraid?

what is the point in leaving a dead drive in the array.  Once you get any write errors to a drive unRAID will stop writing to it so no point in keeping it in the array.    You also increase the chance of encountering a multiple drive failure scenario which will lead to data loss.

 

I meant keeping it in till it dies.  As in reallocated sectors could go into the 100's and you'd still keep running with it.

[BRiT]

What's wrong with leaving a drive in the array until it's totally dead? Dual parity support might make this safer no?  Reading the freenas zfs forum that's what a lot of those users do.  Or is this tactic not really advisable with unraid?

what is the point in leaving a dead drive in the array.  Once you get any write errors to a drive unRAID will stop writing to it so no point in keeping it in the array.    You also increase the chance of encountering a multiple drive failure scenario which will lead to data loss.

 

I meant keeping it in till it dies.  As in reallocated sectors could go into the 100's and you'd still keep running with it.

 

Heck, don't even bother with parity. It's your data, take whatever risks you want with your data.

[itimpi]

What's wrong with leaving a drive in the array until it's totally dead? Dual parity support might make this safer no?  Reading the freenas zfs forum that's what a lot of those users do.  Or is this tactic not really advisable with unraid?

what is the point in leaving a dead drive in the array.  Once you get any write errors to a drive unRAID will stop writing to it so no point in keeping it in the array.    You also increase the chance of encountering a multiple drive failure scenario which will lead to data loss.

.

 

I meant keeping it in till it dies.  As in reallocated sectors could go into the 100's and you'd still keep running with it.

my experience is that if the reallocated sectors keeps growing then the drive is likely to fail.  There is nothing intrinsically wrong with having a significant number of reallocated sectors as long as the number stays relatively stable.  It is regular increases which are a sign of trouble.

 

As has been mentioned of far more importance is the value for pending sectors as any non-zero value means you may not be able to recover a different disk that fails without some level of data corruption

[TUMS]

What's wrong with leaving a drive in the array until it's totally dead? Dual parity support might make this safer no?  Reading the freenas zfs forum that's what a lot of those users do.  Or is this tactic not really advisable with unraid?

what is the point in leaving a dead drive in the array.  Once you get any write errors to a drive unRAID will stop writing to it so no point in keeping it in the array.    You also increase the chance of encountering a multiple drive failure scenario which will lead to data loss.

.

 

I meant keeping it in till it dies.  As in reallocated sectors could go into the 100's and you'd still keep running with it.

my experience is that if the reallocated sectors keeps growing then the drive is likely to fail.  There is nothing intrinsically wrong with having a significant number of reallocated sectors as long as the number stays relatively stable.  It is regular increases which are a sign of trouble.

 

As has been mentioned of far more importance is the value for pending sectors as any non-zero value means you may not be able to recover a different disk that fails without some level of data corruption

 

Thanks.  I'd probably be tempted to run it up into the 100's (reallocated sectors) if I had dual parity.  I mean if I can get a few more months out of the drive then hell, why not. Especially if I had a backup server.  Without that I wouldn't try it. Also depends if the drive is still under warranty or not, if it is, probably best to get it replaced sooner rather than later.

[garycase]

As noted above, there's nothing intrinsically wrong or "bad" with a high reallocated sector count as long as it's not continuing to grow.  If it's growing, then that's a good indication that the drive is failing; and while you COULD wait for the actual failure, it seems that the prudent thing to do is to be a bit more pro-active.  The whole idea of SMART is to provide a health check for your drives -- so if it's showing bad things happening, I'd think you'd want to take some action based on that.

 

[TUMS]

As noted above, there's nothing intrinsically wrong or "bad" with a high reallocated sector count as long as it's not continuing to grow.  If it's growing, then that's a good indication that the drive is failing; and while you COULD wait for the actual failure, it seems that the prudent thing to do is to be a bit more pro-active.  The whole idea of SMART is to provide a health check for your drives -- so if it's showing bad things happening, I'd think you'd want to take some action based on that.

 

Thanks garycase, agree totally.  I just don't want to throw away perfectly good drives when I don't have to for a few bad sectors.  Sometimes it's hard to tell how long a drive is going to last if all that's showing bad in smart is reallocated sectors.  I've had drives with 50+ reallocated sectors (in desktop pc's) and they still operated just fine for months/years afterwords.  One nice thing about having a backup server, takes a little bit of the pressure off.  You can afford to take a few more risks than you would otherwise.

[garycase]

... One nice thing about having a backup server, takes a little bit of the pressure off.  You can afford to take a few more risks than you would otherwise.

 

It's certainly true that if you have proper backups the risk of waiting for an actual failure is somewhat mitigated.  Between that and dual parity you have very little likelihood of data loss when a drive fails.

 

... it continually surprises me, however, how many folks consider their data important enough to build a fault-tolerant server but not important enough to back up !!

 

[daze]

Useful thread, my older drives are picking up single reallocated sectors. I'm keeping an eye on them, and if that value increase, I'm replacing these 5 year old 1 to 2TB WD green drives. But not before time. I have dual parity, and only one drive is showing 1 reallocation at the moment.

[garycase]

On ‎5‎/‎3‎/‎2018 at 11:40 AM, daze said:

Useful thread, my older drives are picking up single reallocated sectors. I'm keeping an eye on them, and if that value increase, I'm replacing these 5 year old 1 to 2TB WD green drives. But not before time. I have dual parity, and only one drive is showing 1 reallocation at the moment.

 

With only a single reallocated sector I wouldn't be at all concerned -- especially if the count doesn't in crease.   And of course with dual parity you're well protected should the drive suddenly decide to fail.    I never replace a drive just because of a small # of reallocated sectors; as long as the count stays static.    If it gets higher than I like, but is still a stable number, I'll replace the drive and use the one with the reallocated sectors for storing off-line backups.

[tr0910]

I have had many single sector issues and since I have a hot spare, I yank the offending drive and rebuild to the spare. I them start a pre clear on the offending drive, and most of the time the offending sector returns good. False alarm??

Sent from my Nexus 6 using Tapatalk

[itimpi]

On 4/3/2016 at 10:13 PM, TUMS said:

What's wrong with leaving a drive in the array until it's totally dead? Dual parity support might make this safer no?  Reading the freenas zfs forum that's what a lot of those users do.  Or is this tactic not really advisable with unraid?

 

The problem arises when a different drive fails.   Unraid reads all data drives plus parity to successfully rebuild a drive. 

 

If the the drive is really dead then you may just get away with the rebuild if you have dual parity.  If it is not dead but just unreliable then it may return bad data during a rebuild which would result in the contents of the rebuilt disk having some level of corruption (and where may not be obvious).

[jowi]

ID# ATTRIBUTE_NAME          FLAGS    VALUE WORST THRESH FAIL RAW_VALUE
  1 Raw_Read_Error_Rate     PO-R--   100   100   016    -    0
  2 Throughput_Performance  P-S---   136   136   054    -    80
  3 Spin_Up_Time            POS---   149   149   024    -    557 (Average 481)
  4 Start_Stop_Count        -O--C-   099   099   000    -    5491
  5 Reallocated_Sector_Ct   PO--CK   100   100   005    -    7
  7 Seek_Error_Rate         PO-R--   100   100   067    -    0
  8 Seek_Time_Performance   P-S---   114   114   020    -    37
  9 Power_On_Hours          -O--C-   089   089   000    -    82897
 10 Spin_Retry_Count        PO--C-   100   100   060    -    0
 12 Power_Cycle_Count       -O--CK   100   100   000    -    140
192 Power-Off_Retract_Count -O--CK   096   096   000    -    5604
193 Load_Cycle_Count        -O--C-   096   096   000    -    5604
194 Temperature_Celsius     -O----   230   230   000    -    26 (Min/Max 18/55)
196 Reallocated_Event_Count -O--CK   100   100   000    -    9
197 Current_Pending_Sector  -O---K   100   100   000    -    0
198 Offline_Uncorrectable   ---R--   100   100   000    -    0
199 UDMA_CRC_Error_Count    -O-R--   200   200   000    -    0

My oldest drive (Hitachi HDS724040ALE640, 4TB) has 7 reallocated sectors, showed up a few days ago. With over 10 years of service, pretty good  if it starts to increase, i will replace it with an 8TB one. This is my parity drive btw.

Edited January 17, 2022 by jowi

[paloooz]

One of my drives was at 7 reallocated sectors for about a year. I acknowledged the error and kept using the drive as normal. Within the last month, the error count grew to 132 and then 164. At that point, I stopped all write operations to the array and began migrating data out of that drive using unbalance. During the migration, the error count grew once more to 1440. There is 1TB of data left to move, and the parity is protecting me. I'm going to replace this drive.

 

In another server, my parity drive now has 24 reallocated sectors. This one is a bit trickier to replace because my data will not be redundant while parity rebuilds. I will report back here in the future with what I experienced as the drives failed. I have never seen a drive fail before, so I want to take these ones to the end.

 

Edited May 20, 2022 by paloooz

[ChristianMingle]

I have a sun F80 drive that i bought used that has 60k hours on it with 117 reallocated sectors on all 4 of the 200GB drives ... im trying to use it and see what happens but im not putting anything valuable on it .. just curious how long itll last lol

[itimpi]

On 5/20/2022 at 6:20 PM, paloooz said:

One of my drives was at 7 reallocated sectors for about a year. I acknowledged the error and kept using the drive as normal. Within the last month, the error count grew to 132 and then 164. At that point, I stopped all write operations to the array and began migrating data out of that drive using unbalance. During the migration, the error count grew once more to 1440. There is 1TB of data left to move, and the parity is protecting me. I'm going to replace this drive.

In such a situation I am not sure you want to ‘move’ the data to other drives as a ‘move’ implies write operations on the drive.    I would prefer to simply ‘copy’ the data elsewhere to avoid any writing to the suspect drive.