1

We have a RAID 5 using a 3ware 9550SX RAID controller that shows one degraded disk. There is a spare drive configured for the array, but it wasn't used automatically as intended.

The current array status is the following:

> /c0 show

Unit  UnitType  Status         %RCmpl  %V/I/M  Stripe  Size(GB)  Cache  AVrfy
------------------------------------------------------------------------------
u0    RAID-5    DEGRADED       -       -       64K     11175.8   OFF    OFF
u1    SPARE     OK             -       -       -       1863.01   -      OFF

Port   Status           Unit   Size        Blocks        Serial
---------------------------------------------------------------
p0     OK               u0     1.82 TB     3907029168    WD-WMAY0108****
p1     OK               u0     1.82 TB     3907029168    WD-WMAY0080****
p2     OK               u0     1.82 TB     3907029168    WD-WMAY0109****
p3     OK               u0     1.82 TB     3907029168    WD-WMAY0080****
p4     DEVICE-ERROR     u0     1.82 TB     3907029168    WD-WMAY0108****
p5     OK               u0     1.82 TB     3907029168    WD-WMAY0109****
p6     OK               u0     1.82 TB     3907029168    WD-WMAY0080****
p7     OK               u1     1.82 TB     3907029168    WD-WCAY0188****

Name  OnlineState  BBUReady  Status    Volt     Temp     Hours  LastCapTest
---------------------------------------------------------------------------
bbu   On           Yes       OK        OK       OK       0      xx-xxx-xxxx

> /c0/u1 show

Unit     UnitType  Status         %RCmpl  %V/I/M  Port  Stripe  Size(GB)
------------------------------------------------------------------------
u1       SPARE     OK             -       -       p7    -       1863.01

> /c0/u0 show

Unit     UnitType  Status         %RCmpl  %V/I/M  Port  Stripe  Size(GB)
------------------------------------------------------------------------
u0       RAID-5    DEGRADED*      -       -       -     64K     11175.8
u0-0     DISK      OK             -       -       p6    -       1862.63
u0-1     DISK      OK             -       -       p5    -       1862.63
u0-2     DISK      DEGRADED       -       -       p4    -       1862.63
u0-3     DISK      OK             -       -       p3    -       1862.63
u0-4     DISK      OK             -       -       p2    -       1862.63
u0-5     DISK      OK             -       -       p1    -       1862.63
u0-6     DISK      OK             -       -       p0    -       1862.63
u0/v0    Volume    -              -       -       -     -       11175.8

When trying to manually trigger a rebuild with tw_cli we encountered the following error:

> /c0/u0 start rebuild disk=p7
Error: (CLI:144) Invalid drive(s) specified.

We tried a few more variations of that command (/c0/u0 start rebuild disk=7 and maint rebuild c0 u0 p7), but it does not seem to be a syntax error or a mistake in specifying the drive. For some reason the RAID controller doesn't want to use this specific drive to rebuild.

Does the spare drive have to be in the same unit? Or could it be an issue that the spare is from a different batch? How can I get the array to use the spare we have to rebuild the array?

2

Before start rebuilding of array, you need configure controller settings. This command will enable automatic rebuilding:

> /c0 set autorebuild=on

Remove bad disk 4 from array:

> /c0/p4 remove

Use next command to identify bad disk by led indicator:

> /c0/p4 set identify=on

Eject bad disk from server and drop it into recycle bin. If array rebuilding won't start automatically then start it manually:

> /c0/u0 start rebuild disk=7

When you will insert new disk as hot spare enter these commands:

> /c0 rescan
> /c0/p4 add type=spare
  • 1
    Notice how Mikhail said disk=7 not disk=p7. 3ware decided on a different convention for disks then the convention used for controllers, units and ports. Fun, eh? – Stefan Lasiewski Aug 31 '16 at 20:58
  • I'm not sure but you can try start rebuilding on unit 1, witch is spare unit ` > /c0/u1 start rebuild disk=7` – Mikhail Khirgiy Sep 1 '16 at 5:34
  • removing the degraded drive triggered the autorebuild, so it is working now. Though I still don't really understand why the configured hot spare wasn't used automatically, and why the manual rebuild command failed. – user17363 Sep 1 '16 at 6:51
  • I think this situation was be because the 4-th disk was in degraded state, not fault. You should test this disk, it may be good. Check its smart attributes. The error on RAID controller was happened by any reason: bad contact between disk and cable, bad disk, bad cable, the disk was ejected and inserted, sudden power off, static electric discharge. May be the bad disk has long spin up time. – Mikhail Khirgiy Sep 1 '16 at 12:16
  • The 3ware firmwares do have bugs, some of which were never fixed. We've had drives fail in ways that confused the controller, and autorebuild would sometimes only start after we pulled a drive. Keep in mind that the 3ware 9550SX is pretty old (10+ years old?). – Stefan Lasiewski Sep 1 '16 at 17:04

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.