SDS: Beide Spiegelhaelften auf "Needs maintenance" HOWTO

[7ami7]

Fehler:
############
Die Disk c1t0d0 brachte Fehler und sollte proaktiv getauscht werden.
Beim Tausch verabschiedete sich dann auch der Slice 3 (/var) der zweiten Spiegelhaelfte.
Der Sync brach einfach ab:

# metattach d30 d31
d30: submirror d31 is attached
# while true; do metastat | grep %; echo; sleep 10; done
    Resync in progress: 3 % done

    Resync in progress: 5 % done

    Resync in progress: 6 % done

    Resync in progress: 8 % done

--> Dat wars dann ...:-(

Beide Spiegelhaelften stehen nun im Maintenance Mode:
(Ausgabe leider nicht mehr verfuegbar)




Loesung:
############
# metadetach d30 d31

# metastat d30
d30: Mirror
    Submirror 1: d32
      State: Needs maintenance
    Pass: 1
    Read option: roundrobin (default)
    Write option: parallel (default)
    Size: 12587712 blocks (6.0 GB)

d32: Submirror of d30
    State: Needs maintenance
    Invoke: after replacing "Maintenance" components:
                metareplace d30 c1t1d0s3 <new device>
    Size: 12587712 blocks (6.0 GB)
    Stripe 0:
        Device     Start Block  Dbase        State Reloc Hot Spare
        c1t1d0s3          0     No      Last Erred   Yes

#format

AVAILABLE DISK SELECTIONS:
       0. c1t0d0 <SUN72G cyl 14087 alt 2 hd 24 sec 424>
          /pci@1c,600000/scsi@2/sd@0,0
       1. c1t1d0 <SUN72G cyl 14087 alt 2 hd 24 sec 424>
          /pci@1c,600000/scsi@2/sd@1,0
Specify disk (enter its number): 1
selecting c1t1d0
[disk formatted]
...
format> analyze
...
analyze> read
Ready to analyze (won't harm SunOS). This takes a long time,
but is interruptable with CTRL-C. Continue? yes

        pass 0
Warning: unable to pinpoint defective block.
   14086/23/304 

        pass 1
Recoverable error during read: block 34714347 (0x211b2eb) (3411/9/195)
ASC: 0x17   ASCQ: 0x8
Recoverable error during read: block 34714348 (0x211b2ec) (3411/9/196)
ASC: 0x17   ASCQ: 0x8
Recoverable error during read: block 34714349 (0x211b2ed) (3411/9/197)
ASC: 0x17   ASCQ: 0x8
Medium error during read: block 34714353 (0x211b2f1) (3411/9/201)
ASC: 0x11   ASCQ: 0x1

Recoverable error during read: block 34714345 (0x211b2e9) (3411/9/193)
ASC: 0x17   ASCQ: 0x8
Medium error during read: block 34714353 (0x211b2f1) (3411/9/201)
ASC: 0x11   ASCQ: 0x1

Recoverable error during read: block 34714320 (0x211b2d0) (3411/9/168)
ASC: 0x17   ASCQ: 0x8
   14086/23/304 

Total of 0 defective blocks repaired.

--> Falls "read" Fehler findet, wiederholen.

analyze> read
Ready to analyze (won't harm SunOS). This takes a long time,
but is interruptable with CTRL-C. Continue? yes

        pass 0
   14086/23/304 

        pass 1
   14086/23/304 

Total of 0 defective blocks repaired.
analyze> quit
...
format> quit

Danach ist ein attachen der 2. Spiegelhaelfte moeglich.

# metattach d30 d31
=====================
d30: submirror d31 is attached
root@ds42101a # while true; do metastat | grep %; echo; sleep 10; done
    Resync in progress: 0 % done

    Resync in progress: 2 % done
    ...
    Resync in progress: 98 % done


# metastat d30
d30: Mirror
    Submirror 0: d31
      State: Okay         
    Submirror 1: d32
      State: Needs maintenance
    Pass: 1
    Read option: roundrobin (default)
    Write option: parallel (default)
    Size: 12587712 blocks (6.0 GB)

d31: Submirror of d30
    State: Okay         
    Size: 12587712 blocks (6.0 GB)
    Stripe 0:
        Device     Start Block  Dbase        State Reloc Hot Spare
        c1t0d0s3          0     No            Okay   Yes


d32: Submirror of d30
    State: Needs maintenance
    Invoke: after replacing "Maintenance" components:
                metareplace d30 c1t1d0s3 <new device>
    Size: 12587712 blocks (6.0 GB)
    Stripe 0:
        Device     Start Block  Dbase        State Reloc Hot Spare
        c1t1d0s3          0     No      Last Erred   Yes


Device Relocation Information:
Device   Reloc  Device ID
c1t0d0   Yes    id1,sd@SFUJITSU_MAT3073N_SUN72G_000509B01REK____AAN0P5301REK
c1t1d0   Yes    id1,sd@SFUJITSU_MAT3073N_SUN72G_000509B01S5V____AAN0P5301S5V


Danach kann die fehlerhafte Disk replaced werden.

# metareplace -e d30 c1t1d0s3
d30: device c1t1d0s3 is enabled