FW: [SunHELP] Solaris 8 / fsck / SDS 4.2.1 / boot failed

Michael Karl mk at lexcom-net.de
Mon Feb 14 06:51:08 CST 2005


Some informations ...

Kernel version: SunOS 5.8 Generic 117350-18 Dec 2004

# more /etc/vfstab
#device         device          mount   FS      fsck    mount   mount
#to mount       to fsck         point   type    pass    at boot options
#
/proc   -       /proc   proc    -       no      -
/dev/md/dsk/d1  -       -       swap    -       no      -
/dev/md/dsk/d0  /dev/md/rdsk/d0 /       ufs     1       no      -
/dev/md/dsk/d6  /dev/md/rdsk/d6 /usr    ufs     1       no      -
/dev/md/dsk/d7  /dev/md/rdsk/d7 /star   ufs     2       yes     -
swap    -       /tmp    tmpfs   -       yes     -
# 

# metastat d7
d7: Trans
    State: Okay    
    Size: 57344700 blocks
    Master Device: d77
    Logging Device: d55

d77: Mirror
    Submirror 0: d70
      State: Okay  
    Submirror 1: d71
      State: Okay  
    Pass: 1
    Read option: roundrobin (default)
    Write option: parallel (default)
    Size: 57344700 blocks

d70: Submirror of d77
    State: Okay    
    Size: 57344700 blocks
    Stripe 0:
        Device     Start Block  Dbase State        Hot Spare
        c0t0d0s7          0     No    Okay


d71: Submirror of d77
    State: Okay    
    Size: 57344700 blocks
    Stripe 0:
        Device     Start Block  Dbase State        Hot Spare
        c0t1d0s7          0     No    Okay


d55: Logging device for d7
    State: Okay    
    Size: 2048754 blocks

d55: Mirror
    Submirror 0: d50
      State: Okay  
    Submirror 1: d51
      State: Okay  
    Pass: 1
    Read option: roundrobin (default)
    Write option: parallel (default)
    Size: 2049012 blocks

d50: Submirror of d55
    State: Okay    
    Size: 2049012 blocks
    Stripe 0:
        Device     Start Block  Dbase State        Hot Spare
        c0t0d0s5          0     No    Okay


d51: Submirror of d55
    State: Okay    
    Size: 2049012 blocks
    Stripe 0:
        Device     Start Block  Dbase State        Hot Spare
        c0t1d0s5          0     No    Okay
        
# format
Searching for disks...done


AVAILABLE DISK SELECTIONS:
       0. c0t0d0 <IBM-IC35L036UCDY10-0-S29A cyl 36313 alt 2 hd 3 sec 658>
          /pci at 1f,4000/scsi at 2/sd at 0,0
       1. c0t1d0 <IBM-IC35L036UCDY10-0-S29A cyl 36313 alt 2 hd 3 sec 658>
          /pci at 1f,4000/scsi at 2/sd at 1,0

... both disks

Current partition table (original):
Total disk cylinders available: 36313 + 2 (reserved cylinders)

Part      Tag    Flag     Cylinders         Size            Blocks
  0       root    wm    1063 -  2619        1.47GB    (1557/0/0)   3073518
  1       swap    wu       0 -  1062        1.00GB    (1063/0/0)   2098362
  2     backup    wm       0 - 36312       34.18GB    (36313/0/0) 71681862
  3 unassigned    wm    2620 -  2640       20.24MB    (21/0/0)       41454
  4 unassigned    wm    2641 -  2848      200.48MB    (208/0/0)     410592
  5 unassigned    wm    2849 -  3886     1000.49MB    (1038/0/0)   2049012
  6        usr    wm    3887 -  7262        3.18GB    (3376/0/0)   6664224
  7 unassigned    wm    7263 - 36312       27.34GB    (29050/0/0) 57344700


The partition 3 have the replicas ... 3 on both disks.

My customer need this setup with Solaris 8 ...
It couldn't be a TransMetaDeviceLogging-problem ... I've tested it.

My HomeOffice UE10 with !Sol9! has the same setup and is working fine.

The segmentation fault of fsck could be a hint ...
why does the manual fsck work ?

# fsck /dev/md/rdsk/d7
** /dev/md/rdsk/d7
** Last Mounted on /star
** Phase 1 - Check Blocks and Sizes
** Phase 2 - Check Pathnames
** Phase 3 - Check Connectivity
** Phase 4 - Check Reference Counts
** Phase 5 - Check Cyl groups
2 files, 9 used, 28221451 free (19 frags, 3527679 blocks,  0.0%
fragmentation)

Thanks Michael 

> Show me your vfstab file please. I am wondering why it works ok with /
> and /usr but not the other filesystem, the only reason that I can
> think of is something wrong in the vfstab file..
> 
> 
> On Mon, 14 Feb 2005 12:34:26 +0100, Michael Karl <mk at lexcom-net.de> wrote:
>> Thank you for the answer ...
>> but if I type  control+d the mirrored filesystem on slice 7
>> (c0t0d0s7 and c0t1d0s7) is correctly mounted 'read/write'!
>> So the vfstab-entry must be correct ...
>> 
>> I forgot to tell you, that the fsck has a segmentation fault
>> during checking filesystems at startup, but after this fsck
>> works fine manual.
>> 
>> I've installed the patches 109576-01 and 110910-03.
>> 
>> Any other hints ?
>> 
>> Best regards Michael
>> 
>> 
>>> Check your /etc/vfstab
>>> 
>>> 
>>> 
>>> On Mon, 14 Feb 2005 09:33:31 +0100, Michael Karl <mk at lexcom-net.de> wrote:
>>>> Hello,
>>>> 
>>>> I'd setting up an Ultra60-Cycle-Clone with Solaris 8/SDS 4.2.1
>>>> and two 36GB-IBM-Disks, also the latest Recommended-Patch-Cluster.
>>>> 
>>>> With SDS 4.2.1 inclusive 108693-25-Patch I have a problem.
>>>> I'm thinking, that I've had this problem in the past, but
>>>> now I don't remember the solution.
>>>> 
>>>> I'd setting up mirroring ...
>>>> it works on / and /usr fine, but not with one more filesystem.
>>>> 
>>>> The boot failed with a fsck-error. 'fsck' in single-user-mode doesn't
>>>> report
>>>> a problem and the start after this with 'crtl+d' is normal. The next boot
>>>> fails with the same error.
>>>> 
>>>> I googled, but find no hints :-(
>>>> 
>>>> Best regards Michael
>>>> _______________________________________________
>>>> SunHELP maillist  -  SunHELP at sunhelp.org
>>>> http://www.sunhelp.org/mailman/listinfo/sunhelp
>>>> 
>>> _______________________________________________
>>> SunHELP maillist  -  SunHELP at sunhelp.org
>>> http://www.sunhelp.org/mailman/listinfo/sunhelp
>> _______________________________________________
>> SunHELP maillist  -  SunHELP at sunhelp.org
>> http://www.sunhelp.org/mailman/listinfo/sunhelp
>> 



More information about the SunHELP mailing list