summery, Sol. 9 SMC - Creating extends and mirroring the root harddisc

From: LOEWENTHAL Simon (sloewenthal_at_gemini.edu)
Date: 08/31/04

  • Next message: ed.rolison_at_itc.alstom.com: "Rediscover disk size and geometry." 
    To: "Sun Mangers" <sunmanagers@sunmanagers.org>
Date: Tue, 31 Aug 2004 10:32:39 -0400

    SUMMARY:

    Hi all of you who responded. Thanks for the advise. Here is whats happened
    so far.

    I thought that it was strange that vold had control of the device, but the
    discs are:
    /pci@1f,0/pci@1/scsi@8
    Target 0
      Unit 0 Disk FUJITSU MAP3147N SUN146G0501
    Target 1
      Unit 0 Disk FUJITSU MAP3147N SUN146G0501

    I have removed the CDrom drive and 'boot -r'ed the system. Now vold doesn't
    try and take control of the harddisc anymore, but a metadb still gives:

    metadb -a -f c0t0d0s7
    metadb: test1: /dev/dsk/c0t0d0s7: No such device or address

    ...But its here, and I can access it via format.
    [root@test1]root$ ls -l /dev/dsk/c0t0d0s7
       2 lrwxrwxrwx 1 root root 45 Aug 27 12:36
    /dev/dsk/c0t0d0s7 -> ../../devices/pci@1f,0/pci@1,1/ide@d/sd@0,0:h

    The slice is 9.94Mb in size. I thought that this was fine for a state dB,
    but Darren recommends >50Mb.
    If I start SMC and display the disc partitions it does not pick up this
    slice. Is the slice too small for SMC to notice? The format cmd says that
    slice 7 only occupies one cylinder or 20352 blocks. With this in mind I
    changed slice 7 to 70mb and ran the metadb command again, but got the same
    results.

    Next I started the SMC and tried to create the state dBs that way, and low
    and behold it works.
    [root@test1]root$ metadb
            flags first blk block count
         a u 16 8192 /dev/dsk/c1t1d0s7
         a u 16 8192 /dev/dsk/c1t0d0s7

    Looks like the cmd line commands won't work at all so, unhappily, I'm going
    to have to use the GUI (yuk:(
    [root@test1]root$ metainit -f d1 1 1 c0t0d0s0
    metainit: test1: /dev/dsk/c0t0d0s0: No such device or address

    Many thanks to the following:
    Martin Foster
    DarrenProg
    William Hathaway
    David.Harrington
    Chad Truhn
    Tobias Nutt

    Rdgs/Atentamente,
    Simon. 

    ---
Simon Loewenthal
Gemini Observatory
Information Systems Group
Video 139.229.33.88
Voice +56 51 205610
-----Original Message-----
From: William D. Hathaway [mailto:wdh@perfectorder.com]
Sent: Tuesday, August 31, 2004 8:16 AM
To: LOEWENTHAL Simon
Subject: Re: metadb/newfs on disc: If vold running = Device busy/ If
vold not running = No such device or address.
Are you sure that your disk is on c0t0d0s0 and not the cdrom?  Vold
having a partition open seems much more likely on a cdrom device, as
well as a cdrom may not have any space for the partition 7.
Can you provide output from the following:
metastat
df -k
prtvtoc /dev/rdsk/*s0
On Mon, 2004-08-30 at 18:25, LOEWENTHAL Simon wrote:
> Dear all,
>
> I'm trying to metadb a slice to a new statedatabase.  As per the table
below
> the slice is 9.94Mb.  Machine is Solaris 9.
>
> Current partition table (original):
> Total disk cylinders available: 14087 + 2 (reserved cylinders)
>
> Part      Tag    Flag     Cylinders         Size            Blocks
>   0       root    wm     105 -   457        3.43GB    (353/0/0)
7184256
>   1       swap    wu       0 -   103        1.01GB    (104/0/0)
2116608
>   2     backup    wm       0 - 14086      136.71GB    (14087/0/0)
286698624
>   3        var    wm     459 -   473      149.06MB    (15/0/0)
305280
>   4 unassigned    wm       0                0         (0/0/0)
0
>   5 unassigned    wm       0                0         (0/0/0)
0
>   6 unassigned    wm       0                0         (0/0/0)
0
>   7 unassigned    wm   14086 - 14086        9.94MB    (1/0/0)
20352
>
> I next used metadb to create the first state dB:
>
> [root@test1]root$ metadb -a -f /dev/rdsk/c0t0d0s7
> metadb: test1: /dev/dsk/c0t0d0s7: No such device or address
>
> OK, Odd. So I tried to do a newfs on the partition:
> [root@test1]root$ newfs /dev/rdsk/c0t0d0s7
> /dev/rdsk/c0t0d0s7: Device busy
>
> An fuser shows me that vold pid is using that partition (Que?)  so I kill
> vold.
>
> [root@test1]root$ /etc/init.d/volmgt stop
> [root@test1]root$ !newfs
> newfs /dev/rdsk/c0t0d0s7
> /dev/rdsk/c0t0d0s7: No such device or address
> [root@test1]root$ !metadb
> metadb -a -f /dev/rdsk/c0t0d0s7
> metadb: test1: /dev/dsk/c0t0d0s7: No such device or address
>
> I reboot just to check its the same, and it is.  Any clues?
>
> Thanks to everyone who replied to the SMC / mirroring / extents question.
> Rdgs/Atentamente,
> Simon.
>
> ---
> Simon Loewenthal
> Gemini Observatory
> Information Systems Group
> Video 139.229.33.88
> Voice +56 51 205610
>
> ---
> Outgoing mail is certified Virus Free.
> Checked by AVG anti-virus system (http://www.grisoft.com).
> Version: 6.0.745 / Virus Database: 497 - Release Date: 27/08/2004
> _______________________________________________
> sunmanagers mailing list
> sunmanagers@sunmanagers.org
> http://www.sunmanagers.org/mailman/listinfo/sunmanagers
--
William D. Hathaway    email: wdh@perfectorder.com
Consultant             AIM:   wdhPO
Perfect Order Inc.     cell:  717-314-5461
---
Incoming mail is certified Virus Free.
Checked by AVG anti-virus system (http://www.grisoft.com).
Version: 6.0.745 / Virus Database: 497 - Release Date: 27/08/2004
---
Outgoing mail is certified Virus Free.
Checked by AVG anti-virus system (http://www.grisoft.com).
Version: 6.0.745 / Virus Database: 497 - Release Date: 27/08/2004
_______________________________________________
sunmanagers mailing list
sunmanagers@sunmanagers.org
http://www.sunmanagers.org/mailman/listinfo/sunmanagers
  • Next message: ed.rolison_at_itc.alstom.com: "Rediscover disk size and geometry."

    Relevant Pages