** Reply to message from Felix Miata <[EMAIL PROTECTED]> on Thu, 01 Nov 2007
14:39:48 -0400


> On 2007/11/01 19:52 (GMT+0200) Stan Goodman apparently typed:
> 
> > I do not know how you got the log file to get all that detail. The one I 
> > have
> 
> All that detail is because I have more partitions than you. Otherwise,
> they're the same output, with mine reduced somewhat by manual editing.

That's what I meant by "my hardware is much simpler than yours".

> > is much simpler, even allowing for the  fact that my hardware is much 
> > simplet
> > than yours. Given its size, I think it's OK to just post it here. Here is 
> > the
> > file:
> 
> >  P-Geo Disk 1 Cyl : 16383 H: 16 S:63  Bps:512   Size : 0x09962B80 = 78533.4 
> > MiB
> >  Geometry D1 from : LVM info (DLAT) sector at 0x3e
> >  L-Geo Disk 1 Cyl : 10011 H:255 S:63  Bps:512   Size : 0x0996055B = 78528.7 
> > MiB
> >  BIOS Int13 limit : 1024, I13X support needed beyond : 8032.5 MiB
> >  MBR crc 054b4eb9 : 0x0c8ca699 = DFSee generic MBR, English messages, I13X
> 
> > DFSee  OS/2  9.02 : executing: fdisk -r-
> > +---+--+-----------------+--+--------+--------+-----------+----------+--------+
> > |ID |Dr|Type, description|ux|Format  |Related |VolumeLabel|LVM Volume|Size 
> > MiB|
> > +---+--<disk  1></dev/hda   >--------+--------+-----------<[ D1 ]    
> > >--------+
> > |01>|  |Prim 0a IBM-BMGR | 2|BMGR    |LVM     |I13Xneeded |., BootMan|     
> > 7.8|
> > |02 |  |Log  82 SunS/SWAP| 5|SWAP    |LinuxV1 |SWAPSPACE2 |, LinuxSwa|   
> > 502.0|
> > |03 |  |Log  83 LinuxNatv| 6|XFS     |Linux   |           |SuSE, SuSE| 
> > 50007.0|
> > |04*|  |Log  83 LinuxNatv| 7|EXT2    |GRUB    |           |SuSE v10.2|     
> > 7.8|
> > |05 |  |FreeSpace Logical|  |-- -- --|-- -- --|- - - - - -|          | 
> > 21987.4|
> > |05*|C:|Log  07 Inst-FSys| 8|HPFS    |IBM 4.50|ECS        |eCS v1.1, |  
> > 1004.0|
> > |06 |D:|Log  35 Warp-LVM | 9|JFS     |IBM 4.50|           |OS/2 Apps,|  
> > 1506.1|
> > |07 |H:|Log  35 Warp-LVM |10|JFS     |IBM 4.50|INFO       |Info, Info|  
> > 2502.3|
> > |08 |W:|Log  35 Warp-LVM |11|JFS     |IBM 4.50|DATAFILES  |DataFiles,|  
> > 1004.0|
> > +---+--+-----------------+--+--------+--------+-----------+----------+--------+
> 
> > Despite what the heading says, I HAVE registered it, but not yet put the key
> > file in -- I am leaving the country in three days, and am panicked with 
> > things
> > left to do, which is why this could not have happened at a worse time.
> 
> > Jan has explained to me that DFSee calls ALL disks e.g. "hda" rather than
> > "sda", because it doesn't see the interface anyway. I did not ask him what
> > happens in a system with both serial and parallel HDs.
> 
> When run from Linux, DFSee can easily tell the difference between PATA and
> SCSI, but I'm not sure about telling the difference between SATA & SCSI.
> 
> > If I have misunderstood your instructions, please give me more detailed 
> > ones,
> > and I will follow them as well as I can.
> 
> This was close enough, except that you didn't include /etc/grub.conf.

That's because I didn't know how to get it, since I have no access to SuSE. I
am very new in Linux, as you have surely detected.

In the interest of saving time, which is at a premium right now, if I can
retrieve a few things that are not backed up, I think it might be best to just
reinstall v10..3 from scratch. I have a Ubuntu v6.06LTS disk, and as I
understand it I can boot with it, and gain access to the /home directory where
those small things reside. Or I could do the same with a Rescue boot from the
10.3 DVD. Can I impose on you, or Joe, or anybody else, to let me run past you
the actions I need to do to accomplish the retrieval of these files?

Joe Morris gave me some hints about using Rescue for a different purpose:

*****
>> mount the root partition of the drive, i.e. mount /dev/md0 /mnt
>> (Not sure if these are still needed in 10.3 or not)
>> mount -o bind /proc /mnt/proc
>> mount -o bind /dev /mnt/dev
>> mount -o bind /sys /mnt/sys
>> cd mnt
>> chroot /mnt
*****

That was for using YaST. Which of the above "mount" lines are necessary for the
different purpose I have now?

Now I want to access the /home directory and also a JFS partition that I use
for transfer between SuSE and OS/2. In normal SuSE operation, that partition
is,  "/mnt/transfer", so I would add the line "mount -o bind /mnt
/mnt/transfer".  Have I generalized correctly from Joe's notes?

I would also add a line like "mount -o bind /home /mnt/home", which would cover
any subdirectories. I would copy the needed files. Being root, I would not have
to deal with permissions.

Is all that correct? 

so I would add another line to the above as follows: <mount -o bind /min>

Thank you for the advice below. I will apply these hints when I reinstall the
OS.

> I don't know what went wrong with the install, or what is wrong now. What
> first I'd do if it was here would be to delete ID 4 and immediately recreate
> it as 200MiB instead of 7.8MiB. Kernels and initrds have gotten so large that
> such small /boot partitions are not safe at upgrade time, or certainly if
> wanting multiple kernels and/or initrds to be available. The RedHat/Fedora
> installer will scream and holler at so small a boot partition, recommending
> at least 75M be allocated to it. I used to allocate 78M for /boot, but now do
> 200M.
> 
> Anyway, after recreating it, you need to rescue boot or boot a Linux live CD
> and run resize2fs on /dev/sda7 to enable access to all of its new size. After
> doing that, I'd try a rescue boot mode attempt to reinstall grub, either
> --batch using the /etc/grub.conf file, or manually from the grub prompt. If
> that wouldn't work, I'd do a reinstall.
> 
> Because you have so much freespace, it would be prudent to use some for a new
> ext3 partition to use from a rescue boot to copy all of /home. That way you
> could do an install from scratch of either 10.2 or 10.3. Ext3 is accessible
> via an OS/2 boot by using the ext2 driver from Hobbes. AFAIK, neither XFS nor
> ReiserFS have OS/2 drivers, and certainly Linux LVM does not.
> 
> If I was starting nearly all over, I'd first back up /home as above, then
> delete all linux partitions except the /home backup. Then I'd make the
> logical first Linux partition a *primary* ext2 of 200M, next a logical swap
> (sda5), then a much smaller root (sda6, minimum 4G, up to maybe 10G, more if
> you're a developer), then permanent /home (sda7).
> 
> I'm not sold on the idea that any other type is better than ext3 for the
> average user.
> 
> Also because you have so much unallocated freespace, you could easily
> allocate 4-10G for another SUSE installation, which could be used as a rescue
> system if nothing else. On future upgrades it could be a fallback when
> installation malfunctions. Or like many do, always have two. Use the oldest
> as the fallback reserve. When a new release comes out, upgrade the older,
> using the previous newer as the fallback. They're usually a lot easier than
> "rescue" boots from CD/DVD.
> -- 
> "   A patriot without religion . . . is as great a
> paradox, as an honest man without the fear of God."
>                                    John Adams
> 
>  Team OS/2 ** Reg. Linux User #211409
> 
> Felix Miata  ***  http://mrmazda.no-ip.com/
> -- 
> To unsubscribe, e-mail: [EMAIL PROTECTED]
> For additional commands, e-mail: [EMAIL PROTECTED]

-- 
Stan Goodman
Qiryat Tiv'on
Israel


"I did not attend his funeral, but I did send a note of approval" -- Mark 
Twain, on hearing of the death of a particularly corrupt politician
-- 
To unsubscribe, e-mail: [EMAIL PROTECTED]
For additional commands, e-mail: [EMAIL PROTECTED]

Reply via email to