bug-ddrescue
[Top][All Lists]
Advanced

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: [Bug-ddrescue] partition size for ddrescue direct to device


From: peter
Subject: Re: [Bug-ddrescue] partition size for ddrescue direct to device
Date: Mon, 26 Sep 2005 07:54:26 -0400
User-agent: Internet Messaging Program (IMP) 3.2.4

Hi Ariel-

thanks for writing back again. i am done with ddrescue for now i think. so it is
safe to use dd without any extra args to copy the data back onto a new drive? i
know the reason to use ddrescue over dd has to do with null blocks, so i'm not
sure of the correct syntax to use.

i tried to mount the disk first, it failed. i did find out the exact partition
size, it can be seen with "pdisk -l"
79892103168 bytes
156039264 blocks

so i used this command to copy it:
sudo dd if=/dev/disk2s12 of=/dev/disk3s3 bs=512 count=156039264

mounting still failed. i then used a utility to rebuild the catalogue file.
mounting worked! i'm not sure if i could use the same utility on the partition
with the wrong size, but i am not going to try because that is a archive of the
data for now.

unfortunately i still cannot boot off this final output partition. it may have
something to do with the firewire drivers that i erased when i used pdisk to
make the manual partition with the exact size. not sure.

i am going to go get a replacement drive from apple, that should be identical to
the internal drive that failed. i am wondering what is the best way to copy the
data from the "too big" partition to this drive. should i use the same command
again?

the author here said to use ddrescue again to copy back to the device:
http://www.macosxhints.com/article.php?story=20050720092514388&query=ddrescue

i also know mac os x which is BSD style has two devices, /dev/disk and
/dev/rdisk which allows "unbuffered access". i am using /dev/disk this whole
time, i hope that is the right way to go.

finally.. i goofed up a bit when i was still getting the hang of dd, making the
copy from the ddrescue output to the new correct size partition, and i ended up
killing it a few times and resuming using the seek and skip options to pick up
where it left off. i hope this is an OK thing to do and i didn't mess up the
copy.

i killed ddrescue a few hours ago to reboot the machine, and stupidly when it
started back up again and os x failed to boot from my final recovered firewire
drive, it tried to boot from the internal disk. i'm not sure if i should go
back to splitting errors with ddrescue or if this will damage the data because
some has changed on the damaged disk. previously the disk had been unmounted
the whole time.

thoughts much appreciated!
thanks again,
peter

Quoting Ariel <address@hidden>:

>
> On Fri, 23 Sep 2005 address@hidden wrote:
>
> > now i am seeing i failed to copy the whole device.. i missed the partition
> > table?
>
> No, you did it right. You don't need the partition table, just the actual
> partition.
>
> Can you try: diskutil info /dev/disk0s3
>
> Hopefully that will give you the real size of the disk.
>
> Also try blockdev --getsize /dev/disk0s3
>
> > should i have copied disk0 -> a new drive with no pre-existing partition
> table?
>
> No. You are fine. Just find out the old size of the partition and create a
> new partition anywhere, just as long as it's the same size.
>
> I wouldn't run ddrescue again for no reason - once a hard disk fails it
> starts getting worse the more you use it. You won't get as much data the
> second time, so keep what you go already.
>
> > if i understand what you're saying i can use dd to copy data off disk0s3
> while
> > ddrescue is currently running?
>
> Yes. To some other partition you are not using. Hopefully on a different
> physical disk, or it will be very slow.
>
>       -Ariel
>
>







reply via email to

[Prev in Thread] Current Thread [Next in Thread]