12:51 p.m.
I have been trying to use the F7 that I copied to the new Hard Drive
but instead I'm getting very strange results. With just the new Hard
Drive mounted I set up Grub and changed grub.conf so the proper root was
showing. But it would not boot and the error was file system not good.
So DD copy must have ruined the file system I put on.
So back to the Old Hard Drive and it has some weird stuff going. I
wanted to fix the /dev/sdb5 file system and here is what happend:
[root@k5di ~]# fsck /dev/sdb5
fsck 1.40.2 (12-Jul-2007)
e2fsck 1.40.2 (12-Jul-2007)
/dev/sdb5 is mounted.
WARNING!!! Running e2fsck on a mounted filesystem may cause
SEVERE filesystem damage.
Do you really want to continue (y/n)? yes
f7: recovering journal
f7: clean, 174545/10246368 files, 3132553/10239421 blocks
[root@k5di ~]# mount -t ext2 /dev/sdb5 /mnt
mount: /dev/sdb5 already mounted or /mnt busy
The last 2 lines say that /dev/sdb5 is mounted to this Old Hard Drive
somehow. I did not do this. /etc/fsack did not do this. So not sure what
is going on. And last the /etc/grub.conf has been changed by something
to the wrong on this computer. Weird!
--
Karl F. Larsen, AKA K5DI
Linux User
#450462 http://counter.li.org.
12:57 p.m.
Somebody in the thread at some point said:
/dev/sdb5 is mounted.
WARNING!!! Running e2fsck on a mounted filesystem may cause
SEVERE filesystem damage.
Do you really want to continue (y/n)? yes
That is like a gun saying "Warning, gun is pointed at your head... Do
you really want to continue firing projectile (y/n) yes"
Whatever was on /dev/sdb5 is fixed "real good" now, bust out the backups
(but don't fsck them!)
-Andy
1:10 p.m.
On Thu, 2007-08-30 at 11:51 -0600, Karl Larsen wrote:
I have been trying to use the F7 that I copied to the new Hard
Drive
but instead I'm getting very strange results. With just the new Hard
Drive mounted I set up Grub and changed grub.conf so the proper root was
showing. But it would not boot and the error was file system not good.
So DD copy must have ruined the file system I put on.
So back to the Old Hard Drive and it has some weird stuff going. I
wanted to fix the /dev/sdb5 file system and here is what happend:
[root@k5di ~]# fsck /dev/sdb5
fsck 1.40.2 (12-Jul-2007)
e2fsck 1.40.2 (12-Jul-2007)
/dev/sdb5 is mounted.
WARNING!!! Running e2fsck on a mounted filesystem may cause
SEVERE filesystem damage.
Do you really want to continue (y/n)? yes
----
at this point - you have to ask yourself...Is there something in my
constitution that is hell bent to self destruction?
----
f7: recovering journal
f7: clean, 174545/10246368 files, 3132553/10239421 blocks
[root@k5di ~]# mount -t ext2 /dev/sdb5 /mnt
mount: /dev/sdb5 already mounted or /mnt busy
The last 2 lines say that /dev/sdb5 is mounted to this Old Hard Drive
somehow. I did not do this. /etc/fsack did not do this. So not sure what
is going on. And last the /etc/grub.conf has been changed by something
to the wrong on this computer. Weird!
----
you can always check what is mounted and where it is mounted by simply
issuing the command 'mount'
you can always 'unmount' filesystems with the 'umount' command but
realize that if you unmount your filesystem with necessary files such
as /usr or /etc - the system is likely to refuse because they are in
use.
you can see what filesystems are set to mount at bootup or later by
checking /etc/fstab and/or what is mounted and when by
checking /var/log/messages
--
Craig White <craig(a)tobyhouse.com>
1:10 p.m.
<snip>
So DD copy must have ruined the file system I put on.
<snip>
Karl F. Larsen, AKA K5DI
You obviously misunderstood what dd does. dd will overwrite
everything (so yes will overwrite the file system table at the
beginning of the partition in addition to everything else). It is a
bit by bit copy. You can use dd to copy a file to a new file which
will do so on an existing file system (must be on an existing file
system). However when running dd against a physical device (i.e.
/dev/sda) or partition (i.e. /dev/sda5) with a destination of another
physical device or partition, then the destination device/partition
does not have to be formatted prior to doing this. dd is copying
everything (and I mean everything - partition table, inode table, file
slack, unallocated space - everything) thus resulting in an exact bit
by bit duplicate of that device onto a new one (or partition onto a
new partition). Hence why I said I didn't think it would work to only
grab 10 gig of a 30 gig partition using dd with if= being your
device/partition and your of= being a device/partition.
Now if your of= was a file name on a mounted file system, that would
be very different (wouldn't do what you want to do, just an aside to
this topic). You can dd a thumb drive to a file on your system (i.e.
if=/dev/sda of=/home/myname/thumb.dd) and then using the loop back
option mount thumb.dd (image of a block device), make changes to it
(add/remove/modify the content), unmount it and then dd it back onto
your thumb drive and the thumb drive would now have all those changes.
You can even use dd to create an empty thumb drive. I've done this
in order to create a scenario for a class and then dd'ed it onto a
bunch of thumb drives to that everybody would be working from the
exact same thumb drive. But I am really straying of topic now. Your
issue was moving your Fedora 7 partition to a new partition - not
something directly on topic seeing achieving this has nothing to do
with Fedora but is generic to Linux, however nevertheless dealing with
a Fedora system so sufficiently on topic to be asked on this list.
Jacques B.
1:24 p.m.
<snip>
[root@k5di ~]# mount -t ext2 /dev/sdb5 /mnt
mount: /dev/sdb5 already mounted or /mnt busy
The last 2 lines say that /dev/sdb5 is mounted to this Old Hard Drive
somehow.
<snip>
Karl F. Larsen, AKA K5DI
No, the last two lines is
telling you that either
/dev/sdb5 is already mounted
OR
the mount point /mnt is busy
Which is not to be confused with meaning something is already mounted
there. You can mount something on top of a currently mounted device
(and repeat if you wish). You will not see the underlying mounted
device mounted at that mount point until after you unmount the last
mounted device (and continue unmounting in reverse order everything
mounted at that mount point). So if the second scenario is applicable
then something else is going on causing /mnt to be busy hence refusing
to mount /dev/sdb5 at that mount point.
As was suggested, type mount with no other options to see what is
mounted (keeping it simple - there is a caveat that most likely does
not apply so I won't go there and further confuse things).
Jacques B.
5:12 p.m.
Jacques B. wrote:
<snip>
> [root@k5di ~]# mount -t ext2 /dev/sdb5 /mnt
> mount: /dev/sdb5 already mounted or /mnt busy
>
> The last 2 lines say that /dev/sdb5 is mounted to this Old Hard Drive
> somehow.
>
<snip>
> Karl F. Larsen, AKA K5DI
>
No, the last two lines is telling you that either
/dev/sdb5 is already mounted
OR
the mount point /mnt is busy
Which is not to be confused with meaning something is already mounted
there. You can mount something on top of a currently mounted device
(and repeat if you wish). You will not see the underlying mounted
device mounted at that mount point until after you unmount the last
mounted device (and continue unmounting in reverse order everything
mounted at that mount point). So if the second scenario is applicable
then something else is going on causing /mnt to be busy hence refusing
to mount /dev/sdb5 at that mount point.
As was suggested, type mount with no other options to see what is
mounted (keeping it simple - there is a caveat that most likely does
not apply so I won't go there and further confuse things).
Jacques B.
Well hang onto your hat. I am typing this on my dd copy which was
booted over my old copy boot up. Someone said that if you use LABLE's
this can and IS happening.
I will take what you said as a way to try dd again. I will take the
file system off the partition and let dd put it back on. Since this is
being typed on the current new F7 it is very close to working.
--
Karl F. Larsen, AKA K5DI
Linux User
#450462 http://counter.li.org.
6:39 p.m.
<snip>
I will take what you said as a way to try dd again. I will take
the
file system off the partition and let dd put it back on. Since this is
being typed on the current new F7 it is very close to working.
Karl F. Larsen, AKA K5DI
Won't change a thing. The end result will be identical. dd will
produce the same results as it overwrites what is there. So whether
you start with a \x00 drive, a fully formatted drive, a partially
formatted, whatever - it WON'T matter. dd overwrites (as it did when
you first tried it, and as it will when you try it again).
You can't dd a slice of a 30 gig partition (if=/dev/sda#) onto a 10
gig partition (of=/dev/sdb#) and expect it to work. IT WILL NOT.
As I stated earlier, dd copies EVERY BIT. So the partition
information at the beginning of that 30 gig partition gets written at
the same byte offset (assuming you didn't use a seek which you would
not in this case anyhow) on the 10 gig partition which will not work.
Fragments of your files could very well be beyond the 10 gig mark on
that 30 gig drive. Those parts of files will not get copied. Again,
you clearly do not understand dd. Trust me that it will not work as
you hope it will. And if there is some messed up data on the 30 gig,
well dd will copy messed up data. It copies bit for bit. It doesn't
care about file systems, file sizes, file names, or any of that. If
the bit is \x45 then it will copy \x45 onto the other drive even if
that is a corrupt piece of data.
Even if it somehow does work because as luck would have it all your
allocated data is in that first 10 gig, it still will at some point
puke when it tries to write a file beyond the 10 gig mark because it
expects that it can seeing the partition info and inode table is for a
30 gig partition.
dd is a very powerful tool. But it does not work for your scenario.
Jacques B.
6:52 p.m.
Jacques B. wrote:
<snip>
> I will take what you said as a way to try dd again. I will take the
> file system off the partition and let dd put it back on. Since this is
> being typed on the current new F7 it is very close to working.
>
> Karl F. Larsen, AKA K5DI
>
Won't change a thing. The end result will be identical. dd will
produce the same results as it overwrites what is there. So whether
you start with a \x00 drive, a fully formatted drive, a partially
formatted, whatever - it WON'T matter. dd overwrites (as it did when
you first tried it, and as it will when you try it again).
You can't dd a slice of a 30 gig partition (if=/dev/sda#) onto a 10
gig partition (of=/dev/sdb#) and expect it to work. IT WILL NOT.
Come on for God's sake! I am way past this stage already. I am
taking this F7 on a 20 GB partition and putting it on a 30 GB partition.
As I stated earlier, dd copies EVERY BIT. So the partition
information at the beginning of that 30 gig partition gets written at
the same byte offset (assuming you didn't use a seek which you would
not in this case anyhow) on the 10 gig partition which will not work.
YES YES
Fragments of your files could very well be beyond the 10 gig mark on
that 30 gig drive. Those parts of files will not get copied. Again,
you clearly do not understand dd. Trust me that it will not work as
you hope it will. And if there is some messed up data on the 30 gig,
well dd will copy messed up data. It copies bit for bit. It doesn't
care about file systems, file sizes, file names, or any of that. If
the bit is \x45 then it will copy \x45 onto the other drive even if
that is a corrupt piece of data.
Again for God's sake! The F7 I am using now is fully updated and
working fine. I want to put the whole thing on a new HARD DRIVE!!!!!
Even if it somehow does work because as luck would have it all your
allocated data is in that first 10 gig, it still will at some point
puke when it tries to write a file beyond the 10 gig mark because it
expects that it can seeing the partition info and inode table is for a
30 gig partition.
dd is a very powerful tool. But it does not work for your scenario.
Jacques B.
Now with your obvious wisdom tell me if the ACTUAL thing I am trying
to do will work. EVER.
--
Karl F. Larsen, AKA K5DI
Linux User
#450462 http://counter.li.org.
9:35 p.m.
Come on for God's sake! I am way past this stage already. I
am
taking this F7 on a 20 GB partition and putting it on a 30 GB partition.
That was not obvious in the thread(s) that you started on this issue.
Am I the only one that missed the change of trying to copy the first
10 gig of a 30 gig onto a new 10 gig partition?
Again for God's sake! The F7 I am using now is fully updated
and
working fine. I want to put the whole thing on a new HARD DRIVE!!!!!
Again, that was not the thread you started on this. You wanted to
copy 10 of a 30 gig partition from /dev/sda6 to a 10 gig partition on
/dev/sdb5.
From your thread you started about 13 hours ago titled dd and cp -a:
"> Karl Larsen wrote:
> I have this computer on /dev/sda and the new hard drive is
> /dev/sdb. This F7 is all in /dev/sda6 and I want to copy /dev/sda6 to
> /dev/sdb5. I tried dd but it failed I think because /dev/sdb5 is
> smaller 10 GB than /dev/sda6 which is 30 GB. It ended with an error
> message."
Now with your obvious wisdom tell me if the ACTUAL thing I am
trying
to do will work. EVER.
Karl F. Larsen, AKA K5DI
What are you ACTUALLY trying to do? Are you now trying to copy an
entire physical drive to a new physical drive? What is the size of
the source drive? The size of the destination drive?
Start a thread and stick to it. And if you decide to abandon a
particular approach and undertake a different one, please be so kind
as to post this fact so people can re-align their advice accordingly.
You have the knack to bit the hand that feeds you... I'm a patient
and tolerant individual and strive to maintain respect in my postings.
I ask no less from others.
Jacques B.
10:10 p.m.
Jacques B. wrote:
> Come on for God's sake! I am way past this stage already.
I am
> taking this F7 on a 20 GB partition and putting it on a 30 GB partition.
>
That was not obvious in the thread(s) that you started on this issue.
Am I the only one that missed the change of trying to copy the first
10 gig of a 30 gig onto a new 10 gig partition?
> Again for God's sake! The F7 I am using now is fully updated and
> working fine. I want to put the whole thing on a new HARD DRIVE!!!!!
>
Again, that was not the thread you started on this. You wanted to
copy 10 of a 30 gig partition from /dev/sda6 to a 10 gig partition on
/dev/sdb5.
>From your thread you started about 13 hours ago titled dd and cp -a:
"> Karl Larsen wrote:
>> I have this computer on /dev/sda and the new hard drive is
>> /dev/sdb. This F7 is all in /dev/sda6 and I want to copy /dev/sda6 to
>> /dev/sdb5. I tried dd but it failed I think because /dev/sdb5 is
>> smaller 10 GB than /dev/sda6 which is 30 GB. It ended with an error
>> message."
>>
> Now with your obvious wisdom tell me if the ACTUAL thing I am trying
> to do will work. EVER.
>
> Karl F. Larsen, AKA K5DI
>
What are you ACTUALLY trying to do? Are you now trying to copy an
entire physical drive to a new physical drive? What is the size of
the source drive? The size of the destination drive?
Start a thread and stick to it. And if you decide to abandon a
particular approach and undertake a different one, please be so kind
as to post this fact so people can re-align their advice accordingly.
You have the knack to bit the hand that feeds you... I'm a patient
and tolerant individual and strive to maintain respect in my postings.
I ask no less from others.
Jacques B.
Here is what I did this time. /dev/sda6 which is this computer has
5098 cylinders of space (the hard drives are identical) and the other
has 5100 cylinders. So it is just a bit larger than the source
partition. This will finish in good time and then I will use fsck to see
if the transfered file system is good. Last time I mounted the new
partition to this computer and it looked so good I was sure it was
perfect. Alas the LABEL problems and many others kept me busy doing
stupid things.
This time I know much more 8-)
--
Karl F. Larsen, AKA K5DI
Linux User
#450462 http://counter.li.org.
10:18 p.m.
Here is what I did this time. /dev/sda6 which is this computer
has
5098 cylinders of space (the hard drives are identical) and the other
has 5100 cylinders. So it is just a bit larger than the source
partition. This will finish in good time and then I will use fsck to see
if the transfered file system is good. Last time I mounted the new
partition to this computer and it looked so good I was sure it was
perfect. Alas the LABEL problems and many others kept me busy doing
stupid things.
This time I know much more 8-)
Karl F. Larsen, AKA K5DI
Ok, that should work fine. I don't use labels but it makes sense what
was previously explained. Where dd does an exact copy, that means the
label on both drives is the same. So if you have both drives
connected at the same time it would cause problems seeing they have
the exact same label. Hence why tune2fs was suggested to change the
label of one of the drives (with only one connected of course - once
the label is changed then you should be able to fire up the computer
with both drives connected).
Jacques B.
5:39 a.m.
Karl Larsen wrote:
Alas the LABEL problems and many others kept me busy doing
stupid things.
You really like to discourage people from attempting to help you with
the above comment.
Anyway, last posting giving you stupid things to try. You got it
altogether and need no replies to your queries.
This time I know much more 8-)
I hope so.
Jim
--
If something has not yet gone wrong then it would ultimately have been
beneficial for it to go wrong.
6:21 a.m.
Jim Cornette wrote:
Karl Larsen wrote:
Alas the LABEL problems and many others kept me busy doing
> stupid things.
You really like to discourage people from attempting to help you with
the above comment.
No the stupid things were of my own doing. Not setting the
bios
properly, operating from the new hard drive when I thought I was on the
old hard drive. I didn't know two sets of the same LABEL will load up odd.
Anyway, last posting giving you stupid things to try. You got it
altogether and need no replies to your queries.
That is not the case. I learned dd is a way to do this from this
list and I am trying to make it work. Lots of help has I hope made this
happen.
>
> This time I know much more 8-)
And it is from my experience and that of the people on this list.
I hope so.
Jim
--
Karl F. Larsen, AKA K5DI
Linux User
#450462 http://counter.li.org.
7:59 p.m.
Karl Larsen wrote:
Jim Cornette wrote:
> Karl Larsen wrote:
> Alas the LABEL problems and many others kept me busy doing
>> stupid things.
>
> You really like to discourage people from attempting to help you with
> the above comment.
No the stupid things were of my own doing. Not setting the bios
properly, operating from the new hard drive when I thought I was on the
old hard drive. I didn't know two sets of the same LABEL will load up odd.
Sorry, my English skills are poor. I read the comments as other people
kept advising you to do stupid things. Many other problems would have
been more clear to me.
> Anyway, last posting giving you stupid things to try. You got it
> altogether and need no replies to your queries.
>
That is not the case. I learned dd is a way to do this from this list
and I am trying to make it work. Lots of help has I hope made this happen.
>>
>> This time I know much more 8-)
>
And it is from my experience and that of the people on this list.
I learn quite a bit from the problems and solutions to the problems
others encounter. I very rarely use some of the programs others need at
the time it is discussed but at least get familiar with the program from
list discussions.
The discussion on the dd program enlightened me on many issues so the
discussion was not wasted.
Jim
--
Everything is controlled by a small evil group to which, unfortunately,
no one we know belongs.
7:06 p.m.
Karl Larsen wrote:
mount: /dev/sdb5 already mounted or /mnt busy
The last 2 lines say that /dev/sdb5 is mounted to this Old Hard Drive
somehow. I did not do this. /etc/fsack did not do this. So not sure what
Yes you did. See your opening mail of the thread titled 'dd and cp -a'
in which you say that you mounted it to /mnt, but didn't mention how. I
am assuming you made it permanent by adding it to fstab. And if you did
this, dd'ing the partition would take this setting over to the new drive.
--
Regards,
विवेक ज. पाटणकर (Vivek J. Patankar)
Registered Linux User #374218
Fedora release 7 (Moonshine)
Linux 2.6.22.4-65.fc7 x86_64
7:17 p.m.
Vivek J. Patankar wrote:
Karl Larsen wrote:
> mount: /dev/sdb5 already mounted or /mnt busy
>
> The last 2 lines say that /dev/sdb5 is mounted to this Old Hard Drive
> somehow. I did not do this. /etc/fsack did not do this. So not sure what
Yes you did. See your opening mail of the thread titled 'dd and cp -a'
in which you say that you mounted it to /mnt, but didn't mention how.
I am assuming you made it permanent by adding it to fstab. And if you
did this, dd'ing the partition would take this setting over to the new
drive.
Guys I am getting a information overload. My question is simple now.
If I do it right is it possible to use dd to make a copy of this F7 to
another Hard Drive?
Also you NEVER mount a partition that your dd transfering too. It
finds it fine.
--
Karl F. Larsen, AKA K5DI
Linux User
#450462 http://counter.li.org.
7:59 p.m.
Karl Larsen wrote:
Vivek J. Patankar wrote:
> Karl Larsen wrote:
>> mount: /dev/sdb5 already mounted or /mnt busy
>>
>> The last 2 lines say that /dev/sdb5 is mounted to this Old Hard Drive
>> somehow. I did not do this. /etc/fsack did not do this. So not sure what
>
> Yes you did. See your opening mail of the thread titled 'dd and cp -a'
> in which you say that you mounted it to /mnt, but didn't mention how.
> I am assuming you made it permanent by adding it to fstab. And if you
> did this, dd'ing the partition would take this setting over to the new
> drive.
>
Guys I am getting a information overload. My question is simple now.
If I do it right is it possible to use dd to make a copy of this F7 to
another Hard Drive?
Also you NEVER mount a partition that your dd transfering too. It
finds it fine.
Wouldn't you lose 10 GB and the disk would appear to be a 20 GB disk
also with dd?
What about making a /boot, a /home, a / and a swap partition on the new
drive, create a filesystem and label the new partitions, then mount the
new / under a /mnt/newdrive directory, create a boot and home partition
on the new / partition so you could mount the to be new home drive under
a /mnt/newdrive/home mount point and then do the same with the soon to
be new boot directory under /mnt/newdrive/boot
Once the new disk partitions are mounted you could use rsync to
replicate your present installation onto the new drive.
I did this once and was fairly successful and got the replicated disk to
start to boot by using the rescue mode to run grub-install on the new
disk. It was a sata drive and the other drive was an IDE so I had other
issues with the attempt. If the target disk was IDE instead of a SATA
drive, it might have worked.
Someone more familiar with rsync and the effects of /proc, /sys and
other not truly directories might be able to answer to whether rsync
would work when transferred to IDE to IDE. I figure the kernel panic
that I got was because of the IDE to SATA attempt.
Using some replication program for Linux would probably be the best
option rather than DD or rsync.
Jim
--
A right is not what someone gives you; it's what no one can take from you.
-- Ramsey Clark
8:24 p.m.
Jim Cornette wrote:
Karl Larsen wrote:
> Vivek J. Patankar wrote:
>> Karl Larsen wrote:
>>> mount: /dev/sdb5 already mounted or /mnt busy
>>>
>>> The last 2 lines say that /dev/sdb5 is mounted to this Old Hard
>>> Drive somehow. I did not do this. /etc/fsack did not do this. So
>>> not sure what
>>
>> Yes you did. See your opening mail of the thread titled 'dd and cp
>> -a' in which you say that you mounted it to /mnt, but didn't mention
>> how. I am assuming you made it permanent by adding it to fstab. And
>> if you did this, dd'ing the partition would take this setting over
>> to the new drive.
>>
> Guys I am getting a information overload. My question is simple
> now. If I do it right is it possible to use dd to make a copy of this
> F7 to another Hard Drive?
>
> Also you NEVER mount a partition that your dd transfering too. It
> finds it fine.
>
Wouldn't you lose 10 GB and the disk would appear to be a 20 GB disk
also with dd?
What about making a /boot, a /home, a / and a swap partition on the
new drive, create a filesystem and label the new partitions, then
mount the new / under a /mnt/newdrive directory, create a boot and
home partition on the new / partition so you could mount the to be new
home drive under a /mnt/newdrive/home mount point and then do the same
with the soon to be new boot directory under /mnt/newdrive/boot
Once the new disk partitions are mounted you could use rsync to
replicate your present installation onto the new drive.
I did this once and was fairly successful and got the replicated disk
to start to boot by using the rescue mode to run grub-install on the
new disk. It was a sata drive and the other drive was an IDE so I had
other issues with the attempt. If the target disk was IDE instead of a
SATA drive, it might have worked.
Someone more familiar with rsync and the effects of /proc, /sys and
other not truly directories might be able to answer to whether rsync
would work when transferred to IDE to IDE. I figure the kernel panic
that I got was because of the IDE to SATA attempt.
Using some replication program for Linux would probably be the best
option rather than DD or rsync.
Jim
Hi Jim, well the third iteration of a dd to the new drive is
ongoing. So When it is done I will not reboot until I have used fsck on
the new file system. Last time I had big trouble even getting a chance
to check the file system :-)
--
Karl F. Larsen, AKA K5DI
Linux User
#450462 http://counter.li.org.
9:25 p.m.
Karl Larsen wrote:
> Using some replication program for Linux would probably be the
best
> option rather than DD or rsync.
>
> Jim
>
Hi Jim, well the third iteration of a dd to the new drive is ongoing.
So When it is done I will not reboot until I have used fsck on the new
file system. Last time I had big trouble even getting a chance to check
the file system :-)
If DD does not work out for you, you might try partimage or ghost 4
Linux. I am not sure if rsync would be successful. I did read a mention
by Alan regarding rsync -a being an option after already have posted a
response to this thread.
Good luck finding it though since there are many different threads with
very similar subject matter.
Jim
--
I switched on the radio and nearly dropped dead
The news was so bad that i fell out of bed
There was a gas strike, oil strike, lorry strike, bread strike
Got to be a superman to survive
Gas bills, rent bills, tax bills, phone bills
I'm such a wreck but i'm staying alive
Superman By "The Kinks"
6:49 a.m.
New subject: DD not working--SUCCESS!
Vivek J. Patankar wrote:
Karl Larsen wrote:
> mount: /dev/sdb5 already mounted or /mnt busy
>
> The last 2 lines say that /dev/sdb5 is mounted to this Old Hard Drive
> somehow. I did not do this. /etc/fsack did not do this. So not sure what
Yes you did. See your opening mail of the thread titled 'dd and cp -a'
in which you say that you mounted it to /mnt, but didn't mention how.
I am assuming you made it permanent by adding it to fstab. And if you
did this, dd'ing the partition would take this setting over to the new
drive.
This is being typed on the copy of F7 I made last night. I had a lot
of things to set up in fstab and grub.conf, and I ran fsck to check the
file system and it said it was fine.
Had to use the Rescue CD because I forgot to set one thing on
grub.conf :-(
But when I got that fixed it booted up EXACTLY like the original version
with a couple of odd things.
There are many things to consider before you make a copy of your
working system with dd. If you do those things it is just scary how
perfect the copy is :-P
I will get this hard drive out of the computer now and the use of dd
was a success.
--
Karl F. Larsen, AKA K5DI
Linux User
#450462 http://counter.li.org.
7:49 a.m.
New subject: DD not working--SUCCESS!
I will get this hard drive out of the computer now and the use of dd
was a success.
Karl F. Larsen, AKA K5DI
Excellent! Thanks for the update. Out of
curiosity did you boot from
the rescue disk when you did your dd, or did you boot from the hard
drive you were cloning? If you booted from the hard drive and then
cloned it while booted from it that would explain some of the quirky
issues you may have faced afterwards. Because the clone would be of a
running system. So booting from it would be comparable to booting
from a system that crashed (I'm making an educated guess at that one).
Jacques B.
9:04 a.m.
New subject: DD not working--SUCCESS!
Jacques B. wrote:
> I will get this hard drive out of the computer now and the
use of dd
> was a success.
>
>
> Karl F. Larsen, AKA K5DI
>
Excellent! Thanks for the update. Out of curiosity did you boot from
the rescue disk when you did your dd, or did you boot from the hard
drive you were cloning?
Hi Jacques, I used the drive I was cloning. That way I
could keep
using the Internet and write things.
If you booted from the hard drive and then
cloned it while booted from it that would explain some of the quirky
issues you may have faced afterwards.
I do not think that is true. Most of the
quirky issues were the
result of my stupidity. And they were expected results if you understood.
Because the clone would be of a
running system. So booting from it would be comparable to booting
from a system that crashed (I'm making an educated guess at that one).
Not a good guess. To use DD you need a computer with dd and a fast
cpu. I did top while dd was working and it was taking 70% of the cpu's
time :-P
Jacques B.
--
Karl F. Larsen, AKA K5DI
Linux User
#450462 http://counter.li.org.
9:50 a.m.
New subject: DD not working--SUCCESS!
Hi Jacques, I used the drive I was cloning. That way I could
keep
using the Internet and write things.
A live CD such as Knoppix or Helix or many
other such distros would
allow you to do the same (especially easy if you have your router
authenticating for your cable/DSL connection with DHCP activated on
it).
> If you booted from the hard drive and then
> cloned it while booted from it that would explain some of the quirky
> issues you may have faced afterwards.
I do not think that is true. Most of the quirky issues were the
result of my stupidity. And they were expected results if you understood.
Doing a dd of a live, running system is a potential problem. Your
system is in constant state of change. Granted if you are doing
nothing else with the system at the time and booted in run level 3 you
minimize this impact. But in an X Windows environment with applets
running in the background chances are something is always happening in
the back ground, some of which could cause writes to the hard drive
(especially where you mentioned you wanted Internet access to keep
writing things and such - lots of disk activity resulting from that).
Imagine when you start the copy and dd reads the partition
information, the superblock, and inode table near the beginning of the
drive and writes it to the new drive. Part way through the process
writes take place on the source drive thereby potentially altering
superblock information, block groups, and inodes.
I haven't looked at file system stuff in about 6 months and I'm not
the primary resource for that topic so I'm not fluent enough to fully
explain the issues. But I do know enough about it to know imaging a
live system could cause (and likely will cause) issues.
It is far safer to boot from a live CD without either drives mounted
(only connected) and then dd from one to the other.
> Because the clone would be of a
> running system. So booting from it would be comparable to booting
> from a system that crashed (I'm making an educated guess at that one).
>
Not a good guess. To use DD you need a computer with dd and a fast
cpu. I did top while dd was working and it was taking 70% of the cpu's
time :-P
dd is a standard binary on any Linux distros I've ever known. So
you'd have access to dd with any of the mainstream live CDs out there
(perhaps some very specific ones may exclude it but that would be the
exception rather than the rule). In the event you hit one that didn't
have it (very unlikely) you could either download one that does
(Knoppix, Helix, Slax, Phlax, and if I dig through my material I could
find many other live CDs that have dd) or download and install dd in
the live environment (again - very unlikely it would come to that).
dd will use lots of CPU time. There are a number of bottle necks in
the process - hard drive speed, motherboard/hard drive controller,
cables, available RAM, and no doubt a number of others. The CPU is
not likely your bottle neck. Chances are it's the reading & writing
to the hard drive. So dd would not need 100% of CPU if it fills up
the RAM quicker than it can read/write. How much CPU time dd uses
will vary from scenario to scenario (all factors noted previously and
many others being considered). Running it off a live CD would impact
its performance seeing you are using some of the RAM for the live CD.
But the stability it affords you vs doing it on a live system is worth
it.
When you were mentioning rescue CD I was under the impression you were
doing it from that environment. Or you were booting from a different
machine where you had your source and target connected (and not
mounted). Where I've always done such tasks from a live CD
environment or with another box depending on my situation I
incorrectly assumed that others would automatically do likewise.
My advice is to boot from a live CD with both drives connected (not
mounted - don't need to be mounted - actually greater assurance of
integrity if not mounted). dd the source drive onto the target drive.
Then remove the source drive and boot with the target drive only
(remember that initially they will be master/slave unless you put them
on two different IDE chains so you will have to then ensure that your
new drive is master or cable select and at the end of the IDE ribbon
cable - I'm assuming you have PATA and not SATA drives). It will boot
just like the original because it will be an exact copy of the
original. If you want both drives connected at the same time then you
will have to change the label on one of them. You can do that from
the live CD environment after dd is finished, or you can do it after
rebooting with the single drive and changing the label on that drive,
shutting down, and then connecting the other and rebooting with both
drives now having unique labels.
Jacques B.
11:16 a.m.
New subject: DD not working--SUCCESS!
Jacques B. wrote:
Doing a dd of a live, running system is a potential problem. Your
system is in constant state of change. Granted if you are doing
nothing else with the system at the time and booted in run level 3 you
minimize this impact. But in an X Windows environment with applets
running in the background chances are something is always happening in
the back ground, some of which could cause writes to the hard drive
(especially where you mentioned you wanted Internet access to keep
writing things and such - lots of disk activity resulting from that).
Imagine when you start the copy and dd reads the partition
information, the superblock, and inode table near the beginning of the
drive and writes it to the new drive. Part way through the process
writes take place on the source drive thereby potentially altering
superblock information, block groups, and inodes.
You are much better off doing this at run level 1 then run level 3.
You can mount the file systems read-only and still function in run
level 1. (I still prefer to use a live CD - the System Rescue CD is
nice for this.)
I haven't looked at file system stuff in about 6 months and
I'm not
the primary resource for that topic so I'm not fluent enough to fully
explain the issues. But I do know enough about it to know imaging a
live system could cause (and likely will cause) issues.
It is far safer to boot from a live CD without either drives mounted
(only connected) and then dd from one to the other.
dd will use lots of CPU time. There are a number of bottle necks in
the process - hard drive speed, motherboard/hard drive controller,
cables, available RAM, and no doubt a number of others. The CPU is
not likely your bottle neck. Chances are it's the reading & writing
to the hard drive. So dd would not need 100% of CPU if it fills up
the RAM quicker than it can read/write. How much CPU time dd uses
will vary from scenario to scenario (all factors noted previously and
many others being considered). Running it off a live CD would impact
its performance seeing you are using some of the RAM for the live CD.
But the stability it affords you vs doing it on a live system is worth
it.
dd should not use lots of CPU time. The bottleneck is disk access
speed. Depending on the drivers used, and some tweaking, the system
may be spending a lot of time in the disk driver routines doing
basically nothing.
I would think the speed would be the same regardless of the amount
of RAM, so booting off a live CD should not make dd any slower,
unless you are really low. (1 meg of free RAM should be plenty.)
I could be wrong, but I thought that dd read one block at a time,
and wrote one block at a time, unless you use the options to tweak
that. Even if the system buffers disk writes to the target drive, it
is still going to be the write speed of the target drive that will
be the problem. You may get some increase in speed if the source and
target drives are on different IDE interfaces. (source on hda,
target on hdc)
In any case, dd is not the best tool for the job. Something like
parted or g4l that understands file systems would be a better
choice. Unless your file system is full, (90% or more) it is going
to be much faster, as they only copy used blocks. (Doing recovery,
or making an image of a cracked system/damaged disk is another story...)
Mikkel
--
Do not meddle in the affairs of dragons,
for thou art crunchy and taste good with Ketchup!
12:50 p.m.
New subject: DD not working--SUCCESS!
Mikkel L. Ellertson wrote:
dd should not use lots of CPU time. The bottleneck is disk access
speed. Depending on the drivers used, and some tweaking, the system
may be spending a lot of time in the disk driver routines doing
basically nothing.
CPU use will depend on your disk controller. IDE controllers have a lot
of CPU overhead for seeks/positioning even if DMA is used for the data
transfer - that's why people still use SCSI for busy servers.
--
Les Mikesell
lesmikesell(a)gmail.com
4:14 p.m.
New subject: DD not working--SUCCESS!
On Fri, 31 Aug 2007 12:50:42 -0500
Les Mikesell <lesmikesell(a)gmail.com> wrote:
Mikkel L. Ellertson wrote:
> dd should not use lots of CPU time. The bottleneck is disk access
> speed. Depending on the drivers used, and some tweaking, the system
> may be spending a lot of time in the disk driver routines doing
> basically nothing.
CPU use will depend on your disk controller. IDE controllers have a lot
of CPU overhead for seeks/positioning even if DMA is used for the data
transfer - that's why people still use SCSI for busy servers.
Thankfully no longer true as of SATA providing your controller does AHCI
or similar and the drives support NCQ
Alan
12:55 p.m.
New subject: DD not working--SUCCESS!
Mikkel L. Ellertson wrote:
Jacques B. wrote:
> Doing a dd of a live, running system is a potential problem. Your
> system is in constant state of change. Granted if you are doing
> nothing else with the system at the time and booted in run level 3 you
> minimize this impact. But in an X Windows environment with applets
> running in the background chances are something is always happening in
> the back ground, some of which could cause writes to the hard drive
> (especially where you mentioned you wanted Internet access to keep
> writing things and such - lots of disk activity resulting from that).
> Imagine when you start the copy and dd reads the partition
> information, the superblock, and inode table near the beginning of the
> drive and writes it to the new drive. Part way through the process
> writes take place on the source drive thereby potentially altering
> superblock information, block groups, and inodes.
>
>
You are much better off doing this at run level 1 then run level 3.
You can mount the file systems read-only and still function in run
level 1. (I still prefer to use a live CD - the System Rescue CD is
nice for this.)
> I haven't looked at file system stuff in about 6 months and I'm not
> the primary resource for that topic so I'm not fluent enough to fully
> explain the issues. But I do know enough about it to know imaging a
> live system could cause (and likely will cause) issues.
>
> It is far safer to boot from a live CD without either drives mounted
> (only connected) and then dd from one to the other.
>
>
> dd will use lots of CPU time. There are a number of bottle necks in
> the process - hard drive speed, motherboard/hard drive controller,
> cables, available RAM, and no doubt a number of others. The CPU is
> not likely your bottle neck. Chances are it's the reading & writing
> to the hard drive. So dd would not need 100% of CPU if it fills up
> the RAM quicker than it can read/write. How much CPU time dd uses
> will vary from scenario to scenario (all factors noted previously and
> many others being considered). Running it off a live CD would impact
> its performance seeing you are using some of the RAM for the live CD.
> But the stability it affords you vs doing it on a live system is worth
> it.
>
>
dd should not use lots of CPU time. The bottleneck is disk access
speed. Depending on the drivers used, and some tweaking, the system
may be spending a lot of time in the disk driver routines doing
basically nothing.
I would think the speed would be the same regardless of the amount
of RAM, so booting off a live CD should not make dd any slower,
unless you are really low. (1 meg of free RAM should be plenty.)
I could be wrong, but I thought that dd read one block at a time,
and wrote one block at a time, unless you use the options to tweak
that. Even if the system buffers disk writes to the target drive, it
is still going to be the write speed of the target drive that will
be the problem. You may get some increase in speed if the source and
target drives are on different IDE interfaces. (source on hda,
target on hdc)
In any case, dd is not the best tool for the job. Something like
parted or g4l that understands file systems would be a better
choice. Unless your file system is full, (90% or more) it is going
to be much faster, as they only copy used blocks. (Doing recovery,
or making an image of a cracked system/damaged disk is another story...)
Mikkel
All I can talk to is my measurements. And when I was making my copy
with dd, top told me my CPU was 70% used by dd :-)
That is a lot of cpu use and I think those suggesting the use of the
Rescue disk I have this:
The Rescue disk does not have dd. It also lacks RAM and will NOT work. I
was getting email and doing things with my computer while dd was sending
it to the other Hard Drive.
It worked just fine.
--
Karl F. Larsen, AKA K5DI
Linux User
#450462 http://counter.li.org.
1:25 p.m.
New subject: DD not working--SUCCESS!
That is a lot of cpu use and I think those suggesting the use of the
Rescue disk I have this:
The Rescue disk does not have dd. It also lacks RAM and will NOT work. I
was getting email and doing things with my computer while dd was sending
it to the other Hard Drive.
It worked just fine.
Karl F. Larsen, AKA K5DI
What lacks RAM? I'm assuming you mean your computer. Anyhow, not
important at this point. On to more important issues.
More accurately you should say "So far it is working just fine".
Imagine photocopying a book. You photocopy the front of the book, the
foreword, the table of content with the list of chapters and headings,
then start photocopying the chapters. While you are doing this you
continue to make changes to the book. You are now at page 20 of 100
photocopying away. All at the same time you change the chapter name
of the chapter on page 45 (now that doesn't match what you photocopied
in the beginning). You remove a paragraph on page 15 (but you already
photocopied it so it's already in your copy). You delete page 19 so
now your book is from page 1-99 but you already copied that page so
it's on the copy. You update your alphabetical index at the end of
the book when you are up to page 50. So the index tells you that
something is on page 20 but it's really page 21 in your copy because
you deleted page 19 after photocopying that page so your updated copy
has it on page 20 but it is on page 21 on the original revision. And
it goes on.
Now dd is doing the exact same thing. It's copying stuff over
complete with pointers to where to find certain data and pointers to
free space on the drive. That is written at the beginning of the
copying process because it resides there. Yet you are still using the
system so what was marked as unallocated space is in use by the time
dd gets to it. And what was marked as used disk space when dd copied
over the index is now unallocated space because you deleted something.
It can make a real mess of things. Problems may manifest themselves
pretty quickly, later on, or apparently not at all if you are lucky
(but that will not mean that stuff isn't misaligned with the index
because it's impossible for that NOT to happen given your scenario).
The choice is yours. Run with it and keep your fingers crossed or
exercise an once of prevention now to avoid a potential pound of
headaches later on and redo the process using a live CD as previously
explained. Or pursue someone else's advice on how to ghost a drive
onto another one (entirely different approach than dd and certainly
more efficient if this is something you will be repeating somewhat
regularly - but again you would not do that on a live system...).
Jacques B.
2:05 p.m.
New subject: DD not working--SUCCESS!
Jacques B. wrote:
> That is a lot of cpu use and I think those suggesting the use of
the
> Rescue disk I have this:
> The Rescue disk does not have dd. It also lacks RAM and will NOT work. I
> was getting email and doing things with my computer while dd was sending
> it to the other Hard Drive.
>
> It worked just fine.
>
>
>
> Karl F. Larsen, AKA K5DI
>
What lacks RAM? I'm assuming you mean your computer. Anyhow, not
important at this point. On to more important issues.
More accurately you should say "So far it is working just fine".
Imagine photocopying a book. You photocopy the front of the book, the
foreword, the table of content with the list of chapters and headings,
then start photocopying the chapters. While you are doing this you
continue to make changes to the book. You are now at page 20 of 100
photocopying away. All at the same time you change the chapter name
of the chapter on page 45 (now that doesn't match what you photocopied
in the beginning). You remove a paragraph on page 15 (but you already
photocopied it so it's already in your copy). You delete page 19 so
now your book is from page 1-99 but you already copied that page so
it's on the copy. You update your alphabetical index at the end of
the book when you are up to page 50. So the index tells you that
something is on page 20 but it's really page 21 in your copy because
you deleted page 19 after photocopying that page so your updated copy
has it on page 20 but it is on page 21 on the original revision. And
it goes on.
Now dd is doing the exact same thing. It's copying stuff over
complete with pointers to where to find certain data and pointers to
free space on the drive. That is written at the beginning of the
copying process because it resides there. Yet you are still using the
system so what was marked as unallocated space is in use by the time
dd gets to it. And what was marked as used disk space when dd copied
over the index is now unallocated space because you deleted something.
It can make a real mess of things. Problems may manifest themselves
pretty quickly, later on, or apparently not at all if you are lucky
(but that will not mean that stuff isn't misaligned with the index
because it's impossible for that NOT to happen given your scenario).
The choice is yours. Run with it and keep your fingers crossed or
exercise an once of prevention now to avoid a potential pound of
headaches later on and redo the process using a live CD as previously
explained. Or pursue someone else's advice on how to ghost a drive
onto another one (entirely different approach than dd and certainly
more efficient if this is something you will be repeating somewhat
regularly - but again you would not do that on a live system...).
Jacques B.
OK. I changed my little paper so 6. says this:
6. Always run dd in the source computer. If your worried about computer
activity during the transfer get out of X windows by using Cont-Alt-F1
and then run dd.
--
Karl F. Larsen, AKA K5DI
Linux User
#450462 http://counter.li.org.
2:46 p.m.
New subject: DD not working--SUCCESS!
Karl Larsen wrote:
OK. I changed my little paper so 6. says this:
6. Always run dd in the source computer. If your worried about computer
activity during the transfer get out of X windows by using Cont-Alt-F1
and then run dd.
Ctrl-Alt-F1 won't exit XWindows, only hide it. You want
Ctrl-Alt-Backspace. And I'm almost certain dd is included in the live cd.
Matt Flaschen
2:02 p.m.
New subject: DD not working--SUCCESS!
On Fri, 2007-08-31 at 11:55 -0600, Karl Larsen wrote:
The Rescue disk does not have dd.
This is false.
Because I found your statement so incredulous, I had to prove it.
Accordingly, I just booted from the F7 rescue CD, and when I got to the
shell prompt, I typed 'dd' and it worked. It would be hard to imagine an
effective rescue CD that lacked such basic utilities.
Not that I think it will make much difference, but I have to echo the
sentiments of others and say that copying a live, mounted partition with
a non-atomic utility like dd is foolhardy at best.
The fact that you did and it seemed to work does not mean that doing so
is safe. We've all done stupid things in our lives that should have
killed us but we somehow made out ok. That doesn't mean I'm going to
recommend to my children to do the same stupid things.
3:32 p.m.
New subject: DD not working--SUCCESS!
Alan M. Evans wrote:
On Fri, 2007-08-31 at 11:55 -0600, Karl Larsen wrote:
> The Rescue disk does not have dd.
>
This is false.
Because I found your statement so incredulous, I had to prove it.
Accordingly, I just booted from the F7 rescue CD, and when I got to the
shell prompt, I typed 'dd' and it worked. It would be hard to imagine an
effective rescue CD that lacked such basic utilities.
Not that I think it will make much difference, but I have to echo the
sentiments of others and say that copying a live, mounted partition with
a non-atomic utility like dd is foolhardy at best.
The fact that you did and it seemed to work does not mean that doing so
is safe. We've all done stupid things in our lives that should have
killed us but we somehow made out ok. That doesn't mean I'm going to
recommend to my children to do the same stupid things.
I suggest you do this: Come up in the Rescue cd and do NOT mount
your computer to the rescue at /mnt/syswhatever. Then see if you have dd.
As for your tone and lack of understanding I think you take the cake
for gall. If you don't believe what I just DID, then stop writing about it.
--
Karl F. Larsen, AKA K5DI
Linux User
#450462 http://counter.li.org.
4:15 p.m.
New subject: DD not working--SUCCESS!
On Fri, 2007-08-31 at 14:32 -0600, Karl Larsen wrote:
I suggest you do this: Come up in the Rescue cd and do NOT mount
your computer to the rescue at /mnt/syswhatever. Then see if you have dd.
I used the rescue disk on a computer that only has Windows 2000
installed, which partition was not mounted. So, yes, dd was on the
rescue disk.
As for your tone and lack of understanding I think you take the
cake
for gall. If you don't believe what I just DID, then stop writing about it.
I didn't say that it didn't work for you. Only that the fact that it
worked (or seemed to work) for you doesn't automatically make it a good
idea.
As for tone and understanding, might I suggest that you stop preaching
to the list about a utility that most of us have used for years and
you've used for days. I understand perfectly what's going on; it's you
who's bumbling through with no real understanding of the tools or
systems you are using, all the while calling ignorant those actually in
the know.
11:22 a.m.
New subject: DD not working--SUCCESS!
Karl Larsen wrote:
Jacques B. wrote:
> Because the clone would be of a
> running system. So booting from it would be comparable to booting
> from a system that crashed (I'm making an educated guess at that one).
>
Not a good guess. To use DD you need a computer with dd and a fast
cpu. I did top while dd was working and it was taking 70% of the cpu's
time :-P
It's one of the oldest Unix programs: it will copy as fast as your
system can go. Yes, a fast system will copy faster.
(BTW, it is a good guess. You are copying straight off the device,
bypassing the filesystem. The primary problem I think is the
possibility of missing cached writes. Things like log files will
also be in an odd state, but that is unlikely to ever be critical.
However backups, and what you are doing is essentially the same
thing, are best done on unmounted systems.)
--
imalone
11:19 a.m.
New subject: DD not working--SUCCESS!
On Fri, 31 Aug 2007, Ian Malone wrote:
Karl Larsen wrote:
> Jacques B. wrote:
> > Because the clone would be of a
> > running system. So booting from it would be comparable to booting
> > from a system that crashed (I'm making an educated guess at that one).
> >
> Not a good guess. To use DD you need a computer with dd and a fast cpu. I
> did top while dd was working and it was taking 70% of the cpu's time :-P
>
It's one of the oldest Unix programs: it will copy as fast as your
system can go. Yes, a fast system will copy faster.
i'm guessing he failed to provide a decent blocksize so dd was using
the default blocksize of 512 bytes. that will slow things down in a
hurry.
rday
--
========================================================================
Robert P. J. Day
Linux Consulting, Training and Annoying Kernel Pedantry
Waterloo, Ontario, CANADA
http://crashcourse.ca
========================================================================
11:36 a.m.
New subject: DD not working--SUCCESS!
Robert P. J. Day wrote:
On Fri, 31 Aug 2007, Ian Malone wrote:
> Karl Larsen wrote:
>> Jacques B. wrote:
>>> Because the clone would be of a
>>> running system. So booting from it would be comparable to booting
>>> from a system that crashed (I'm making an educated guess at that one).
>>>
>> Not a good guess. To use DD you need a computer with dd and a fast cpu. I
>> did top while dd was working and it was taking 70% of the cpu's time :-P
>>
> It's one of the oldest Unix programs: it will copy as fast as your
> system can go. Yes, a fast system will copy faster.
i'm guessing he failed to provide a decent blocksize so dd was using
the default blocksize of 512 bytes. that will slow things down in a
hurry.
How would you go about estimating a decent blocksize (other than by
testing)? My first instinct would be to go for some percentage of
the drive's cache. (And it had better be a factor of the amount you
want to copy I suppose.)
--
imalone
12:16 p.m.
New subject: DD not working--SUCCESS!
How would you go about estimating a decent blocksize (other than by
testing)? My first instinct would be to go for some percentage of
the drive's cache. (And it had better be a factor of the amount you
want to copy I suppose.)
--
imalone
Available RAM. Use as big a block size as can possibly be read into
RAM (once you start caching to a drive you just as well be writing to
the destination drive otherwise you slow it down even more). We
played around with this a while back and it certainly makes a
difference (stands to reason as was already explained).
The one caveat would be if there are errors on the drive. If that is
the case, a full block is dropped (so if your block size is 4096
bytes, then 4K gets dropped instead of 512 bytes - now imagine
dropping a 256 meg block of data...). ddrescue will use two block
sizes - a larger one and a smaller one. If it hits an error it drops
back to the smaller one until it gets pass the error and then ramps
back up to the larger block size.
Jacques
7:51 p.m.
New subject: DD not working--SUCCESS!
imalone:
> How would you go about estimating a decent blocksize (other than
by
> testing)? My first instinct would be to go for some percentage of
> the drive's cache.
Jacques B:
Available RAM. Use as big a block size as can possibly be read into
RAM (once you start caching to a drive you just as well be writing to
the destination drive otherwise you slow it down even more). We
played around with this a while back and it certainly makes a
difference (stands to reason as was already explained).
The one caveat would be if there are errors on the drive. If that is
the case, a full block is dropped (so if your block size is 4096
bytes, then 4K gets dropped instead of 512 bytes - now imagine
dropping a 256 meg block of data...). ddrescue will use two block
sizes - a larger one and a smaller one. If it hits an error it drops
back to the smaller one until it gets pass the error and then ramps
back up to the larger block size.
I had a little play around with blocksizes for dd a while ago, while
zeroing a drive (I was zeroing out a bad drive to see it it'd goad the
drive into automatically sorting out the bad blocks, by itself - it
did). The first attempt just started out using the default, and that
took ages. After a bit of fiddling, I settled for using about a meg.
Of course my situation was a bit different, but it's to show that you'd
tweak the option to suit your circumstances.
--
[tim@bigblack ~]$ uname -ipr
2.6.22.4-65.fc7 i686 i386
Using FC 4, 5, 6 & 7, plus CentOS 5. Today, it's FC7.
Don't send private replies to my address, the mailbox is ignored.
I read messages from the public lists.
6:38 a.m.
New subject: DD not working--SUCCESS!
Tim wrote:
imalone:
>> How would you go about estimating a decent blocksize (other than by
>> testing)? My first instinct would be to go for some percentage of
>> the drive's cache.
>>
Jacques B:
> Available RAM. Use as big a block size as can possibly be read into
> RAM (once you start caching to a drive you just as well be writing to
> the destination drive otherwise you slow it down even more).
The
discussion of block size says your in a hurry to do it. Here is
a modified HowTo for dd.
If you want to copy something big from one partition to another the
old dd method is for you. You have to do it right. This means that:
1. The destination partition MUST be at least a byte larger than the
source partition where the data is coming from. This is essential!
2. The dd copies everything including the file system.
3. Be ready to check the file system of the copy with fsck.
4. If you are making a copy of an entire working Fedora system make sure
you change all the entries in /etc/grub.conf and /etc/fstab files to the
new partitions of the copy before you try to run it.
5. To copy a partition from the source, called /dev/source to a
destination partition called /dev/destination use the standard dd form:
# dd if=/dev/source of=/dev/destination
6. Make certain that all Hard Drives are registered in bios. Put the F7
Rescue CD in, or your F7 DVD in, and select Rescue. When it asks if you
want to mount your fedora system do not just go to the Prompt. At the
prompt do your dd transfer.
7. Take your time doing all this. It will save you problems later. There
are a lot of ways to make errors.
8. Think of how dd works this way, dd see's the source partition as just
a pile of bytes. It takes a few bytes each cycle and puts those bytes on
the destination partition. When done dd reports how many bytes it found
and how many it put on destination. They are the same large number.
--
Karl F. Larsen, AKA K5DI
Linux User
#450462 http://counter.li.org.
9:51 a.m.
New subject: DD not working--SUCCESS!
Karl Larsen wrote:
If you want to copy something big from one partition to another the
old dd method is for you. You have to do it right. This means that:
1. The destination partition MUST be at least a byte larger than the
source partition where the data is coming from. This is essential!
This isn't true. Can you explain why you think it is?
3. Be ready to check the file system of the copy with fsck.
If your source partition is cleanly unmounted, the destination will be
clean as well.
8. Think of how dd works this way, dd see's the source partition
as just
a pile of bytes. It takes a few bytes each cycle and puts those bytes on
the destination partition. When done dd reports how many bytes it found
and how many it put on destination. They are the same large number.
This means it is important for the source to not change during the copy
so it should be unmounted.
--
Les Mikesell
lesmikesell(a)gmail.com
10:12 a.m.
New subject: DD not working--SUCCESS!
Les Mikesell wrote:
Karl Larsen wrote:
>
> If you want to copy something big from one partition to another
> the old dd method is for you. You have to do it right. This means that:
>
> 1. The destination partition MUST be at least a byte larger than the
> source partition where the data is coming from. This is essential!
This isn't true. Can you explain why you think it is?
Yes I can. My first
try with dd I tried to put a 40GB partition into
a 20GB partition and dd errored out. Then I read man dd.
> 3. Be ready to check the file system of the copy with fsck.
If your source partition is cleanly unmounted, the destination will be
clean as well. Perhaps so but I like to run fsck just to be sure
nothing is wrong.
> 8. Think of how dd works this way, dd see's the source partition as
> just a pile of bytes. It takes a few bytes each cycle and puts those
> bytes on the destination partition. When done dd reports how many
> bytes it found and how many it put on destination. They are the same
> large number.
This means it is important for the source to not change during the
copy so it should be unmounted.
It does and the method using the Rescue mode allows this.
--
Karl F. Larsen, AKA K5DI
Linux User
#450462 http://counter.li.org.
10:18 a.m.
New subject: DD not working--SUCCESS!
On Saturday 1 September 2007 4:12:02 pm Karl Larsen wrote:
Les Mikesell wrote:
> Karl Larsen wrote:
>> If you want to copy something big from one partition to another
>> the old dd method is for you. You have to do it right. This means that:
>>
>> 1. The destination partition MUST be at least a byte larger than the
>> source partition where the data is coming from. This is essential!
>
> This isn't true. Can you explain why you think it is?
Yes I can. My first try with dd I tried to put a 40GB partition into
a 20GB partition and dd errored out. Then I read man dd.
Erm, what.
Please explain how the fact that you cannot squeeze a 40G partition into a 20G
one, something I would have thought obvious, leads you to the conclusion that
the destination has to be bigger than the source.
All this proves is 40 > 20 : hardly ground a breaking discovery...
For the record the destination has to be THE SAME size or bigger. There is no
requirement (I am aware of) that it must be at least a byte bigger....
Chris
10:37 a.m.
New subject: DD not working--SUCCESS!
Chris Jones wrote:
On Saturday 1 September 2007 4:12:02 pm Karl Larsen wrote:
> Les Mikesell wrote:
>
>> Karl Larsen wrote:
>>
>>> If you want to copy something big from one partition to another
>>> the old dd method is for you. You have to do it right. This means that:
>>>
>>> 1. The destination partition MUST be at least a byte larger than the
>>> source partition where the data is coming from. This is essential!
>>>
>> This isn't true. Can you explain why you think it is?
>>
> Yes I can. My first try with dd I tried to put a 40GB partition into
> a 20GB partition and dd errored out. Then I read man dd.
>
Erm, what.
Please explain how the fact that you cannot squeeze a 40G partition into a 20G
one, something I would have thought obvious, leads you to the conclusion that
the destination has to be bigger than the source.
All this proves is 40 > 20 : hardly ground a breaking discovery...
For the record the destination has to be THE SAME size or bigger. There is no
requirement (I am aware of) that it must be at least a byte bigger....
Chris
I think this is a stupid discussion. To you Chris, "How do you make
sure the destination partition is EXACTLY equal in byte size to the
source partition?".
--
Karl F. Larsen, AKA K5DI
Linux User
#450462 http://counter.li.org.
10:41 a.m.
New subject: DD not working--SUCCESS!
I think this is a stupid discussion. To you Chris, "How do
you make
sure the destination partition is EXACTLY equal in byte size to the
source partition?".
How about having two completely identical disks, with one single partition on
each ? That would do it.
My point was just the destination disk doesn't have to be bigger. It can be
the same. You claimed it had to be a little bit bigger, by one byte. This is
incorrect.
Chris
10:42 a.m.
New subject: DD not working--SUCCESS!
Around 04:37pm on Saturday, September 01, 2007 (UK time), Karl Larsen scrawled:
I think this is a stupid discussion. To you Chris, "How do
you make
sure the destination partition is EXACTLY equal in byte size to the
source partition?".
Define them as the same size when you create them?
Steve
--
A: Because it messes up the order in which people normally read text.
Q: Why is top-posting a bad thing?
16:41:15 up 27 days, 5:05, 2 users, load average: 0.00, 0.01, 0.00
10:50 a.m.
New subject: DD not working--SUCCESS!
On Sat, 2007-09-01 at 09:37 -0600, Karl Larsen wrote:
"How do you make sure the destination partition is EXACTLY equal
in
byte size to the source partition?".
It's as simple as this: The size of the destination must not be less
than the size of the source. That covers equality, as well as ensuring
that any mismatch is in the right direction.
--
[tim@bigblack ~]$ uname -ipr
2.6.22.4-65.fc7 i686 i386
Using FC 4, 5, 6 & 7, plus CentOS 5. Today, it's FC7.
Don't send private replies to my address, the mailbox is ignored.
I read messages from the public lists.
10:57 a.m.
New subject: DD not working--SUCCESS!
Karl Larsen wrote:
Chris Jones wrote:
> On Saturday 1 September 2007 4:12:02 pm Karl Larsen wrote:
>
>> Les Mikesell wrote:
>>
>>> Karl Larsen wrote:
>>>
>>>> 1. The destination partition MUST be at least a byte
larger than the
>>>> source partition where the data is coming from. This is essential!
>>>>
>>
>
> Erm, what.
> Please explain how the fact that you cannot squeeze a 40G partition
> into a 20G one, something I would have thought obvious, leads you to
> the conclusion that the destination has to be bigger than the source.
> All this proves is 40 > 20 : hardly ground a breaking discovery...
>
> For the record the destination has to be THE SAME size or bigger.
> There is no requirement (I am aware of) that it must be at least a
> byte bigger....
I think this is a stupid discussion. To you Chris, "How do
you make
sure the destination partition is EXACTLY equal in byte size to the
source partition?".
As pointed out you define them this way. However I think there are two
more important things to look at here. First, why specify it must be at
least a byte larger when this is not the case. Secondly this approach
(using dd) only really makes sense when cloning the partition. This is
/why/ you specify it the same size, because for whatever reason you want
an exact copy of the partition. Other methods involving copying are
often more time efficient and do not suffer such restrictions on the
partition sizes or inefficiencies should you copy to a slightly larger
partition. (Depending of course on what you're doing. Fixing a broken
fs is example that has already been given of a good time to clone the
partition.)
--
imalone
10:35 a.m.
New subject: DD not working--SUCCESS!
On Sat, 1 Sep 2007, Karl Larsen wrote:
Les Mikesell wrote:
> Karl Larsen wrote:
> >
> > If you want to copy something big from one partition to another the old
> > dd method is for you. You have to do it right. This means that:
> >
> > 1. The destination partition MUST be at least a byte larger than the
> > source partition where the data is coming from. This is essential!
>
> This isn't true. Can you explain why you think it is?
Yes I can. My first try with dd I tried to put a 40GB partition
into a 20GB partition and dd errored out. Then I read man dd.
just kill me now, lord.
rday
--
========================================================================
Robert P. J. Day
Linux Consulting, Training and Annoying Kernel Pedantry
Waterloo, Ontario, CANADA
http://crashcourse.ca
========================================================================
4:23 p.m.
New subject: DD not working--SUCCESS!
Here is my final paper on using dd. Feel free to shoot at it.
If you want to copy something big from one partition to another the
old dd method is for you. You have to do it right. This means that:
1. The destination partition MUST be as large or larger than the source
partition where the data is coming from. This is essential!
2. The dd copies everything including the file system.
3. Be ready to check the file system of the copy with fsck.
4. If you are making a copy of an entire working Fedora system make sure
you change all the entries in /etc/grub.conf and /etc/fstab files to the
new partitions of the copy before you try to run it.
5. To copy a partition from the source, called /dev/source to a
destination partition called /dev/destination use the standard dd form:
# dd if=/dev/source of=/dev/destination
6. Make certain that all Hard Drives are registered in bios. Put the F7
Rescue CD in, or your F7 DVD in, and select Rescue. When it asks if you
want to mount your fedora system do not just go to the Prompt. At the
prompt do your dd transfer.
7. Take your time doing all this. It will save you problems later. There
are a lot of ways to make errors.
8. Think of how dd works this way, dd see's the source partition as just
a pile of bytes. It takes a few bytes each cycle and puts those bytes on
the destination partition. When done dd reports how many bytes it found
and how many it put on destination. They are the same large number.
9. The default number of bytes per Block is defaulted to 150. The info
DD says this:
`bs=BYTES'
Set both input and output block sizes to BYTES. This makes `dd'
read and write BYTES per block, overriding any `ibs' and `obs'
settings.
So if in a hurry you can use bs=15000 which will be a good thing because
the IDE hard drive controllers are quite slow. So rather than sending
just 150 bytes per change, you send 15,000. A transfer will look like:
# dd bs=15000 if=/dev/sda6 of=/dev/sdb5
I have not tried this. No reason to think it will not work.
--
Karl F. Larsen, AKA K5DI
Linux User
#450462 http://counter.li.org.
11:40 p.m.
New subject: DD not working--SUCCESS!
On Sat, 1 Sep 2007, Karl Larsen wrote:
Here is my final paper on using dd. Feel free to shoot at it.
8. Think of how dd works this way, dd see's the source partition
as
just a pile of bytes.
what is a "pile" of bytes?
9. The default number of bytes per Block is defaulted to 150.
only in your imagination, karl.
The info DD says
this:
`bs=BYTES'
Set both input and output block sizes to BYTES. This makes `dd'
read and write BYTES per block, overriding any `ibs' and `obs'
settings.
So if in a hurry you can use bs=15000 which will be a good thing because the
IDE hard drive controllers are quite slow. So rather than sending just 150
bytes per change, you send 15,000. A transfer will look like:
# dd bs=15000 if=/dev/sda6 of=/dev/sdb5
I have not tried this. No reason to think it will not work.
karl, why in god's name are you writing documentation ostensibly to
help others do things that you haven't even verified yourself?
please, karl, do everyone a favour, and stop trying to do them any
more favours. all you're going to do is lead someone into doing
something disastrously disastrous.
and please, karl, for the last time, get a blog where you can publish
all this incorrect and potentially damaging guidance. the amount of
everyone's time you've wasted in "finally" producing documentation
that is still incorrect is really quite breathtaking.
rday
--
========================================================================
Robert P. J. Day
Linux Consulting, Training and Annoying Kernel Pedantry
Waterloo, Ontario, CANADA
http://crashcourse.ca
========================================================================
6:11 a.m.
New subject: DD not working--SUCCESS!
Robert P. J. Day wrote:
On Sat, 1 Sep 2007, Karl Larsen wrote:
> Here is my final paper on using dd. Feel free to shoot at it.
>
> 8. Think of how dd works this way, dd see's the source partition as
> just a pile of bytes.
>
what is a "pile" of bytes?
> 9. The default number of bytes per Block is defaulted to 150.
>
only in your imagination, karl.
> The info DD says
> this:
> `bs=BYTES'
> Set both input and output block sizes to BYTES. This makes `dd'
> read and write BYTES per block, overriding any `ibs' and `obs'
> settings.
> So if in a hurry you can use bs=15000 which will be a good thing because the
> IDE hard drive controllers are quite slow. So rather than sending just 150
> bytes per change, you send 15,000. A transfer will look like:
>
> # dd bs=15000 if=/dev/sda6 of=/dev/sdb5
>
> I have not tried this. No reason to think it will not work.
>
karl, why in god's name are you writing documentation ostensibly to
help others do things that you haven't even verified yourself?
please, karl, do everyone a favour, and stop trying to do them any
more favours. all you're going to do is lead someone into doing
something disastrously disastrous.
and please, karl, for the last time, get a blog where you can publish
all this incorrect and potentially damaging guidance. the amount of
everyone's time you've wasted in "finally" producing documentation
that is still incorrect is really quite breathtaking.
rday
Thanks for your help rday.
--
Karl F. Larsen, AKA K5DI
Linux User
#450462 http://counter.li.org.
6:36 a.m.
New subject: DD not working--SUCCESS!
On Sun, 2 Sep 2007, Karl Larsen wrote:
... snip ...
Thanks for your help rday.
your fundamental problem, karl, is that you insist on publishing (and
re-publishing and re-re-publishing) documentation that is simply
wrong. and one of the things about documentation is that *bad*
documentation is even worse than *no* documentation -- if anyone had
decided to follow your early advice, chances are good they would have
screwed themselves thoroughly.
and *that's* the most compelling reason for you to get your own blog
-- so you can publish inaccurate and dangerous tutorials where there's
less chance of them hurting someone. because as long as you keep
posting them here, you're encouraging someone to actually believe you
know what you're talking about. and that is a guaranteed recipe for
disaster.
so, please, karl, for the sake of people who might be tempted to
listen to you and follow your advice, stop. just stop. and leave
writing helpful HOWTOs to those people who a) know what they're doing,
and b) have actually done it before. you don't need to be *writing*
documentation, karl; you need to be *reading* it.
rday
--
========================================================================
Robert P. J. Day
Linux Consulting, Training and Annoying Kernel Pedantry
Waterloo, Ontario, CANADA
http://crashcourse.ca
========================================================================
6:56 a.m.
New subject: DD not working--SUCCESS!
Robert P. J. Day wrote:
On Sun, 2 Sep 2007, Karl Larsen wrote:
... snip ...
> Thanks for your help rday.
>
your fundamental problem, karl, is that you insist on publishing (and
re-publishing and re-re-publishing) documentation that is simply
wrong. and one of the things about documentation is that *bad*
documentation is even worse than *no* documentation -- if anyone had
decided to follow your early advice, chances are good they would have
screwed themselves thoroughly.
and *that's* the most compelling reason for you to get your own blog
-- so you can publish inaccurate and dangerous tutorials where there's
less chance of them hurting someone. because as long as you keep
posting them here, you're encouraging someone to actually believe you
know what you're talking about. and that is a guaranteed recipe for
disaster.
so, please, karl, for the sake of people who might be tempted to
listen to you and follow your advice, stop. just stop. and leave
writing helpful HOWTOs to those people who a) know what they're doing,
and b) have actually done it before. you don't need to be *writing*
documentation, karl; you need to be *reading* it.
rday
First, the PAPER is in Gedit. It is written for me. With the
exception of 9. all that is there has been tested by me and it works
fine. I'm sorry that your the protector of the WORLD from bad writing.
But one thing your really good at rday is damming someones writing
without telling them what is wrong with it. If there is something wrong
with 9. what is it? Most of 9. I took out of info dd.
--
Karl F. Larsen, AKA K5DI
Linux User
#450462 http://counter.li.org.
7:01 a.m.
New subject: DD not working--SUCCESS!
On Sun, 2 Sep 2007, Karl Larsen wrote:
... my condemnation snipped ...
First, the PAPER is in Gedit. It is written for me.
but by posting it here, you are implicitly promoting it as good
advice for readers of this list. and that's simply irresponsible.
With the exception of 9. all that is there has been tested by me and
it works fine.
i'm not talking about the latest version (which, as i pointed out,
*still* has a blatant error in it), i'm talking about all the
*earlier* versions, which numerous readers kept correcting of their
faults.
a lot of us (me included) have invested time in writing
documentation that we're willing to share with the world, but we
generally take the time to polish it first -- we don't keep posting
revised versions, asking "ok, so what about *this* one? does *this*
one work?"
I'm sorry that your the protector of the WORLD from bad writing.
it's a tough job, but maybe it will keep someone from borking their
system because they followed your instructions. and if they do, are
you willing to be held responsible? hmmmmm?
But one thing your really good at rday is damming someones writing
without telling them what is wrong with it. If there is something
wrong with 9. what is it? Most of 9. I took out of info dd.
and i quote, verbatim:
"9. The default number of bytes per Block is defaulted to 150."
here, karl ... let me introduce you to google:
http://www.google.com/search?q=dd+default+block+size&sourceid=navclie...
and, just because i'm feeling masochistic, karl, i'm going to ask
you one more time -- why do you not get your own blog? that way, you
can publish to your heart's content, and you'll have a permanent
record of everything you write, and you can wander off-topic as much
as you want, and ***no one will care***.
***why***, karl, for god's sake, do you keep inflicting your mostly
pointless and oftentimes erroneous meanderings on this mailing list?
how many people need to ask you to get a blog before you take it
seriously? really, how many? give us a number, and we'll see what we
can do.
rday
--
========================================================================
Robert P. J. Day
Linux Consulting, Training and Annoying Kernel Pedantry
Waterloo, Ontario, CANADA
http://crashcourse.ca
========================================================================
7:39 a.m.
New subject: DD not working--SUCCESS!
El Domingo, 2 de Septiembre de 2007 14:01, Robert P. J. Day escribió:
***why***, karl, for god's sake, do you keep inflicting your mostly
pointless and oftentimes erroneous meanderings on this mailing list?
how many people need to ask you to get a blog before you take it
seriously? really, how many? give us a number, and we'll see what we
can do.
He doesn't understand either what a blog or a mailing list is.
--
Manuel Arostegui Ramirez.
Electronic Mail is not secure, may not be read every day, and should not
be used for urgent or sensitive issues.
12:11 p.m.
New subject: DD not working--SUCCESS!
On Sun, 2007-09-02 at 14:39 +0200, Manuel Arostegui Ramirez wrote:
El Domingo, 2 de Septiembre de 2007 14:01, Robert P. J. Day
escribió:
>
> ***why***, karl, for god's sake, do you keep inflicting your mostly
> pointless and oftentimes erroneous meanderings on this mailing list?
> how many people need to ask you to get a blog before you take it
> seriously? really, how many? give us a number, and we'll see what we
> can do.
>
He doesn't understand either what a blog or a mailing list is.
Pardon me,
but while Karl is sometimes annoying, the repeated piling on is really
beginning to get to me. I like this list and I do use it. Where Karl
is wrong, tell him what is wrong and why. Then let it go.
Karl, quit baiting for trolls.
Now, please try to be civil, try to give information that is helpful,
and where Karl is wrong, edit the information and correct it. That
won't take any longer than this continual bickering. In addition, this
stuff will prevent others, especially new folks, from joining the list.
They will know that some of their questions are old to us, and when they
see someone continually taken to task, why would they risk saying
anything. Think about it.
For me, I'm a crusty old codger and generally state my mind, so it
doesn't prevent me from speaking up, but most people are much more
timid, especially when it comes to highly technical things like OS's and
computers. We do want new people on Linux, don't we?
Regards,
Les H
12:29 p.m.
New subject: DD not working--SUCCESS!
El Domingo, 2 de Septiembre de 2007 19:11, Les escribió:
Pardon me,
but while Karl is sometimes annoying, the repeated piling on is really
beginning to get to me. I like this list and I do use it. Where Karl
is wrong, tell him what is wrong and why. Then let it go.
Loads of folks, included myself, already did it. He seems to either don't
understand or just overlook the issue.
Karl, quit baiting for trolls.
Now, please try to be civil, try to give information that is helpful,
and where Karl is wrong, edit the information and correct it. That
won't take any longer than this continual bickering.
If I were him and I would set up a wiki and give here the URL, then, all the
list members interested in helping him to write a good doc, could move to the
wiki and work on the document there, editing and disscusing the issues in the
proper place. Just my opinion.
Open a new thread called "No" is not much helpful.
In addition, this
stuff will prevent others, especially new folks, from joining the list.
They will know that some of their questions are old to us, and when they
see someone continually taken to task, why would they risk saying
anything. Think about it.
Well, don't think so actually, I think that new people who join the list wants
solution, and they don't read the whole list before send their emails.
For me, I'm a crusty old codger and generally state my mind, so it
doesn't prevent me from speaking up, but most people are much more
timid, especially when it comes to highly technical things like OS's and
computers.
Maybe.
We do want new people on Linux, don't we?
Of course, at least, I want.
All the best
Manuel
--
Manuel Arostegui Ramirez.
Electronic Mail is not secure, may not be read every day, and should not
be used for urgent or sensitive issues.
7:53 a.m.
New subject: LEAVE Karl along (WAS Re: DD not working--SUCCESS!)
On Sun, 2007-09-02 at 08:01 -0400, Robert P. J. Day wrote:
***why***, karl, for god's sake, do you keep inflicting your
mostly
pointless and oftentimes erroneous meanderings on this mailing list?
how many people need to ask you to get a blog before you take it
seriously? really, how many? give us a number, and we'll see what we
can do.
<Steps on soap box>
You and everyone else that keeps jumping on Karl's back just needs to
please QUIT! You (not you personally rjay) guys/girls that keep harping
on him and jumping on him just need to stop. You all are being more
annoying then HE is by yelling at him. I have emailed him off list week
or two ago, and explained a few things and did it in a nice, compelling
and calm email. And since then, he has somewhat (maybe not totally yet)
calmed his emails down to a point, and trying to follow the *list rules*
as they are so called. He wrote a frickin paper (and it was to HIM not
really US), to outline about dd, and he was just posting to help pick at
it a little. Help correct it and leave it alone. That is up to him to
follow your advice. If you try and he don't, then fine leave it alone
and don't reply anymore.
Please, he has calmed it down some, I certainly noticed, so let him be
and do what he's going. Sheesh, he's doing heck of a lot better at
participating then the people that get on here and do nothing but moan
and complain. At least he is trying to participate and be a part of
Fedora.
<off soap box>
Sorry for stealing your email rjay and using it to reply to, just was
the one that caught me going initially. Have a nice day all :)
--
Mike Chambers
Madisonville, KY
"Best little town on Earth!"
7:50 a.m.
New subject: LEAVE Karl along (WAS Re: DD not working--SUCCESS!)
On Sun, 2 Sep 2007, Mike Chambers wrote:
... He wrote a frickin paper (and it was to HIM not really US), to
outline about dd, and he was just posting to help pick at it a
little. Help correct it and leave it alone. That is up to him to
follow your advice. If you try and he don't, then fine leave it
alone and don't reply anymore.
i appreciate your playing peacemaker, mike, but there's a
significant difference between:
1) posting a tutorial-in-progress and asking readers for help in
filling in the blanks, and
2) posting an apparently completed tutorial and getting readers to
correct the perpetual (and potentially dangerous) errors therein.
and as long as karl insists on following strategy 2), he should get
smacked for it. and, on that note, i've had my say, and will now go
off and add karl to the killfile. life is just too short, as they
say.
rday
========================================================================
Robert P. J. Day
Linux Consulting, Training and Annoying Kernel Pedantry
Waterloo, Ontario, CANADA
http://crashcourse.ca
========================================================================
8:37 a.m.
New subject: LEAVE Karl along (WAS Re: DD not working--SUCCESS!)
On Sun, 2007-09-02 at 08:50 -0400, Robert P. J. Day wrote:
On Sun, 2 Sep 2007, Mike Chambers wrote:
> ... He wrote a frickin paper (and it was to HIM not really US), to
> outline about dd, and he was just posting to help pick at it a
> little. Help correct it and leave it alone. That is up to him to
> follow your advice. If you try and he don't, then fine leave it
> alone and don't reply anymore.
i appreciate your playing peacemaker, mike, but there's a
significant difference between:
1) posting a tutorial-in-progress and asking readers for help in
filling in the blanks, and
2) posting an apparently completed tutorial and getting readers to
correct the perpetual (and potentially dangerous) errors therein.
and as long as karl insists on following strategy 2), he should get
smacked for it. and, on that note, i've had my say, and will now go
off and add karl to the killfile. life is just too short, as they
say.
rday
========================================================================
Robert P. J. Day
Linux Consulting, Training and Annoying Kernel Pedantry
Waterloo, Ontario, CANADA
http://crashcourse.ca
========================================================================
I haven't had any bandwidth to keep up on this thread...looks like I
missed a "spirited discussion." ;-]
-- Mark C. Allman, PMP
-- Allman Professional Consulting,
Inc.
-- www.allmanpc.com, 617-947-4263
BusinessMsg -- the secure, managed, J2EE/AJAX Enterprise IM/IC solution.
6:37 p.m.
New subject: LEAVE Karl along (WAS Re: DD not working--SUCCESS!)
On Sun, 2007-09-02 at 08:50 -0400, Robert P. J. Day wrote:
there's a significant difference between:
1) posting a tutorial-in-progress and asking readers for help in
filling in the blanks, and
2) posting an apparently completed tutorial and getting readers to
correct the perpetual (and potentially dangerous) errors therein.
and as long as karl insists on following strategy 2), he should get
smacked for it.
Seconded! On any mail list, there always seems to be one or two who do
not understand what they're doing, but then they proceed to try and
explain something that they don't know to someone else, wasting that
other person's time. And they usually discuss the subject as if they're
an expert on it (presenting wrong information as fact, etc.).
I know I don't like having my time wasted, and I've a particular bugbear
about incompetent teaching. I've seen Karl, for one example, give some
incredibly stupid advice on occasions. That sort of thing DOES NEED
countering. Firstly to correct the wrong advice, and secondly to stop
the person from *continually* doing so.
And therein lay the big problem, the *continual* doing so... By all
means the unititiated should join in conversations to learn about
something new. But they shouldn't start giving out advice when they
don't know what they're talking about. He's had polite messages, to no
effect. So the ante does need upping until it has the required effect.
--
[tim@bigblack ~]$ uname -ipr
2.6.22.4-65.fc7 i686 i386
Using FC 4, 5, 6 & 7, plus CentOS 5. Today, it's FC7.
Don't send private replies to my address, the mailbox is ignored.
I read messages from the public lists.
8:11 a.m.
New subject: LEAVE Karl along (WAS Re: DD not working--SUCCESS!)
El Domingo, 2 de Septiembre de 2007 14:53, Mike Chambers escribió:
On Sun, 2007-09-02 at 08:01 -0400, Robert P. J. Day wrote:
> ***why***, karl, for god's sake, do you keep inflicting your mostly
> pointless and oftentimes erroneous meanderings on this mailing list?
> how many people need to ask you to get a blog before you take it
> seriously? really, how many? give us a number, and we'll see what we
> can do.
<Steps on soap box>
You and everyone else that keeps jumping on Karl's back just needs to
please QUIT! You (not you personally rjay) guys/girls that keep harping
on him and jumping on him just need to stop. You all are being more
annoying then HE is by yelling at him. I have emailed him off list week
or two ago, and explained a few things and did it in a nice, compelling
and calm email. And since then, he has somewhat (maybe not totally yet)
calmed his emails down to a point, and trying to follow the *list rules*
as they are so called. He wrote a frickin paper (and it was to HIM not
really US), to outline about dd, and he was just posting to help pick at
it a little. Help correct it and leave it alone. That is up to him to
follow your advice. If you try and he don't, then fine leave it alone
and don't reply anymore.
Please, he has calmed it down some, I certainly noticed, so let him be
and do what he's going. Sheesh, he's doing heck of a lot better at
participating then the people that get on here and do nothing but moan
and complain. At least he is trying to participate and be a part of
Fedora.
<off soap box>
Sorry for stealing your email rjay and using it to reply to, just was
the one that caught me going initially. Have a nice day all :)
Yep, I noticed he calmed down, but I wonder why after all those previus
replies (previous to this thread), even the mail you have said you sent to
him, why does he still keep using this mailing list as his personal blog?
I appreciate people who tried to get involved in all the linux stuff,
especially in Fedora. Writting manuals, translating things etc...but I don't
understand how he tries to write a howto which is wrong and may make people
to get confused when they read it (since google will index it).
I do understand if he wants to write some PERSONAL documentation (as he said),
but, I don't see the point in sending it to the list.
I've also noticed Karl is trying to help new people here as much as he can,
and that's great, but from that to write useful papers...
All the best.
Manuel
--
Manuel Arostegui Ramirez.
Electronic Mail is not secure, may not be read every day, and should not
be used for urgent or sensitive issues.
9:53 p.m.
New subject: LEAVE Karl along (WAS Re: DD not working--SUCCESS!)
On 9/2/07, Mike Chambers <mike(a)miketc.com> wrote:
On Sun, 2007-09-02 at 08:01 -0400, Robert P. J. Day wrote:
> ***why***, karl, for god's sake, do you keep inflicting your mostly
> pointless and oftentimes erroneous meanderings on this mailing list?
> how many people need to ask you to get a blog before you take it
> seriously? really, how many? give us a number, and we'll see what we
> can do.
<Steps on soap box>
You and everyone else that keeps jumping on Karl's back just needs to
please QUIT!
There is rarely an excuse to be spawning threads like an out of control process.
--
Fedora 7 : sipping some of that moonshine
( www.pembo13.com )
7:20 a.m.
New subject: DD not working--SUCCESS!
First, the PAPER is in Gedit.
why is that relevant. I'm writing this email using kmail, does that also
matter ?
It is written for me.
In which case why did you post it to this list. Keep it private, to yourself.
The problem is you post things to this list, like your 'papers' that are
specific to your hardware in a manor that suggests they are a solution for
all. This is really dangerous since, as Robert pointed out people might read
your information, try it on their machines and wonder why it doesn't work for
them.
As long as your insist on posting incorrect or misleading information here,
people will shoot it down. There is nothing personal in this, just we don't
want anyone using your information and screwing their systems because of it.
Chris
8:04 a.m.
New subject: DD not working--SUCCESS!
On Sun September 2 2007, Robert P. J. Day wrote:
your fundamental problem, karl, is that you insist on publishing
(and
re-publishing and re-re-publishing)
The only thing of note about your arrogant posts are a couple of words from
your sig - you are indeed what you say, an annoying pedant. If you are truly
concerned about the people who will hose systems due to bad advice, correct
the paper - JUST DO IT. This incessant hollering at one individual, and your
role in the incitement to mob behavior on the part of a tiny handful of
others last week are not helpful at all - if you're so concerned about being
helpful, be so. Cut your hypocrisy - there are plenty of others who post
erroneous information - are you going to go around policing them all - do you
think you can? You don't like Karl - so be it - who cares? There's about ten
or so of you -- this has become about personality, and despite all the
posturing to the contrary, that is pretty obvious. To the thousands of others
who read this list and the hundreds that are continuing the business of
helping each other as usual, ignoring you obnoxious ten, you're a pimple on
your own collective arse.
What matters it an iota, whether you've written tomes of clear helpful
documentation ... you have not charity
--
Claude Jones
Brunswick, MD, USA
8:16 a.m.
New subject: DD not working--SUCCESS!
El Domingo, 2 de Septiembre de 2007 15:04, Claude Jones escribió:
The only thing of note about your arrogant posts are a couple of words from
your sig - you are indeed what you say, an annoying pedant. If you are
truly concerned about the people who will hose systems due to bad advice,
correct the paper - JUST DO IT. This incessant hollering at one individual,
and your role in the incitement to mob behavior on the part of a tiny
handful of others last week are not helpful at all - if you're so concerned
about being helpful, be so. Cut your hypocrisy - there are plenty of others
who post erroneous information - are you going to go around policing them
all - do you think you can? You don't like Karl - so be it - who cares?
Well, do be that radical. From my point of view this is not a matter of who
likes or doesn't like Karl.
Since none of us knows him personally, we can only disscuss about his posts.
Some weeks ago his posts where slighty off topic or just...annonying.
Most likely the sort of posts you might need to write in a blog, not in a
mailing list where people try to help each other.
As Mike wrote, he calmed down, and he seems to understand how, more or less,
this list works, which is great.
Cheers
Manuel
--
Manuel Arostegui Ramirez.
Electronic Mail is not secure, may not be read every day, and should not
be used for urgent or sensitive issues.
8:38 a.m.
New subject: DD not working--SUCCESS!
On Sun September 2 2007, Manuel Arostegui Ramirez wrote:
Well, do be that radical. From my point of view this is not a matter
of who
likes or doesn't like Karl.
I can't speak to your specific point of view - but, I am speaking to general
appearances. The insane hostility that came out in recent days...
Since none of us knows him personally, we can only disscuss about
his
posts. Some weeks ago his posts where slighty off topic or
just...annonying.
So what? Does that justify the invective that was heaped on him
Most likely the sort of posts you might need to write in
a blog, not in a mailing list where people try to help each other.
yeah, yeah...
As Mike wrote, he calmed down, and he seems to understand how, more
or
less, this list works, which is great.
You speak of Karl in the third person, like he was some vegetable - "he didn't
taste so good before, but now I put some salt on him and he's much better..."
You snipped the most important point I make, in my view, which is the
appalling lack of charity exhibited here -- witness Mr. Day's remarks
about "slapping him around" in his last huffy post. Do you feel all warm and
fuzzy with that metaphor?
--
Claude Jones
Brunswick, MD, USA
8:56 a.m.
New subject: DD not working--SUCCESS!
El Domingo, 2 de Septiembre de 2007 15:38, Claude Jones escribió:
> Since none of us knows him personally, we can only disscuss
about his
> posts. Some weeks ago his posts where slighty off topic or
> just...annonying.
So what? Does that justify the invective that was heaped on him
He was told hundred of times to avoid those emails. He said nothing, and
despite of that, he kept sending offtopics.
I do understand people hostility to him, when you tell the same thing three,
four, five times and nothing change...
> Most likely the sort of posts you might need to write in
> a blog, not in a mailing list where people try to help each other.
yeah, yeah...
> As Mike wrote, he calmed down, and he seems to understand how, more or
> less, this list works, which is great.
You speak of Karl in the third person, like he was some vegetable - "he
didn't taste so good before, but now I put some salt on him and he's much
better..."
Hehehe I admit that was funny :-)
On the other hand, I think you know what I meant.
It's great and would be better if he definetly uses the list to help people
and not as his personal blog. I'm glad he has already started to change his
behaviour.
You snipped the most important point I make, in my view, which is the
appalling lack of charity exhibited here -- witness Mr. Day's remarks
about "slapping him around" in his last huffy post. Do you feel all warm
and fuzzy with that metaphor?
The first emails sent to Karl, I guess they were polite and nice. Trying to
make him understand what is this list about and stuff.
As I said above, no all of us has the same patience, and it's normal in a
mailing list with lot of people that some of the get mad when their mails, in
the beggining, polite ones, have no effect.
Anyways, just to set things straight, I want to be clear, I have nothing
against Karl, I just don't understand some kinda emails he sent to the list
some time before.
Cheers
Manuel
--
Manuel Arostegui Ramirez.
Electronic Mail is not secure, may not be read every day, and should not
be used for urgent or sensitive issues.
6:40 p.m.
New subject: DD not working--SUCCESS!
On Sun, 2007-09-02 at 15:16 +0200, Manuel Arostegui Ramirez wrote:
As Mike wrote, he calmed down, and he seems to understand how, more
or
less, this list works, which is great.
I don't see much evidence of any change.
--
[tim@bigblack ~]$ uname -ipr
2.6.22.4-65.fc7 i686 i386
Using FC 4, 5, 6 & 7, plus CentOS 5. Today, it's FC7.
Don't send private replies to my address, the mailbox is ignored.
I read messages from the public lists.
8:54 a.m.
New subject: DD not working--SUCCESS!
Around 02:04pm on Sunday, September 02, 2007 (UK time), Claude Jones scrawled:
On Sun September 2 2007, Robert P. J. Day wrote:
> your fundamental problem, karl, is that you insist on publishing (and
> re-publishing and re-re-publishing)
The only thing of note about your arrogant posts are a couple of words from
your sig - you are indeed what you say, an annoying pedant. If you are truly
concerned about the people who will hose systems due to bad advice, correct
the paper - JUST DO IT. This incessant hollering at one individual, and your
role in the incitement to mob behavior on the part of a tiny handful of
others last week are not helpful at all - if you're so concerned about being
helpful, be so. Cut your hypocrisy - there are plenty of others who post
erroneous information - are you going to go around policing them all - do you
think you can? You don't like Karl - so be it - who cares? There's about ten
or so of you -- this has become about personality, and despite all the
posturing to the contrary, that is pretty obvious. To the thousands of others
who read this list and the hundreds that are continuing the business of
helping each other as usual, ignoring you obnoxious ten, you're a pimple on
your own collective arse.
As (presumably) one of the ten pimples, I think there are more than ten
of us who believe that Karl is perverting the purpose of the list, and
that it is right we try and change this behaviour, by pointing this out.
Just because he is stubborn does not mean we have to accept this
behaviour, which if it was copied by others on the list would lead to a
total breakdown to the usefulness of the list.
Karl's streams of consciousness should be published on a blog, not the
mailing list.
If Karl wants to develop documentation he should publish it on a web
page or wiki, and invite people from here to comment on / correct it.
That way a *correct* final version can be developed, without incorrect
intermediate versions being recorded on the list archives forever, where
they could easily mislead others.
Where Karl has raised a problem on the list he should stick to the same
thread during its resolution. Raising multiple threads on the same
subject just leads to confusion for everyone.
Lastly, and specifically aimed at you Claude, calling others on the list
"a pimple on your own collective arse" ruins any argument you may be
making that we are being rude, nasty or hurtful to Karl and that it is
wrong for us to do so. Luckily (it seems so far) the remainder of this
list don't resort to childish invective - presumably they are more
intelligent and/or mature than you.
Steve
--
A: Because it messes up the order in which people normally read text.
Q: Why is top-posting a bad thing?
14:31:35 up 28 days, 2:55, 1 user, load average: 0.00, 0.00, 0.02
9:27 a.m.
New subject: DD not working--SUCCESS!
On Sun September 2 2007, Steve Searle wrote:
As (presumably) one of the ten pimples,
most perceptive of you
I think there are more than ten
of us who believe that Karl is perverting the purpose of the list, and
that it is right we try and change this behaviour, by pointing this out.
you are an intemperate few - I have noted no great rush to pile on to your
defense - I have noticed your incessant attempts to annoint yourselves as
spokesmen for the 'silent' masses
Just because he is stubborn does not mean we have to accept this
behaviour, which if it was copied by others on the list would lead to a
total breakdown to the usefulness of the list.
Karl's streams of consciousness should be published on a blog, not the
mailing list.
If Karl wants to develop documentation he should publish it on a web
page or wiki, and invite people from here to comment on / correct it.
That way a *correct* final version can be developed, without incorrect
intermediate versions being recorded on the list archives forever, where
they could easily mislead others.
Where Karl has raised a problem on the list he should stick to the same
thread during its resolution. Raising multiple threads on the same
subject just leads to confusion for everyone.
note - I have not attempted to defend Karl in most of these discussions - it's
never been the issue for me
Lastly, and specifically aimed at you Claude, calling others on the
list
"a pimple on your own collective arse" ruins any argument you may be
making that we are being rude, nasty or hurtful to Karl and that it is
wrong for us to do so. Luckily (it seems so far) the remainder of this
list don't resort to childish invective - presumably they are more
intelligent and/or mature than you.
ruins my argument in your intemperate brains - it was not you I hoped to
convince - you, in particular, have used some rather nasty formulations and
lame attempts at sophomoric humor repeatedly in your attacks, so, you are the
on the receiving end now - you chose to respond to me, I assume you don't
like it, good!
--
Claude Jones
Brunswick, MD, USA
5:23 p.m.
New subject: DD not working--SUCCESS!
Claude Jones wrote:
On Sun September 2 2007, Steve Searle wrote:
> I think there are more than ten
> of us who believe that Karl is perverting the purpose of the list, and
> that it is right we try and change this behaviour, by pointing this out.
you are an intemperate few - I have noted no great rush to pile on to your
defense - I have noticed your incessant attempts to annoint yourselves as
spokesmen for the 'silent' masses
I will step up and defend Steve. He is not one of the "intemperate
few" but one of the many (like myself) who normally stays relatively
silent. If we are all standing around having a conversation, we expect
each person to be polite and follow the unwritten protocol called "good
manners". These manners have been (hopefully) taught to us by our
parents, society, etc. so we can function in a civil manner towards one
another. When we have people join our group who are absolutely
determined to not follow the established protocol (i.e. good manners),
then usually some of the group will try to gently or not-so-gently
inform this person of what is expected. Just because someone does not
want to exercise good manners does not mean that we as a group have to
lower our standards for what we expect everyone to do. Nor does it mean
that we should just sit back and allow others to continually abuse the list.
Patrick
1:46 p.m.
New subject: DD not working--SUCCESS!
Robert P. J. Day wrote:
On Fri, 31 Aug 2007, Ian Malone wrote:
> Karl Larsen wrote:
>
>> Jacques B. wrote:
>>
>>> Because the clone would be of a
>>> running system. So booting from it would be comparable to booting
>>> from a system that crashed (I'm making an educated guess at that one).
>>>
>>>
>> Not a good guess. To use DD you need a computer with dd and a fast cpu. I
>> did top while dd was working and it was taking 70% of the cpu's time :-P
>>
>>
> It's one of the oldest Unix programs: it will copy as fast as your
> system can go. Yes, a fast system will copy faster.
>
i'm guessing he failed to provide a decent blocksize so dd was using
the default blocksize of 512 bytes. that will slow things down in a
hurry.
rday
I used the default block size Robert. What size would you use and why?
--
Karl F. Larsen, AKA K5DI
Linux User
#450462 http://counter.li.org.
1:32 p.m.
New subject: DD not working--SUCCESS!
Ian Malone wrote:
--
Karl F. Larsen, AKA K5DI
Linux User
#450462 http://counter.li.org.
Karl Larsen wrote:
> Jacques B. wrote:
>> Because the clone would be of a
>> running system. So booting from it would be comparable to booting
>> from a system that crashed (I'm making an educated guess at that one).
>>
> Not a good guess. To use DD you need a computer with dd and a fast
> cpu. I did top while dd was working and it was taking 70% of the
> cpu's time :-P
>
It's one of the oldest Unix programs: it will copy as fast as your
system can go. Yes, a fast system will copy faster.
(BTW, it is a good guess. You are copying straight off the device,
bypassing the filesystem.
Please explain. I am taking bytes off and putting them on file
systems. Both Hard Drives need to be running.
The primary problem I think is the
possibility of missing cached writes. Things like log files will
also be in an odd state, but that is unlikely to ever be critical.
However backups, and what you are doing is essentially the same
thing, are best done on unmounted systems.)
I can say that my success was with
me sending and receiving emails
and additions to the web browser while dd on this computer was doing
it's thing. No problems.
3:16 p.m.
New subject: DD not working--SUCCESS!
No, you're taking bytes off and putting them on drives. Neither
drive
should be mounted.
Matt Flaschen
Further to this, dd is a powerful utility that allows you to write
directly to a device. Whereas cp, rsync, and other such commands
require a mounted file system. dd copies from one drive to the other,
oblivious to the actual content. Just that byte 1 on /dev/sda5 is
\x65, so write \x65 to byte 1 on /dev/sdb6. Byte 2 on /dev/sda5 is
\x80, so write \x80 to byte 2 on /dev/sdb6. All this happens at the
physical layer. Whereas cp for example will look at the information
in the inode to determine where to find the file, and how many bytes
to copy so it requires a file system in order to work.
Even though a drive is not mounted, so long as it is connected
properly dd can read/write to it. Because it is doing so at the
physical layer...
I realize there may be some information overload going on here. But
hopefully my analogy of photocopying the book illustrates the danger
of dd'ing a live system. In some scenarios that would have to be done
(computer forensic investigation of a server that cannot be taken down
for example). But not in your scenario. Picture a mechanic trying to
work on an engine that is running. Now imagine trying to rebuild that
engine while it's running. In the case of copying an entire operating
system and user files as you are doing, dd is rebuilding a complexe
system (granted using a very simple process). That cannot be done
with any assurance that it will work properly if the system being
copied is running hence in a state of constant change. It's not like
using rsync or other such tools that work at the file system level and
will place a lock on a file when doing something with it so that it
cannot be changed while it is copying it.
If after everything everybody has told you you still do not consider
dd'ing a live system as a poor and risky practice then nobody will
ever convince you.
Jacques B.
3:42 p.m.
New subject: DD not working--SUCCESS!
Jacques B. wrote:
> No, you're taking bytes off and putting them on drives.
Neither drive
> should be mounted.
>
> Matt Flaschen
>
Further to this, dd is a powerful utility that allows you to write
directly to a device. Whereas cp, rsync, and other such commands
require a mounted file system. dd copies from one drive to the other,
oblivious to the actual content. Just that byte 1 on /dev/sda5 is
\x65, so write \x65 to byte 1 on /dev/sdb6. Byte 2 on /dev/sda5 is
\x80, so write \x80 to byte 2 on /dev/sdb6. All this happens at the
physical layer. Whereas cp for example will look at the information
in the inode to determine where to find the file, and how many bytes
to copy so it requires a file system in order to work.
Even though a drive is not mounted, so long as it is connected
properly dd can read/write to it. Because it is doing so at the
physical layer...
I realize there may be some information overload going on here. But
hopefully my analogy of photocopying the book illustrates the danger
of dd'ing a live system. In some scenarios that would have to be done
(computer forensic investigation of a server that cannot be taken down
for example). But not in your scenario. Picture a mechanic trying to
work on an engine that is running. Now imagine trying to rebuild that
engine while it's running. In the case of copying an entire operating
system and user files as you are doing, dd is rebuilding a complexe
system (granted using a very simple process). That cannot be done
with any assurance that it will work properly if the system being
copied is running hence in a state of constant change. It's not like
using rsync or other such tools that work at the file system level and
will place a lock on a file when doing something with it so that it
cannot be changed while it is copying it.
If after everything everybody has told you you still do not consider
dd'ing a live system as a poor and risky practice then nobody will
ever convince you.
Jacques B.
I of course see what your saying. It would be best if you have a
computer dedicated to running dd. You plug in 2 hard drives and boot up
hoping bios takes care of the hard drives :-)
On the computer you say dd if=hd1 of=hd2 and neither Hard Drive is
doing anything else.
But I don't have that many computers and hard drives. So I do what I
can and it appears to have worked.
--
Karl F. Larsen, AKA K5DI
Linux User
#450462 http://counter.li.org.
3:50 p.m.
New subject: DD not working--SUCCESS!
On Fri, 31 Aug 2007, Karl Larsen wrote:
I of course see what your saying. It would be best if you have a
computer dedicated to running dd. You plug in 2 hard drives and boot up
hoping bios takes care of the hard drives :-)
On the computer you say dd if=hd1 of=hd2 and neither Hard Drive is
doing anything else.
But I don't have that many computers and hard drives. So I do what I
can and it appears to have worked.
So far as I can tell, the important thing
is not having a separate computer.
It is not mounting any partition on either drive.
A live CD can boot and run without using a hard drive.
It makes its file systems in ram.
--
Mike hennebry(a)web.cs.ndsu.NoDak.edu
"Horse guts never lie." -- Cherek Bear-Shoulders
5:08 p.m.
New subject: DD not working--SUCCESS!
Michael Hennebry wrote:
On Fri, 31 Aug 2007, Karl Larsen wrote:
> I of course see what your saying. It would be best if you have a
> computer dedicated to running dd. You plug in 2 hard drives and boot up
> hoping bios takes care of the hard drives :-)
>
> On the computer you say dd if=hd1 of=hd2 and neither Hard Drive is
> doing anything else.
>
> But I don't have that many computers and hard drives. So I do what I
> can and it appears to have worked.
>
So far as I can tell, the important thing
is not having a separate computer.
It is not mounting any partition on either drive.
A live CD can boot and run without using a hard drive.
It makes its file systems in ram.
I don't have enough RAM for a live CD to do dd. The Rescue CD does
not have dd. I'm wondering if the F7 DVD might have dd? It has it
somewhere.
--
Karl F. Larsen, AKA K5DI
Linux User
#450462 http://counter.li.org.
6:21 p.m.
New subject: DD not working--SUCCESS!
On Fri, 31 Aug 2007, Karl Larsen wrote:
Michael Hennebry wrote:
> So far as I can tell, the important thing
> is not having a separate computer.
> It is not mounting any partition on either drive.
> A live CD can boot and run without using a hard drive.
> It makes its file systems in ram.
>
>
I don't have enough RAM for a live CD to do dd. The Rescue CD does
not have dd. I'm wondering if the F7 DVD might have dd? It has it
somewhere.
You seem to have a rare rescue CD.
One can get away with mounting a source partition read only.
You can use the dd to copy a read only disk.
Do you have a swap partition?
Some live CDs can be told to use that.
It doesn't matter much if a swap partition
changes while it's being copied.
I think that with the right boot parameters
you can come up with everything mounted read only.
In that case, running dd should be possible and reliable.
--
Mike hennebry(a)web.cs.ndsu.NoDak.edu
"Horse guts never lie." -- Cherek Bear-Shoulders
6:49 p.m.
New subject: DD not working--SUCCESS!
Michael Hennebry wrote:
On Fri, 31 Aug 2007, Karl Larsen wrote:
> Michael Hennebry wrote:
>
>> So far as I can tell, the important thing
>> is not having a separate computer.
>> It is not mounting any partition on either drive.
>> A live CD can boot and run without using a hard drive.
>> It makes its file systems in ram.
>>
>>
>>
> I don't have enough RAM for a live CD to do dd. The Rescue CD does
> not have dd. I'm wondering if the F7 DVD might have dd? It has it
> somewhere.
>
You seem to have a rare rescue CD.
One can get away with mounting a source partition read only.
You can use the dd to copy a read only disk.
Do you have a swap partition?
Some live CDs can be told to use that.
It doesn't matter much if a swap partition
changes while it's being copied.
I think that with the right boot parameters
you can come up with everything mounted read only.
In that case, running dd should be possible and reliable.
I am WRONG. My F7 Rescue CD Rom has dd built in. I am wanting to try
it soon to see how a RAM mounted system Will work. I may need to load a
system just to have one working fine to put on the New Hard Drive. I
think tomorrow
Both Hard Drives will be running but not mounted or doing a thing. It
should be a perfect copy 8-)
--
Karl F. Larsen, AKA K5DI
Linux User
#450462 http://counter.li.org.
4:48 p.m.
New subject: DD not working--SUCCESS!
Karl Larsen wrote:
>
> If after everything everybody has told you you still do not consider
> dd'ing a live system as a poor and risky practice then nobody will
> ever convince you.
>
I of course see what your saying. It would be best if you have a
computer dedicated to running dd. You plug in 2 hard drives and boot up
hoping bios takes care of the hard drives :-)
On the computer you say dd if=hd1 of=hd2 and neither Hard Drive is
doing anything else.
But I don't have that many computers and hard drives. So I do what I
can and it appears to have worked.
Just boot that computer with any live CD like Knoppix or your Fedora
install disk with 'linux rescue' at the boot prompt, and don't mount any
partitions from the hard disks involved in the copy. That way you are
sure that the directories and free space map don't change separately
from the data in the files during the time it takes to complete the copy.
--
Les Mikesell
lesmikesell(a)gmail.com
10:04 a.m.
New subject: DD not working--SUCCESS!
I neglected to mention. If your original drive has problems (i.e.
bad/failing sectors), you may want to look at an alternative such as
ddrescue (http://www.garloff.de/kurt/linux/ddrescue/). I have
successfully used ddrescue to recover a dead floppy on two different
occasions. Of course it can do the same for a hard drive.
Again, run from a live CD environment so find a live CD which has it
installed, or install it in the live environment after booting up
(either by getting it online, or downloading it to a flash drive first
and then booting to your live CD, mounting the thumb drive and
installing ddrescue from it).
Jacques B.
10:06 a.m.
New subject: DD not working--SUCCESS!
Just found out there is GNU ddrescue available at
http://www.gnu.org/software/ddrescue/ddrescue.html (I've never used
the GNU version but will certainly look at it when I get a chance and
probably start using it).
Jacques B.
10:22 a.m.
New subject: DD not working--SUCCESS!
Jacques B. wrote:
I neglected to mention. If your original drive has problems (i.e.
bad/failing sectors), you may want to look at an alternative such as
ddrescue (http://www.garloff.de/kurt/linux/ddrescue/). I have
successfully used ddrescue to recover a dead floppy on two different
occasions. Of course it can do the same for a hard drive.
Again, run from a live CD environment so find a live CD which has it
installed, or install it in the live environment after booting up
(either by getting it online, or downloading it to a flash drive first
and then booting to your live CD, mounting the thumb drive and
installing ddrescue from it).
Jacques B.
You can also get it by running yum install dd_rescue.
Mikkel
--
Do not meddle in the affairs of dragons,
for thou art crunchy and taste good with Ketchup!
6088
days inactive
6092
days old
86 comments
24 participants
participants (24)
-
Alan Cox -
Alan M. Evans -
Andy Green -
Arthur Pemberton -
Chris Jones -
Claude Jones -
Craig White -
Howard Howell -
Ian Malone -
Jacques B. -
Jim Cornette -
Karl Larsen -
Les Mikesell -
Manuel Aróstegui -
Mark C. Allman -
Matthew Flaschen -
Michael Hennebry -
Mike Chambers -
Mikkel -
Patrick -
Robert P. J. Day -
Steve Searle -
Tim -
Vivek J. Patankar