Happy with image backup :)
Mark Conrad
best seems to be working very well.
I have 105 backups to my credit so far, and near as I can determine the
105th backup is just as good as the 1st one.
All my applications work as well, all the configurations of my
applications are intact.
The _good_ version 2.1.1 of DiskWarrior never complains when I use it
to check out the directory of my OS 10.2.6 after a restore operation.
I am getting so lately I don't even bother checking OS ten after a
restore, because by now I have a lot of faith in the restore operation.
All my files are packed in butt-tight after any five minute restore of
the 3.06 GBs of files in my present smallish OSX implementation. No
fragmentation whatever of my files in the 12.00 GB partition I set
aside for OS-X. That means the "heads" of my disk drive do not have
to whip themselves to death picking up ump-teen pieces of a scattered
file.
Less wear and tear on my disk drive.
On the down side, OS X seems to get "confused" by my restore
instruction, so temporarily (for the last 105 backups) I am ignoring
the onscreen warning that appears immediately after the five minute
restore operation:
"Your disk was not put away properly, there may be damage to your files"
I have tried everything to try to get rid of that warning, but nothing
works, so I will pass it off as a false warning.
After all, nothing dire has happened after 105 restores.
Below are the two one-line commands I use for backup and restore, the
individual pieces of these one-line commands are seperated by
semicolons, which tells the computer to execute all the pieces one
immediately after the other.
To Backup-
dd if=/dev/rdisk0 of=/Volumes/Util-4/OSX-Backup-8-2/Bak bs=524288
count=1000;dd if=/dev/rdisk0 of=/Volumes/Util-4/OSX-Backup-8-2/Bak2
bs=524288 iseek=1000 count=1000;dd if=/dev/rdisk0
of=/Volumes/Util-4/OSX-Backup-8-2/Bak3 bs=524288 iseek=2000
count=1000;dd if=/dev/rdisk0 of=/Volumes/Util-4/OSX-Backup-8-2/Bak4
bs=524288 iseek=3000 count=1000;dd if=/dev/rdisk0
of=/Volumes/Util-4/OSX-Backup-8-2/Bak5 bs=524288 iseek=4000
count=1000;dd if=/dev/rdisk0 of=/Volumes/Util-4/OSX-Backup-8-2/Bak6
bs=524288 iseek=5000 count=1000;dd if=/dev/rdisk0
of=/Volumes/Util-4/OSX-Backup-8-2/Bak7 bs=524288 iseek=6000
count=1000;dd if=/dev/rdisk0 of=/Volumes/Util-4/OSX-Backup-8-2/Bak8
bs=524288 iseek=24574 count=5
To Restore-
dd if=/Volumes/Util-4/OSX-Backup-8-2/Bak of=/dev/rdisk0 bs=524288
count=1000;dd if=/Volumes/Util-4/OSX-Backup-8-2/Bak2 of=/dev/rdisk0
bs=524288 oseek=1000 count=1000;dd
if=/Volumes/Util-4/OSX-Backup-8-2/Bak3 of=/dev/rdisk0 bs=524288
oseek=2000 count=1000;dd if=/Volumes/Util-4/OSX-Backup-8-2/Bak4
of=/dev/rdisk0 bs=524288 oseek=3000 count=1000;dd
if=/Volumes/Util-4/OSX-Backup-8-2/Bak5 of=/dev/rdisk0 bs=524288
oseek=4000 count=1000;dd if=/Volumes/Util-4/OSX-Backup-8-2/Bak6
of=/dev/rdisk0 bs=524288 oseek=5000 count=1000;dd
if=/Volumes/Util-4/OSX-Backup-8-2/Bak7 of=/dev/rdisk0 bs=524288
oseek=6000 count=1000;dd if=/Volumes/Util-4/OSX-Backup-8-2/Bak8
of=/dev/rdisk0 bs=524288 oseek=24574 count=5
The one-line _backup_ command has never given me any problems, i.e.
the partitions of my internal drive never try to remount themselves.
The restore command is another story, all the partitions of my internal
drive remount themselves immediately after the five-minute command
finishes executing itself.
I temporarily tried the variation below to try to get rid of any
tendency of the partitions to remount themselves _during_ execution
of the restore command:
dd if=/Volumes/Util-4/OSX-Backup-8-2/Bak of=/dev/rdisk0 bs=524288
count=1000;umount -f /Volumes/OSX;umount -f /Volumes/Tiny;umount -f
/Volumes/OS9;umount -f /Volumes/Spare;umount -f /Volumes/Storage;dd
if=/Volumes/Util-4/OSX-Backup-8-2/Bak2 of=/dev/rdisk0 bs=524288
oseek=1000 count=1000;umount -f /Volumes/OSX;umount -f
/Volumes/Tiny;umount -f /Volumes/OS9;umount -f /Volumes/Spare;umount -f
/Volumes/Storage;dd if=/Volumes/Util-4/OSX-Backup-8-2/Bak3
of=/dev/rdisk0 bs=524288 oseek=2000 count=1000;umount -f
/Volumes/OSX;umount -f /Volumes/Tiny;umount -f /Volumes/OS9;umount -f
/Volumes/Spare;umount -f /Volumes/Storage;dd
if=/Volumes/Util-4/OSX-Backup-8-2/Bak4 of=/dev/rdisk0 bs=524288
oseek=3000 count=1000;umount -f /Volumes/OSX;umount -f
/Volumes/Tiny;umount -f /Volumes/OS9;umount -f /Volumes/Spare;umount -f
/Volumes/Storage;dd if=/Volumes/Util-4/OSX-Backup-8-2/Bak5
of=/dev/rdisk0 bs=524288 oseek=4000 count=1000;umount -f
/Volumes/OSX;umount -f /Volumes/Tiny;umount -f /Volumes/OS9;umount -f
/Volumes/Spare;umount -f /Volumes/Storage;dd
if=/Volumes/Util-4/OSX-Backup-8-2/Bak6 of=/dev/rdisk0 bs=524288
oseek=5000 count=1000;umount -f /Volumes/OSX;umount -f
/Volumes/Tiny;umount -f /Volumes/OS9;umount -f /Volumes/Spare;umount -f
/Volumes/Storage;dd if=/Volumes/Util-4/OSX-Backup-8-2/Bak7
of=/dev/rdisk0 bs=524288 oseek=6000 count=1000;umount -f
/Volumes/OSX;umount -f /Volumes/Tiny;umount -f /Volumes/OS9;umount -f
/Volumes/Spare;umount -f /Volumes/Storage;dd
if=/Volumes/Util-4/OSX-Backup-8-2/Bak8 of=/dev/rdisk0 bs=524288
oseek=24574 count=5
I got onscreen update messages that the partitions were _already_
unmounted, during execution of this temporary restore command.
The warning message about improper "put-away" still occured at the end,
so I gave up on the variation directly above.
I will just ignore the warning message.
Mark-
Mark Conrad
<nos...@iam.invalid> wrote:
> The warning message about improper "put-away" still occured at the end,
> so I gave up...
Anyone know how to de-activate the AutoDiskMount deamon for a period of
time, then re-activate it later?
...or maybe OS X will defeat such efforts by trying periodically to
re-activate the deamon, if it is shut off, even temporarily.
Mark-
to...@aplawrence.com
>For those like myself who can't afford a raid backup system, the next
>best seems to be working very well.
>I have 105 backups to my credit so far, and near as I can determine the
>105th backup is just as good as the 1st one.
>All my applications work as well, all the configurations of my
>applications are intact.
>The _good_ version 2.1.1 of DiskWarrior never complains when I use it
>to check out the directory of my OS 10.2.6 after a restore operation.
>I am getting so lately I don't even bother checking OS ten after a
>restore, because by now I have a lot of faith in the restore operation.
>All my files are packed in butt-tight after any five minute restore of
>the 3.06 GBs of files in my present smallish OSX implementation. No
>fragmentation whatever of my files in the 12.00 GB partition I set
>aside for OS-X. That means the "heads" of my disk drive do not have
>to whip themselves to death picking up ump-teen pieces of a scattered
>file.
>Less wear and tear on my disk drive.
>On the down side, OS X seems to get "confused" by my restore
>instruction, so temporarily (for the last 105 backups) I am ignoring
>the onscreen warning that appears immediately after the five minute
>restore operation:
>"Your disk was not put away properly, there may be damage to your files"
I'm impressed that you have had no issues, but this would scare the
heck out of me.
Your system is running while this is dd'ing in either direction. Maybe
*you* aren't doing anything, but background tasks are running, and I
would think it would be easy to get the disk into an inconsistent
state. Apparently that hasn't happened, but I would think that
sooner or later dd'ing mounted drives is going to bite you.
Were I going to use this, it would be in single-user mode only.
--
to...@aplawrence.com Unix/Linux/Mac OS X resources: http://aplawrence.com
Get paid for writing about tech: http://aplawrence.com/publish.html
Mark Conrad
> >On the down side, OS X seems to get "confused" by my restore
> >instruction, so temporarily (for the last 105 backups) I am ignoring
> >the onscreen warning that appears immediately after the five minute
> >restore operation:
>
> >"Your disk was not put away properly, there may be damage to your files"
>
>
> I'm impressed that you have had no issues, but this would scare the
> heck out of me.
>
> Your system is running while this is dd'ing in either direction. Maybe
> *you* aren't doing anything, but background tasks are running, and I
> would think it would be easy to get the disk into an inconsistent
> state. Apparently that hasn't happened, but I would think that
> sooner or later dd'ing mounted drives is going to bite you.
I am booted from my external disk, so no background tasks whatever are
running on the internal disk, which has all its partitions unmounted.
_All_ the disk activity, so far as "changing" anything once written to
the disk is concerned, happens on the external disk.
(during the entire restore operation, that is)
In other words, once I write to the internal disk from my backup file,
nothing comes along and changes what I initially wrote to the internal
disk.
The internal disk gets written to sequencially by big sectors (524288
byte sectors) - - - none of these big sectors get "re-written" to by
anything during the restore operation, so the state of the internal
disk never becomes inconsistant.
The only thing unusual are the five big sectors I write to at the very
end of the restore operation. Those five sectors "bracket in" the
spare volume directory at the very end of my 12 GB OS-X partition.
That spare directory consists of 187 (512-byte) sectors which I think
are used to check against the main directory by utilities such as
DiskWarrior, in order to ensure that the main directory has not been
corrupted.
"You did not put-away the disk properly, your files may be corrupted"
As far as that warning message is concerned, I believe it is a bogus
indication, because when I temporarily included forced "umount"
instructions within the restore command, Terminal kept complaining that
the partitions were already unmounted, all during the restore
operation.
I will verify this, if I can ever find out how to disable the
AutoDiskMount deamon for the duration of my restore operation.
Anyone know how to do this?
Mark-
(105 successful OS 10.2.6 "image" type disk restores successfully
completed so far, without any signs of corruption - - - average time to
restore 3.06 GBs of files onto a 12 GB partition is five minutes)
to...@aplawrence.com
>In article <bh3apd$i0a$1...@pcls4.std.com>, <to...@aplawrence.com> wrote:
>> >On the down side, OS X seems to get "confused" by my restore
>> >instruction, so temporarily (for the last 105 backups) I am ignoring
>> >the onscreen warning that appears immediately after the five minute
>> >restore operation:
>>
>> >"Your disk was not put away properly, there may be damage to your files"
>>
>>
>> I'm impressed that you have had no issues, but this would scare the
>> heck out of me.
>>
>> Your system is running while this is dd'ing in either direction. Maybe
>> *you* aren't doing anything, but background tasks are running, and I
>> would think it would be easy to get the disk into an inconsistent
>> state. Apparently that hasn't happened, but I would think that
>> sooner or later dd'ing mounted drives is going to bite you.
>I am booted from my external disk, so no background tasks whatever are
>running on the internal disk, which has all its partitions unmounted.
Oh, OK. I missed that. This is fine then.
Mark Conrad
> >I am booted from my external disk, so no background tasks whatever are
> >running on the internal disk, which has all its partitions unmounted.
>
>
> Oh, OK. I missed that. This is fine then.
You got it okay, I just forgot to mention all the relevant details.
I think I know why this warning message is thrown out.
"The disk was not put away properly, disk files may be corrupted"
Consider: OS X sees no internal disk partitions mounted on the desktop
during the restore operation, then all of a sudden the AutoDiskMount
deamon mounts all the partitions of the internal disk at the end of the
restore operation.
Writing the spare volume directory at the very end of the restore
operation seems to trigger the AutoDiskMount deamon.
Whatever flag in the bowels of the OS that keeps track of
mounts/unmounts never gets reset "properly", because the internal disk
was not mounted the "usual" way. (by a power-on or a restart)
More experimentation is called for. I will try restoring the spare
volume directory _first_ instead of last, then do a forced "umount"
in case the AutoDiskMount deamon gets triggered, then restore all the
rest of the sectors.
Perhaps that will get rid of the warning message.
Lotsa fun :)
Mark-
Mark Conrad
<nos...@iam.invalid> wrote:
> More experimentation is called for. I will try restoring the spare
> volume directory _first_ instead of last, then do a forced "umount"
> in case the AutoDiskMount deamon gets triggered, then restore all the
> rest of the sectors.
>
> Perhaps that will get rid of the warning message.
>
> Lotsa fun :)
I will _never_ understand this computer stuff!
The above plan of mine worked, for no ryme or reason.
The seperate warning-message window no longer comes up at the end of a
restore operation, at least for the seven additional restores that I
tried the new version on.
I won't feel like I have thrown that warning message until I get
another 40 or so additional restores under my belt.
To Restore- (newest version as of 8/10/03)
Remember, this is all one line, no carriage returns.
dd if=/Volumes/Util-4/OSX-Backup-8-2/Bak8 of=/dev/rdisk0 bs=524288
oseek=24574 count=5;umount -f /Volumes/OSX;umount -f
/Volumes/Tiny;umount -f /Volumes/OS9;umount -f /Volumes/Spare;umount -f
/Volumes/Storage;dd if=/Volumes/Util-4/OSX-Backup-8-2/Bak
of=/dev/rdisk0 bs=524288 count=1000;dd
if=/Volumes/Util-4/OSX-Backup-8-2/Bak2 of=/dev/rdisk0 bs=524288
oseek=1000 count=1000;dd if=/Volumes/Util-4/OSX-Backup-8-2/Bak3
of=/dev/rdisk0 bs=524288 oseek=2000 count=1000;dd
if=/Volumes/Util-4/OSX-Backup-8-2/Bak4 of=/dev/rdisk0 bs=524288
oseek=3000 count=1000;dd if=/Volumes/Util-4/OSX-Backup-8-2/Bak5
of=/dev/rdisk0 bs=524288 oseek=4000 count=1000;dd
if=/Volumes/Util-4/OSX-Backup-8-2/Bak6 of=/dev/rdisk0 bs=524288
oseek=5000 count=1000;dd if=/Volumes/Util-4/OSX-Backup-8-2/Bak7
of=/dev/rdisk0 bs=524288 oseek=6000 count=1000
Next I will try removing the "umount" commands above, to see if the
warning message still stays away.
Anyhow, all this arm-waving and jumping up and down does not seem to be
corrupting or damaging the restored OS X partition in any way, and it
still takes only five minutes to do a full image restore of OS X.
Mark-
(112 restores of OS 10.2.6 so far with no corruption that I can detect,
DiskWarrior version 2.1.1 is still happy)
Mark Conrad
<nos...@iam.invalid> wrote:
> Next I will try removing the "umount" commands above, to see if the
> warning message still stays away.
Yup, the warning message stays away with the "umount" lines removed.
This latest "good" version has not displayed the warning message yet:
(in ten runs of this latest version of restore below)
To Restore -
dd if=/Volumes/Util-4/OSX-Backup-8-2/Bak8 of=/dev/rdisk0 bs=524288
oseek=24574 count=5;dd if=/Volumes/Util-4/OSX-Backup-8-2/Bak
of=/dev/rdisk0 bs=524288 count=1000;dd
if=/Volumes/Util-4/OSX-Backup-8-2/Bak2 of=/dev/rdisk0 bs=524288
oseek=1000 count=1000;dd if=/Volumes/Util-4/OSX-Backup-8-2/Bak3
of=/dev/rdisk0 bs=524288 oseek=2000 count=1000;dd
if=/Volumes/Util-4/OSX-Backup-8-2/Bak4 of=/dev/rdisk0 bs=524288
oseek=3000 count=1000;dd if=/Volumes/Util-4/OSX-Backup-8-2/Bak5
of=/dev/rdisk0 bs=524288 oseek=4000 count=1000;dd
if=/Volumes/Util-4/OSX-Backup-8-2/Bak6 of=/dev/rdisk0 bs=524288
oseek=5000 count=1000;dd if=/Volumes/Util-4/OSX-Backup-8-2/Bak7
of=/dev/rdisk0 bs=524288 oseek=6000 count=1000
I have established that even the "bad" version of restore only displays
the warning message at the end of approx' _half_ of the restore runs,
so the problem with the warning message is definately intermittant in
nature.
Mark-
(119 restores of OS 10.2.6 so far, with no corruption that I can detect,
Mark Conrad
<eno...@idontcare.com> wrote:
> Yes, U R a GIANT intellect!
I am? Gee, thanks, and here I thought I was just bumbling along.
> Yes, U R a GIANT intellect! What a IDIOT!
Oh now, don't feel so bad about being wrong, you were bound to make a
mistake about me, given your mentality.
Mark-
Mark Conrad
<eno...@idontcare.com> wrote:
> Yeah, MORON, try a THIRD response!
You can't even count, this is only my 2nd response to you in this
thread.
Is brain damage prevelant in your family, or did it start with you.
Mark-
Vincent Vega
Mark Conrad <nos...@iam.invalid> wrote:
> In article <cCvZa.23924$qg3.1...@twister.tampabay.rr.com>, Enough
> <eno...@idontcare.com> wrote:
[who gives a shit]
Killfile, please.