[kris2fer]

Hi

I have a 250gb hdd which was created from the original 30 + 15hdd image.

I would like to now like to backup this drive saving season passes but not recordings. The commands i am using are-

PM fat32 hdd
PS empty
SM tivo drive
SS cdrom


The commands i am using are-

mkdir /mnt/dos
mount /dev/hda1 /mnt/dos
then

mfsbackup -l 32 -6so /mnt/dos/tivo.bak /dev/hdc

it then says

source drive size is 39 hours (which is the orignal drive)
-upgraded 283 hours ( which is the 250gb drive)
backup image will be 39 hours (THIS IS WHERE I THINK IT IS GOING WRONG IT SHOULD SAY "BACK UP IMAGE SHOULD BE 283 HOURS")

The image created is corrupt and wont restore to a new 400gb hdd

Is this correct
mfsbackup -l 32 -6so /mnt/dos/tivo.bak /dev/hdc



ps i can restore my original 40gb image to the new 400gb hdd but i would like to keep all my season passes etc from the 250gb hdd.


thanks kris2fer

 

[mikerr]

mfsbackup -l 32 -6so /mnt/dos/tivo.bak /dev/hdc

it then says

source drive size is 39 hours (which is the orignal drive)
-upgraded 283 hours ( which is the 250gb drive)
backup image will be 39 hours (THIS IS WHERE I THINK IT IS GOING WRONG IT SHOULD SAY "BACK UP IMAGE SHOULD BE 283 HOURS")

That's correct, the -s option (i.e. -6so ) shrinks the backup back to the original size:

-s ..... Shrinks the volume set.
This will divorce two drive installations, and will allow the backup to be restored to a drive of the same size as the original, or larger, and be re-expanded with the -x option of restore.

Your restore command should be

restore -x -r 4 -s 300 -zpi /mnt/dos/tivo.bak /dev/hdc

On the other hand, if you had them both connected at the same time,
then you could do this to copy on the fly (no disk image needed):

i.e. with old drive on primary master (hda)
CD on slave (hdb)
new drive on secondary master (hdc)

mfsbackup -l 32 -6so - /dev/hda | restore -x -r 4 -s 300 -zpi - /dev/hdc

don't forget to run copykern afterwards to initialise the swap.

 

[kris2fer]

I keep getting the same error

I have made a compressed and uncompressed backup using

mfsbackup -l 32 -6so /mnt/dos/tivo.bak /dev/hdc or
mfsbackup -l 32 -so /mnt/dos/tivo.bak /dev/hdc

and then restoreing using

restore -x -r 4 -s 300 -zpi /mnt/dos/tivo.bak /dev/hdc


The restore gets to 100% complete and then says

"RESTORE FAILED: ERROR RESTORING MFS DATA"

What am i doing wrong ??


kris2fer

 

[blindlemon]

Corruption on the source drive. Have you tried running mfsassert -please from the bash prompt?

 

[kris2fer]

The source drive is working fine in my other tivo, this wont corrupt it in any way will it ??

 

[blindlemon]

I've not seen mfsassert -please cause problems but of course YMMV

Just because drive works doesn't mean it is 100% perfect. I would be inclined to try mikerr's suggestion of doing a direct copy from one drive to another (but without compression) before trying mfsassert though...

 

[kris2fer]

mfsbackup -l 32 -6so - /dev/hda | restore -x -r 4 -s 300 -zpi - /dev/hdc

I'll give it a go

 

[blindlemon]

Leave out the 6

 

[kris2fer]

mfsbackup -l 32 -so - /dev/hda | restore -x -r 4 -s 300 -zpi - /dev/hdc
------------------- ^---------^--------------------------^-------

are these minus's needed between -so & /dev and -zpi & /dev

also i have tried 2 keyboards and i can get "|" in windows but not when booting up in linux i get <> instead

 

[blindlemon]

Yes, the '-' are the desination and source of the backup (stdout/stdin).

The | is shift+# or shift+= on my machines.

 

[ciper]

Why are you suggesting him to use "-s 300" for the restore? It does not match any of the conventional recommendations.

 

[blindlemon]

200 should be enough for a 400gb drive but I normally use 450 to allow for the addition of a 500gb "B" drive later if required

 

[ciper]

You will get a kick out of this from my S1

bash-2.02# cat /proc/meminfo
        total:    used:    free:  shared: buffers:  cached:
Mem:  14274560 14077952   196608 500805632    86016  3686400
Swap: 1471266816  8323072 1462943744

 

[kris2fer]

Still could not get it to work and did not want to risk any corruption using mfsassert, so i used the image from the original 40gb hdd and re-entered season passes.

All is working fine now. Thanks for the help


Kris2fer

 

[jrg]

and re-entered season passes.

Just a note for anyone else reading this later, as it's probably a bit late for you now!

There's a module for TivoWeb, called 'backup.itcl' - which you can find at http://www.boygenius.co.uk/files/tivoweb-backup-readme.htm - that will backup your Season Passes and some other stuff. Well worth a go, if you have TivoWeb installed.

 

[laurence]

Yes, the '-' are the desination and source of the backup (stdout/stdin).

The | is shift+# or shift+= on my machines.

Following the steveconrad guide, he says:

"If you have a twin drive machine - enter the following instead of the above command:

mfsbackup -l 32 -6so /mnt/dos/tivo.bak /dev/hdc /dev/hdb

(The l in the command is a lowercase "L" by the way.)"

So I've been typing a lower case L - is that wrong?

 

[Cainam]

I think you are mixing up 2 different commands. In the command

mfsbackup -l 32 -so - /dev/hda | restore -x -r 4 -s 300 -zpi - /dev/hdc


the one just to the left of "32" is a lower case" L". The one to the left of "restore" is the shift+# or shift+=

 

[laurence]

OK, I'm learning slowly! Thanks.