Recent changes to bugs

#3 The decision on what streams to backup is very dumb

jkozee — Sat, 13 Jun 2015 03:37:34 -0000

status: open --> closed
Group: --> Release 2.0

mfstool restore -h does not cover -r option

Anonymous — Tue, 11 Nov 2008 01:36:18 -0000

In version (mfstools-2.0), the help output for
$ mfstool restore -h
does not include a line for the "-r" option

connors@hpl.hp.com

Swap > 127 not initialized

Anonymous — Fri, 19 Aug 2005 21:26:06 -0000

The "-s" option on the restore allows you to specify
the new swap size. This is necesaru for larger drive
support. When the size specified is greater than 127Mb,
the swap space is not initialized properly, but no
errors are generated. The swap space IS allocated, but
the header is wrong.

When the Tivo is started it fails to boot because the
swap space won't initialize. Workaround is to run
another software called "tpip" with the following flags

tpip v1.1 : tpip -s --swapped /dev/<drive>
tpip v1.2 : tpip -s -1 /dev/<drive>

tpip seems to write out the right header info so that
swap is properly initialized. Tested with "-s 192"
worked fine. Other sizes reported working as well.

segv in mfstool restore

Wayne Feick — Fri, 13 May 2005 02:36:53 -0000

At restore.c:901, you need to change

if (*dev2)

if (dev2 && *dev2)

Large Drive Support

Anonymous — Mon, 07 Feb 2005 01:25:22 -0000

It appears MFSTools only supports a partition size of
274gb. With 300gb, 400gb and larger drives out now
this really needs to be addressed. If anyone has a quick
fix to do a single drive 400gb upgrade using the full
capacity I'd like to hear it.

backup -o does not support large files

John — Mon, 01 Nov 2004 02:51:21 -0000

when I do a backup -9ao foo /dev/hdb, it craps out at
the 2GB mark reporting both a failure and a success.

I'm using the bootable CD of MFS 2.0.

Cannot mount partition on restored disk

Anonymous — Tue, 17 Aug 2004 13:12:35 -0000

Source: 80GB Maxtor from Hitachi SD-H400 DVD/Tivo
Destination: 250GB Hitachi

Restore appears to succeed and I can boot the
destination disk in Tivo, but I cannot mount partition 7
of the restored disk. Partition 7 of the source disk is
mountable and both disk's partitions are recognized as
the same at boot time. pdisk reports some differences in
the sizes of the destination disk's partitions, but both
disk's partition 7s are reported as Ext2 Root 2.

I have tried this with both a disk to disk copy (with and
without expansion) and restore from a compressed (no
streams) backup. Used both sleeper iso and mfstools2
iso from pvrfiles.com

Can't backup on Mac OS X

Anonymous — Mon, 09 Aug 2004 20:00:43 -0000

I checked out and built the 'mac' cvs tag for the tools.

I can't actually back up my Tivo drive.

So I see in System Profiler that my orig drive is mounted as disk0

I thought I could do this:
% sudo ./mfstool backup -6so /Volumes/Hallucinatory/tivobackup/
tivo.bak
/dev/disk0
Password:
/dev/disk0s10: Unknown error: 0
mfs_load_volume_header: mfsvol_read_data: Input/output error

Ok, I was getting permissions errors on /dev/disk0 and /dev/
disk0s<something>
that the tool was adding onto my /dev/disk0 I gave as an
argument..

I wasn't able to get it to run as root under Xcode, so I chowned
/dev/disk0 back to me.. which helped it get farther.. I'm possibly
doing the
wrong tactic at all so maybe this debugging stuff isn't useful..

I got down into tivo_read_partition_table

and here:
/* Find out what the magic is in the bootblock. */
switch (htons (*(unsigned short *) buf))
{
case TIVO_BOOT_MAGIC:
/* It is the right magic. Do nothing. */
break;
case TIVO_BOOT_AMIGC:
/* It is the right magic, but it is byte-swapped. Enable byte-
swapping. */
table->vol_flags |= VOL_SWAB;
break;
default:
/* Wrong magic. Bail. */
close (*fd);
free (table);
return 0;
}

It's hitting the TIVO_BOOT_AMIGC case. I even tried commenting
out that line
but it didn't make a difference in the following..

So it goes a few more lines and fails here:
/* If it doesn't have the magic, it's not hip. No more partitions. */
if (htons (part->signature) !=
MAC_PARTITION_MAGIC)
{
break;
}

and then thinks it has no partitions and bails out.

Restore problems to a larger drive.

Anonymous — Thu, 08 Jul 2004 00:53:54 -0000

I noticed that when I tried to do a Restore (xpi) to a
larger drive that I would get a report that the
destination is not large enough. This occurred with
particular models of Maxtor drives. Particularly the
DiamondMax Plus 9 120GB ATA133 drives. The same
thing happened with the 160GB model of the same
drive. The source drive was 100GB Seagate Barracuda.
Is there any way to override this error. I am sure it is
just a misreporting.

signalguru@yahoo.com

mfsinfo segfaults

Stephen M. Moraco — Thu, 17 Jun 2004 00:38:20 -0000

run mfsinfo against two drive pair, nodma kernel both
drives hdx=bswap. In my case I'm seeing Zone Map
primary and secondary corrupt and checksum bad
messages after which i get the segfault.

I figure the segfault part is a bug. I figure that the
zone map stuff is the not-so-good state of my drives :(