Unified mfs_* tools (originally tridge vplay)

[Jamie]

Was a merged or x86 version ever released? The latest release on sourceforge says "September 2006".

Nope, but the 64 bit source is in sourceforge, so can check it out and compile it yourself.

[mike_s]

Nope, but the 64 bit source is in sourceforge, so can check it out and compile it yourself.

You know that you're responding to a .sig spammer, don't you?

[1fatboy]

No. A stock S3 has a 32bit MFS file system. The kernel has nothing to do with it. Only the TiVoHD and Premiere have 64bit MFS.

I've got a Tivo HD with a corrupt volume header that does the GSOD reboot loop. I was hoping to extract some shows off the failed drive using PC extraction. Do I need to build the mfs 64bit sources on Linux x86 to extract the streams?

I've read where Spike said a corrupt volume header error by mfs_info means the super block header is missing. I'm not sure if I'm wasting my time or should just cut my losses.

[Jamie]

I've got a Tivo HD with a corrupt volume header that does the GSOD reboot loop. I was hoping to extract some shows off the failed drive using PC extraction. Do I need to build the mfs 64bit sources on Linux x86 to extract the streams?

I've read where Spike said a corrupt volume header error by mfs_info means the super block header is missing. I'm not sure if I'm wasting my time or should just cut my losses.

Are your shows still encrypted? If so, I'm not sure PC side extraction is really that useful to you.

If they are unencrypted, then, yes, getting the source and compiling yourself is the way to go.

[1fatboy]

Are your shows still encrypted? If so, I'm not sure PC side extraction is really that useful to you.

If they are unencrypted, then, yes, getting the source and compiling yourself is the way to go.

Yep, they're all unencrypted. I'll give it a try. Btw, I replaced the two disks with a single 2 Tb disk tonight. I'm running your custom kernel so thought it was a good opertunity to reduce chance of drive failure by 50% Its back up, now just to see how much can be recovered off the old disks. Thanks!

[1fatboy]

I was able to get some shows, mostly the newly recorded ones. Unfortunitually, a lot of the older recordings have atleast 1 or 2 parts of the TMF that have a fsid of size 0. On the upside the mfs_utils 64bit build worked like a champ and I've got another way to extract. Again, thanks.

[Jamie]

Glad you were able to recover something anyway.

I see that I have a local patch to export.c, that I think came from bcc [ref]. it relates to how the part size is determined. You could try it to see if it helps.

[1fatboy]

Glad you were able to recover something anyway.

I see that I have a local patch to export.c, that I think came from bcc [ref]. it relates to how the part size is determined. You could try it to see if it helps.

I got pretty excited to see the patch, but unfortunately it had the same result (size=0). I might add some debug to the new method tonight to see what the Ty master header contains. I feel like the data "should" still be there, but really wont know for sure until we can see if the structs contain anything meaningful.

It was really cool to see some shows extract off, but the bad part was they weren't the shows I wanted most. That's how it usually works out though.

[vandebo]

I wanted to learn the MFS format so I could investigate live streaming in progress recordings. I figured a good way to do that was to merge the 32bit and 64bit branches. It would be great if you could apply this to the sourceforge repro. Let me know if there are parts you'd like done a different way.

This patch is against the 64bit branch and merges all changes from head after the branch point. I tested the following commands using vserver on a TivoHD and a S1 Tivo: mfs_info mfs_ls mfs_dumpobj mfs_streams mfs_bitmap vplay

head-64bit_merge.patch.gz

--
Steve

[Jamie]

Cool! Nice work! I'll have a look and see if there is anything I'd suggest you change. Otherwise, if you have a sourceforge account, I can set you up as a developer on the project and you can commit the changes yourself. It will be nice to have the two branches unified again. That's a cleanup project I never seemed to have time for.

[Soapm]

I'm not getting the full Now Playing List when I launch TyTool9r18.zip on my PC. Would this be the thread to ask for help or should I use the tytool thread?

[Jamie]

This is a thread in the FILES forum, so it is not the appropriate place to ask questions, it is a place to post files.

If it is TyTools related, you should probably start asking there, although as far as I know, no one is currently supporting that project.

It does not look like a support thread was ever created for mfs-utils, so either start a thread in the Newbie forum, or in the Extraction, Insertion & Streaming Support forum, depending on how technical the question is. A basic "I can't get this to work" question should probably be in the Newbie forum.

[vandebo]

The two branches on sourceforge cvs have now been merged. Builds from 64 bit filesystems should be done from mainline. If you observer a confirmed regressions (works on the 64 bit branch or trunk before the merge, but not after the merge), report it here.

--
Steve

[Jamie]

Great. Thanks!

I'll post updated binaries soon. I probably can't provide Mac binaries anymore, unless someone else wants to build them.

[Soapm]

Nevermind, just found README.MFS_DEVICE

export MFS_DEVICE="/dev/sdc" is what I needed