TiVo Community Forums Archives - GSOD reboot cycle: bad B drive, got zonemap assert/corrupt. How restore recordings?

TiVoCommunity.com
(c)opyright 1995-2005 All rights reserved
indexcheckTC
This area is a static history of posts in the TiVo Community Forum Archive.
This archive history was made for the simple indexing of search sites like Google.

Pages:1

GSOD reboot cycle: bad B drive, got zonemap assert/corrupt. How restore recordings?
(Click here to view the original thread with full colors/images)

Posted by: cgould

My Sony SV2000 3.0 standalone (expanded w/ 2nd 40g maxtor drive in Aug) had a stuttering/hanging problem for a few months, then recently got into a GSOD/reboot cycle.

I was able to restore OK my old pre-upgrade 30g image to a new WD80g drive, so the Tivo's back online at least- but that won't get back any of my recordings.

Can anyone advise a path to get them back?
or am I stuck?

Problem history/solution attemp details:

It turns out the added "B" Maxtor drive died (sent for RMA replacement last month of warranty, phew!.)
I tried doing mfsbackup, and the compressed image BU'd ok w/ no errors, 816mb in size.
I then tried doing mfsbackup -Tao/mfsrestore of both A/B drives to the new drive, to try and keep my old recordings, and it got a bunch of I/O errors in the middle, but otherwise went OK.

However, on booting w/ the new drive, I got the same GSOD reboot cycle. I checked the tivo log files w/ bash login, I see stuff like:

tverr log file:

Dec 17 19:03:45 (none) FsVolume[74]: Assert failed: mapM.iPageMapM == iPageMap at fszone.C line 115 in FsZone::FsZone(bool, enum FsZoneOwner, )
Dec 17 19:03:47 (none) TmkAssertionFailure[74]: mapM.iPageMapM == iPageMap (FsZone::FsZone(), line 115 (fszone.C))
Dec 17 19:03:47 (none) fsfix[74]: Tmk Fatal Error: Thread fsfix <74> died due to signal -2
Dec 17 19:03:47 (none) fsfix[74]: Tmk Thread Backtrace: 1aa90ac 1aa79c8 1a962a8 1bb47b8 1bb3ab0 1d7fc04 1d80198 1d801f8 1d7ea10 1bb20f8 1a7d80c 1ce610c 1800134

kernel log file:

Dec 27 16:31:01 (none) kernel: Filesystem assert: mapM.iPageMapM == iPageMap at fszone.C line 115 in FsZone::FsZone()
Dec 27 16:31:02 (none) kernel: Filesystem flagged as inconsistent!
Dec 27 16:31:05 (none) kernel: Tmk Assertion Failure: mapM.iPageMapM == iPageMap
Dec 27 16:31:05 (none) kernel: FsZone::FsZone(), line 115 (fszone.C)
Dec 27 16:31:05 (none) kernel: Tmk Fatal Error: Thread fsfix <74> died due to signal -2

I only have a 70g (30g+40g) tivo, so I don't think it's the normal swap problem. I used -s127 on the first expansion, and I didn't see any swap errors in log files.

I then tried restoring only the "divorced" compressed-from-A&B backup (no recordings), onto the 80g drive, and got an error from mfsrestore:
"zone map corrupted, loading backup, failed- aborting"

So, it seems like something outside the recordings itself, is corrupted. (mfs file system?)

I haven't tried a dd w/ noconv/noerr from old B to newyet, but it seems like something else is bad since even the "divorced" backup failed...

are things totally gone?
Any way for me to recover my old recordings?

In the meantime, I've bought a UPS for the Tivo, since I think many recent GSOD/stuttering problems in the Bay Area could be related to storm-caused bad power.

Thanks for any help!
Colin

Posted by: cgould

PS, a couple followup questions to this, though maybe more hack rather than upgrade:

1. how can I start my tivo w/o the "myworld" program, eg to try an fsfix on its own? (I have a bash prompt installed)

2. Has anyone gotten the elseed program (callerID display) to work on a Sony SV2000 standalone? tivo-only, not on any network...

Thanks!

Posted by: bsnelson

Put "runmyworld=false" in your boot parms, and this will cause myworld to not be started.

Having said that, I fear that the recordings are toast.

Brad