TiVo Community Forums Archives - Virgin Restore Green Screens

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

Virgin Restore Green Screens
(Click here to view the original thread with full colors/images)

Posted by: mithril

A couple of weeks ago the original drive in my Phillips HDR 212 failed and I got the GSOD. After 24 hours the drive was unable to repair itself so I removed it and replaced it with an 80G Maxtor that had been imaged from an HDR 312 2.0.1 backup. Later I added a second 80G Maxtor by TiVo Madding the two drives together, the system has since been upgraded to 3.001.

Now I want to restore my original configuration for the sake of canonality. I performed a low level reformat on my Quantum Fireball and have performed extensive tests doing writes and scans for 8 hours and the drive tests out fine.

Last night I restored a "virgin" 2.01 image that I got from the net. The image had been made with MFSTools 1 and I restored it with MFSTools 2.0. Everything looked pretty good except that I couldn't mount partition 7.

I installed the drive in my TiVo and went through the guided setup which went well, until it started processing the initial guide data, at that point it froze then crashed and started to reboot. Just after getting the grey "Almost There" screen it went GSOD on me. Since I was early in the procedure I pulled the disk, re-imaged it, performed an mkefs2 on partition 7 and tarred over a copy of everything in parition 4 and tried again. Once again after it started processing the guide data it GSODed again, this time I let it go the 2.5 hours and I could hear it thrashing the disk and download stuff from TiVo and it rebooted many times but was never able to correct itself. At that point it was getting late so I gave up and put my large drives back in.

I'm wondering if this has to do with the MFSTool 2.0 problem that I've read about and if I should restore with MFSTool 1 instead or if I should have been a little more patient. Also should I be concerned if one of the Linux ext2 filesystems is umountable? I've looked for a FAQ that describes how and why a TiVo is laid out and what should and should not be a concern but haven't found one. Is there a place I can find this stuff?

Thanks
Bob

Posted by: Robert S

You shouldn't expect both hdx4 and hdx7 to be mountable, one will contain the active root partition and the other will be empty, waiting to be formatted to take the next software upgrade. MFS Tools doesn't backup the inactive partition.

I don't know why you GSOD'd, but you can't recover from a GSOD if you've used MFS Tools 2.0. Try MFS Tools 1 - you may still GSOD, but you should recover.

Posted by: mithril

I tried again using mfstools 1.0 and it GSOD on me again. I decided to let it run all night to see what it would do and this morning I gleaned these errors out of the Otvdebuglog file:

Aug 8 23:36:53 (none) MFS[70]: Filesystem repair at 25.1
Aug 8 23:36:53 (none) MFS[70]: Filesystem sync at 25.2
Aug 8 23:36:53 (none) MFS[70]: Transaction log has been processed
Aug 8 23:39:41 (none) mfscheck[70]: Checking reference counts
Aug 8 23:39:41 (none) mfscheck[70]: ** WARNING: FIXING: file 2 refcount (changing from 4 to 2) object reasonable
Aug 8 23:39:41 (none) mfscheck[70]: ** WARNING: FIXING: file 5 refcount (changing from 4 to 2) object reasonable
Aug 8 23:39:41 (none) mfscheck[70]: ** WARNING: FIXING: file 6 refcount (changing from 4 to 2) object reasonable
Aug 8 23:39:41 (none) mfscheck[70]: ** WARNING: FIXING: file 7 refcount (changing from 4 to 2) object reasonable

It goes on like this for about 1200 lines but reports that everything has been corrected. After that I see errors like this:

Aug 8 23:44:37 (none) mfscheck[70]: ** ERROR: Object 49347 size mismatch: MFS
size = 652, obj size = 0
Aug 8 23:44:40 (none) MFS[70]: copy_bytes = 1024, actual = 0
Aug 8 23:44:41 (none) MFS[70]: FsFile::Get failed -- errFsIncompleteOperation

AND

Aug 8 23:45:34 (none) fsfix[69]: ** WARNING: can't read inode 10752 primary, blech!
Aug 8 23:45:34 (none) fsfix[69]: Rewriting primary inode page
Aug 8 23:45:34 (none) fsfix[69]: Writing node page 21504
Aug 8 23:45:41 (none) MFS[69]: copy_bytes = 524288, actual = 262144
Aug 8 23:45:41 (none) fsfix[69]: ** WARNING: can't read inode 20992 secondary, blech!

I then get a bunch of these:

Aug 8 23:49:21 (none) fsfix[69]: ** WARNING: allocated extent 956278 size 8 is marked free, file may be corrupt
Aug 8 23:49:22 (none) fsfix[69]: ** WARNING: allocated extent 987878 size 8 is marked free, file may be corrupt
Aug 8 23:49:22 (none) fsfix[69]: ** WARNING: allocated extent 919494 size 8 is marked free, file may be corrupt
Aug 8 23:49:22 (none) fsfix[69]: ** WARNING: allocated extent 988886 size 8 is marked free, file may be corrupt
Aug 8 23:49:22 (none) fsfix[69]: ** WARNING: allocated extent 997854 size 8 is marked free, file may be corrupt
Aug 8 23:49:22 (none) fsfix[69]: ** WARNING: allocated extent 792950 size 8 is marked free, file may be corrupt
Aug 8 23:49:23 (none) fsfix[69]: ** WARNING: allocated extent 920086 size 8 is marked free, file may be corrupt

I then get a few of these:

Aug 8 23:55:06 (none) mfscheck[70]: Destroying file 49347
Aug 8 23:55:06 (none) mfscheck[70]: RefCount::DestroyFile 49347 errFsNoLinks
Aug 8 23:55:06 (none) mfscheck[70]: Destroying file 49351
Aug 8 23:55:06 (none) mfscheck[70]: RefCount::DestroyFile 49351 errFsNoLinks

Followed by a few more file mismatches.

The rest of the file contains these:

Aug 8 23:55:06 (none) mfscheck[70]: Link 45876 /Server/1262
(errOk)
Aug 8 23:55:06 (none) mfscheck[70]: Link 49153 /Server/218375
(errOk)
Aug 8 23:55:06 (none) mfscheck[70]: Link 45877 /Server/1387
(errOk)
Aug 8 23:55:06 (none) mfscheck[70]: Link 2 /ObjectType/Actor
(errOk)
Aug 8 23:55:06 (none) mfscheck[70]: Link 2 /Server/38598
(errOk)
Aug 8 23:55:07 (none) mfscheck[70]: Link 49154 /Server/218615
(errOk)
Aug 8 23:55:07 (none) mfscheck[70]: Link 45878 /Server/1408
(errOk)
Aug 8 23:55:07 (none) mfscheck[70]: Link 49155 /Server/35719
(errOk)
Aug 8 23:55:07 (none) mfscheck[70]: Link 45879 /Server/1424
(errOk)

I have no idea what this is trying to tell me. Can anyone throw me a clue?

Is it possible that despite a low level reformat and hours of read/write tests my drive is junk?

Thanks
-Bob

Posted by: Otto

It mainly says that you have a bad image. :p

Posted by: mithril

Well, Sunday I bought a new hard drive, restored the image and viola, everything worked fine. So this leads me to believe that the Quantum Fireball that came in my TiVo has failed on some fundemental level. I did notice that the drive was locked and reporting on 9MB of space which seems like a plausible reason why I might get errors of file size truncation. I have the drive running in my compter as my C: drive and for that app it seems fine, though when I try to run MaxBlaster I get a firmware error and it quits. MaxBlaster will work if the drive is anything but C:

So, that as they say is that.

-Bob