TiVo Community Forums Archives - Series 2 60-hour upgrade problem

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

Series 2 60-hour upgrade problem
(Click here to view the original thread with full colors/images)

Posted by: wkirke

I've messed up my Series 2 60-hour machine, and my backup doesn't seem to work either.

I tried using Tiger's MFS tools 2.0 boot cd to backup & then add a 2nd harddrive to the 60-gig hard drive in my Series 2 60-hour TiVo, and it now it's stuck at "Welcome. Powering up...".

The steps I followed were:
1) I put the original 60-gig drive in my PC as secondary master, booted the MFS tools 2.0 CD, and followed the Hinsdale instructions to make a backup onto my C: drive (hda1 mounted at /dos).

2) I replaced the original 60-gig drive with a new 120 gig WD hard drive, restored the image to test the backup, installed the drive in the TiVo, and it booted just fine. System information said it was still 60 hours (as expected).

3) I then wanted to use the 120 gig drive as an expansion for the original 60-gig drive, so I put the original 60-gig as hdc, and the new 120-gig as hdb. I booted the MFS tools 2.0 cd, and ran mfsadd -x /dev/hdc /dev/hdb. I put the drives back in the TiVo, and it stuck on the "Welcome. Powering up..."

I've checked the jumpers, and the 60 gig drive is definetly Master (I've tried it at both the Master setting, and the cable select setting), and the 120 gig drive is definitely slave (tried it both at Slave, and at cable select).

I used WD's hard disk tools to verify that the 120 gig drive is ok. I've run the short and long tests several times with no problems.

4) I removed the 120 gig drive, and with just the 60-gig drive the Series 2 machine now repeatedly reboots after getting to "Welcome. Powering up..." (I guess it wants the 2nd hard drive -- is there any way to undo an mfsadd?)

5) I decided to restore my backup to the 120 gig drive and just live with a single drive machine, but it doesn't work (I swear it did when I first tested it). It now just goes to the "Welcome. Powering up..." screen and sits. (I tried the drive jumpers at both Master and cable-select.)

6) I picked up an 80 gig drive, and decided to try and restore the backup onto the 80 and 120 gig drive using this command:
mfstool restore -s 127 -zxpbi /dos/tivo2-60.bak /dev/hdc /dev/hdd

This restored 816 megs of backup, then adds /dev/hdc14-/dev/hdc15 for 81 hours (24 more), then adds dev/hdd2-/dev/hdd3 for 255 hours (144 more). However, this set now sits at the "Welcome. Powering up..." screen.

I've tried several variations of restores, using both one & two drives, with various combinatinos of not using the -x, not using -z, not using -p, not using -b (yes, there's lots of combinations here), and nothing seems to work. Most combinations hang at "Welcome. Powering up...", but some advance to "Almost There" before hanging.

7) I checked the partition tables using pdisk -l (dash ell), and it doesn't look right for anything restored from my backup image. Specifically, the original 60-hour drive has a 1 byte length for both primary & secondary bootstrap partitions, and a 4 meg length for both primary & secondary kernel partitions. Whereas the partitions set up by mfsrestore have Bootstrap 2 and Kernel 2 partitions that look like Series 1 machines (2 megs each). This concerns me because if the restored image were to work now, it might not survive a software upgrade. Perhaps this is just a problem with my backup image, and it showes up as an ancillary symptom of whatever really is the problem I'm having? Can anyone who has a working Series 2 drive made from an mfstools backup check it's partition table to see if the primary and secondary partitions are the same sizes? I'm worried that this could be a lurking problem for anyone running with a Series 2 drive built from a backup.

I'm running out of things to try to get my Series 2 machine working again. Any suggestions? I've looked for a 60-hour image, but the ftp site doesn't seem to have any Philips Series 2 images at all. Does anyone know if an 80-hour image would work on an 80-gig drive put into the 60-hour series 2 machine? Note, I have not tried restoring my backup onto the original 60-hour drive, as I'm trying to keep it is pristine as possible. (Yeah, I guess I should have set it on the shelf instead of doing mfsadd to it. But, I've upgraded so many series 1 machines that I guess I got bold/careless with my series 2 upgrade.)

Any help/suggestions you folks could give me would be greatly appreciated.

-Bill Kirke

Posted by: wkirke

I got a 60 hour image with 3.0, and have managed to restore it to my 120 gig drive, and my TiVo is finally working again.

Interestingly, the partition table was the same as I had noticed before -- bootstrap 1 = 1 byte, kernel 1 = 4 meg, while bootstrap 2 = 2 meg & kernel 2 = 2 meg. However, the drive worked, repeated guided setup just fine, and downloaded new program listings.

I let it sit overnight, and then forced a call, and it upgraded from 3.0 to 3.2. It seems to be working just fine, so apparently the partition problem does not affect the upgrade.

It took me 3 tries to get the image to restore and work. The first attempt was on an 80 gig Maxtor drive with the jumper set to Master. After restoring, I put it in the Tivo, and consistently got an over temperature error (I even removed the cover and put a large fan blowing lots of air accross the motherboard & power supply).

My second attempt was on my 120 gig WD drive set to Master. This just hung at "Welcome. Powering up..." (I gave up after 45 minutes.)

For my third attempt, I noticed that my PC bios wasn't recognizing the WD 120 set as master, so I changed the jumper to "cable select" (kept it still pluged into the same Secondary Master cable position), and the PC bios now could see it. I also executed "dd if=/dev/zero of=/dev/hdc bs=1024k" to zero the whole drive (took about an hour). Then, I did a restore via "mfsrestore -s 127 -bzpi /mnt/dev/tivofiles/tivo.bak /dev/hdc". This worked!

It's possible my entire 3-day problem was caused by the 120 gig drive not working in any position except cable-select. I had seen this once before with a western digital drive, but did not remember that until now.

Hope this helps anyone else.
-Bill Kirke