1 TB Tivo Issues

Discussion in 'TiVo Upgrade Center' started by jugbugs, Jun 30, 2006.

  1. jugbugs

    jugbugs New Member

    12
    0
    Jan 10, 2006

    Advertisements

    Okay I have spent about 3 hours reading around the forum to get caught up (as much as possible) since the last time I did an upgrade to this TiVo.

    Here's my history and what I want to do...

    I have a 40hr TCD540040 TiVo and I did my first upgrade from 40 to a 40 + 80 using weaknees upgrade tool and mfsbackup -Tao | mfsrestore -s 127 -r 4 -xzpi. I did this just to see if I could do it before I went all out (or quasi-all out).

    Next I did a 40 + 80 to 250 + 250 using the exact same method. No problems what so ever.

    Now I think I have a questionable drive and am ready for a replacement ( I think I want to replace both drives with either 250 + 250 or 500 + 500 )

    I have since attemped a backup | restore from 250 + 250 to 500 + 500 and am now getting GSOD followed by prompt reboots (rinse and repeat) and am at a loss as to where to go.

    I have heard ideas about increasing swap size above 127 using tpip.

    Will this solve my problem, or is there something else I should be doing?

    Any help is GREATLY appreciated!!
     
  2. JamieP

    JamieP Member

    991
    0
    Aug 2, 2004
    I would not expect a standarded size swap (127MB) to cause a problem like this. There's no reason you should get a GSOD on a fresh restore if everything was done right. There are good reasons to use a larger swap, but I don't think it is the root of your problem. See this for details.

    It really sounds like you didn't use -r 4, though I know you said you did.

    Can you post the result of mfsinfo, run on the PC right after you do the restore to the new disks?
     
  3. jugbugs

    jugbugs New Member

    12
    0
    Jan 10, 2006
    Wow, okay, great! There is some good info there that I wasn't able to find on my own. I just have a couple more conceptual questions about some of the switches.

    To do a dual drive backup there are two ways I have seen it done.

    1) # mfsbackup -f 9999 -1so /mnt/dos/tivo.bak /dev/hdc /dev/hdd

    and

    2) # mfsbackup -f 9999 -6so /mnt/dos/tivo.bak /dev/hdc /dev/hdd

    First, what does the -f 9999 do?

    and

    What does to -6so and -1so do? Difference?

    Moving on with the restore of the above image from a dual drive tivo. Can I go to any combination of drives that I want from here? Example, backup came from 250 + 250 GB tivo, could go back to just one 40 GB from here or any combination thereof?

    And the correct syntax is...

    # mfsrestore -s 127 -r 4 -zxpi /mnt/backup.bak /dev/hdZ /dev/hdZZ

    ?

    Thanks again...
     
  4. JamieP

    JamieP Member

    991
    0
    Aug 2, 2004
    This post has a good summary of the mfstools command line options. "-f" is used to leave some things (e.g. recordings) out of the backup. This is important if you want to make a small backup, or if you are trying to "shrink" the backup to drop upgrade partitions. It's definitely a hack. Many people believe using "-l" is safer (specifying the cutoff by size rather than by fsid #). How do you know whether the background animations you want to keep all have low fsid #s? In practice, -f 9999 seems to work fine on drives with stock tivo software images.

    -6 verses -1 controls the compression level. In the past, people have had problems with the higher compression levels, and Weaknees did a study indicating that -1 did almost as well as -6 in size, and did better in terms of runtime performance ([ref1], [ref2-Weaknees blog entry dated Feb 20,2004]).

    As long as you drop recordings and "shrink" in the backup phase, you can drop down to anything that is at least as large as the stock drive size on the box where the backup was made.
    I agree if you have a process that works, keep doing it. Ironically, that's the same argument many people use to keep using 127MB swap partitions. Many people still use 127MB swap partitions with large drives without problem. Upgrade vendors PTVUpgrade and Weaknees come to mind. I'm personally in the "better safe than sorry" school, so I always make large swap partitions on any drives I image. Too much swap space never hurt anyone, in my view, as long as you initialize it correctly.

    It could well be that the small swap prevented the GSOD from completing, but it still seems to me that the root cause of the problem is something else. A small swap shouldn't cause a GSOD, but may prevent the tivo from recovering from it.

    I didn't mean to assert that the OP didn't use "-r 4" in spite of saying he did. It was just pointing out that it would cause the same symptoms.
     
  5. mchips

    mchips TiVo Fanatic

    904
    0
    Feb 21, 2003
    Midwest

    Advertisements

    Looks like you must have started your reply just as I was removing my posts... I felt he may not have been listening to me, so I had decided to excuse myself from this discussion and let you guide him instead... :)

    I agree... that's why you hear so many "but it was working great until" a software update or a GSOD, and I've read and heard the theories as to why that is with a small swap file... I agree that there could have been another underlying issue that caused the GSOD, and then was unable to recover from it because of what options he did or didn't use during the backup/restore...

    I force a GSOD after my upgrades as a test...

    I agree here as well, which is why I didn't mention the "-r 4" since he included it in his command line, so I figured he must have used it, but perhaps he didn't... and I was then interested in his answer, because if he did, then that wasn't the cause of the GSOD... I know without it, it causes the same symptoms, which is why I use all three (-r 4, increased swap file + tpip)...

    It works, seems recommended by so many trustworthy people, so I use it, and haven't bothered with running any tests myself to see what happens if I vary from any of these settings... and therefore am just basing all of my assertions on this from what I've seen others report, and I now have your posts and additional links to add to that mix... :)
     
  6. jugbugs

    jugbugs New Member

    12
    0
    Jan 10, 2006
    Yeah, and I totally understand. It never hurts to second guess even yourself sometimes. It is quite possible that I left the -r 4 out even when I didn't want to. It wouldn't bo the first time that I have made a msitake.

    I am currently re-backing up (using -f 9999) again and then I going to restore using the -r 4 for sure and a swap size of 512. Then I am going to tpip and all should be well.

    If this doesn't work I am going to try this and you might be able to tell me if my logic is right here...

    I ahve tryed to just backup | restore to drives using the same size drives that are currently in there but it complains about size being too small. I am guessing it is just a little off so I will try this command and it should leave to unused space out...

    #mfsbackup -ao - /dev/hd1 /dev/hd2 | mfsrestore -s 256 -r 4 -xzpi - /dev/hd3 /dev/hd4

    My thinking is that leaving off the T will not preserve unused space but I still unsure as to what the o does above. Maybe you can tell me?

    Thank you so much for all your help... I found it very difficult to find definitive answers grepping through this forum.

    Although, all the needed information to do this stuff is (for the most part) in the sticky threads, these threads are over 50 pages long and that's alot of reading. Perhaps we should remake some stickys with the first post of which including VERY detailed and up-to-date information along these line? I might even be willing to help.

    Thanks again, I'll let ya know how it went.

    -Cheers-
     

Share This Page

spam firewall

Advertisements