Imminent HD failiure :-0

Discussion in 'TiVo Series 1 - UK' started by Neil F, Jan 6, 2006.

  1. Jan 6, 2006 #1 of 12
    Neil F

    Neil F New Member

    33
    0
    Feb 27, 2002
    United Kingdom

    Advertisements

    Hi everyone,

    I was hoping for a bit of advice from the more experienced TIVO hackers on here.

    I had a hard drive failure in about September so replaced the failed B drive (original 30gig A drive) with a 250gig. I made the 120gig which was my A drive the B in the new set up.

    Anyway, I come to watch CBB4 launch last night and blank screen on the TIVO and no response to the peanut.

    After a few reboots I got it to come back to life and upon checking the kernal log found this entry occured a few times:-

    Jan 5 10:40:05 (none) kernel: Stopping immediate on Ide
    Jan 5 10:40:05 (none) kernel: IDE: tivo DMA engine aborted
    Jan 5 10:40:05 (none) kernel: hda: irq timeout: status=0xd0 { Busy }
    Jan 5 10:40:05 (none) kernel: hda: irq timeout: error=0xd0 { BadSector UncorrectableError SectorIdNotFound }, secCnt=208, LBAsect=13684944
    Jan 5 10:40:05 (none) kernel: hda: disabled DMA
    Jan 5 10:40:05 (none) kernel: hda: ide-tivo re-enabled DMA
    Jan 5 10:40:05 (none) kernel: hdb: disabled DMA
    Jan 5 10:40:05 (none) kernel: hdb: ide-tivo re-enabled DMA
    Jan 5 10:40:06 (none) kernel: ide0: reset: success

    So it looks like I have a bad sector?? I take it hda will be the master drive in the TIVO?

    The few times it occured, it always said LBAsect=13684944. Will Tivo map out this bad sector?

    I have just backed up my season passes and tarballed my /var/hack/ as a 1st precaution but things look bad :( The 'clouds' animation that plays in the background of the menus now has a pixellation glitch in it that repeats every few secs.

    Anyone any suggestions on how to fix this most painlessly? It's took me ages to get all the hacks running since I rebuilt a few months ago and I don't know If I could face doing it all again so soon hehe

    thanks
    Neil
     
  2. Jan 6, 2006 #2 of 12
    Glesgie

    Glesgie New Member

    194
    0
    Feb 3, 2003
    London
    Yes hda is the Master drive.

    A copy of the drive using ddrescue is your best hope for retaining your data and recordings.
     
  3. Jan 6, 2006 #3 of 12
    Neil F

    Neil F New Member

    33
    0
    Feb 27, 2002
    United Kingdom
    Hiya

    Thanks for the reply.

    I take it that backup would be the same size as the original though?

    I have 250gig + 120gig drives, So I take it I would need a 370gig drive to put the image on? Or does it copy direct from old A drive to new A drive and I could leave my B drive as is?

    To be honest, losing my recordings wouldn't be the end of the world for me. I'm more worried about losing all the hours I've put into getting mode├ś, turbonet, tivowebplus, hackman, endpad etc etc etc working nice :(

    ta
    Neil
     
  4. Jan 6, 2006 #4 of 12
    blindlemon

    blindlemon tivoheaven.co.uk

    5,577
    0
    May 12, 2002
    Malmesbury, UK
    With just one bad sector there's a good chance you will be able to make a minimal MFSTools backup of the system - so just preserving your settings and hacks etc.

    If I didn't want the recordings I would try that first as it only takes 10 minutes if it works :)

    If you have to use dd_rescue then you can leave the "B" drive untouched. However, if the corruption has damaged your system files or the MFS then you may still end up with a non-working system.
     
  5. Jan 6, 2006 #5 of 12
    Neil F

    Neil F New Member

    33
    0
    Feb 27, 2002
    United Kingdom

    Advertisements

    Thank you for that info.

    I'm just looking at the hard drive selection i can buy locally. I notice thread in the past have said you can't replace manufacturer 'A' with 'B' because their drive is slightly smaller. With this in mind could you suggest a good replacement from these as I have no idea how to work that out and wouldnt like to buy one thats smaller than what I have.

    the failing drive I have in is a maxtor 250gig 7L250S0. this has a 5 year warranty so it will be going back!

    Drives in stock that I can get locally:-

    250 GB Samsung SP2514N 7200rpm 8Mb Cache IDE ATA133 (3yr Warranty)
    Manufacturers Code: SP2514N

    250 GB Western Digital Caviar 8MB Cache 7200rpm ATA100 Fluid Bearings (JB)
    Manufacturers Code: WD2500JB

    250 GB Seagate Barracuda® 8MB Cache 7200rpm ATA100 (Manufacturer's 5yr Warranty)
    Manufacturers Code: ST3250823A

    250 GB Maxtor DiamondMax 10 lead free 7200rpm 16MB ATA133 Cache
    Manufacturers Code: 6L250R0

    I don't suppose you have any idea which is big enough to allow me to restore my image? Or is there a way round restoring to a smaller drive these days?

    Many thanks
    Neil
     
  6. Jan 6, 2006 #6 of 12
    blindlemon

    blindlemon tivoheaven.co.uk

    5,577
    0
    May 12, 2002
    Malmesbury, UK
    Unfortunately, the only drive likely to be as big as your failed Maxtor is the other Maxtor :(

    The drive I would recommend out of the above would be the Seagate, but that's definitely smaller. However, if you can get a divorced minimal backup (-l32 -so) then you should be able to restore that to any drive >= 30GB.
     
  7. Jan 6, 2006 #7 of 12
    Neil F

    Neil F New Member

    33
    0
    Feb 27, 2002
    United Kingdom
    Thanks again blindlemon. Your a star!

    I'll go with the seagate then and hope i can get a 'divorced minimal backup'! I don't fancy getting another maxtor...3 months out of a brand new drive and it was meant to be industrial strength for 24/7 use!

    I may be back with more questions about divirced backups later hehe

    Once again your help is much appreciated - Thanks

    Neil
     
  8. Jan 6, 2006 #8 of 12
    Neil F

    Neil F New Member

    33
    0
    Feb 27, 2002
    United Kingdom
    Hi

    I think i need a bit more advice now. I have connected the drives up to my PC, run mfstool 2 CD and created a backup but it says:-

    Source drive is 30 hours
    - upgraded to 283 hours
    -upgraded to 427 hours

    Backup image will be 427 hours

    I would have expected the image to be 30 hours for it to fit on my new disk? Is that correct?
    I tried the mfsbackup command with:-

    mfsbackup -l 32 -so /mnt/dos/tivo.bak /dev/hdc /dev/hdb
    and
    mfsbackup -l 32 -6so /mnt/dos/tivo.bak /dev/hdc /dev/hdb

    but both said 427 hours. What is the relevance of the '6'? Anyone any suggestions before I revert back tp my older image with no hacks etc :(
     
  9. Jan 6, 2006 #9 of 12
    blindlemon

    blindlemon tivoheaven.co.uk

    5,577
    0
    May 12, 2002
    Malmesbury, UK
    Your backup has failed to divorce due to small recordings scattered over various data partitions.

    You need to put the drive back in your TiVo, telnet in and run my DeleteAll script from the bash prompt to clear out all the recordings. Then when you retry the backup it should divorce cleanly (discarding the additional partitions) and give you a 30 hour image.

    The 6 is the compression factor for the backup if you use the built-in compression. However, it's buggy, which is why people sometimes get a "compression error" when trying to restore a vital backup - so I now always leave it out. An uncompressed backup will be about 1.4GB in size (compared to 350mb for a compressed one) but can be zipped down to around 350mb if you need to - and Winzip doesn't suffer from compression errors, AFAIK :)
     
  10. Neil F

    Neil F New Member

    33
    0
    Feb 27, 2002
    United Kingdom
    Fantastic! I will get right onto that :)

    Thanks for explaining the '6' too. I have an old 5gig FAT drive I'm using, so size doesnt matter (so they say lol)

    Thanks again and I'll let you know how I get on..
    Neil
     
  11. Neil F

    Neil F New Member

    33
    0
    Feb 27, 2002
    United Kingdom
    Thanks a million Blindlemon :D

    All went well and it's up and running. One problem, I'm still left with the glitchy graphics in the menus. I think its either the clouds anim but its hard to tell exactly because its at the top near the title of the menu.
    I've read about this 'kickstart' that reinstalls the operating system and wonder if that would be a good option. I might give it a try. at least I have a good 30hour backup if it all goes nasty! :)

    Neil
     
  12. Neil F

    Neil F New Member

    33
    0
    Feb 27, 2002
    United Kingdom
    Ok I tried the following and i'm now stuck in a reboot loop..I think i better face the music and restore from my clean image and use the tarball of /var/hack that i have :rolleyes:

    bash-2.02# mfscheck
    EventSwitcher stopped
    mfscheck scan begins
    Segmentation fault
    bash-2.02# mfsassert -please
    Filesystem assert: false && "mfsassert" at fsassert_main.C line 37 in int fsasse
    rt_main(int, char **)
    Filesystem flagged as inconsistent!
    Tmk Assertion Failure: false && "mfsassert"
    int fsassert_main(int, char **), line 37 (fsassert_main.C)
    Tmk Fatal Error: Thread mfsassert <177> died due to signal -2
    1a344ec 1a32a44 1a1a604 1a160f8 1c28cd4 1800134
     

Share This Page

spam firewall

Advertisements