bad block on hda - am i hosed?

[ruwach]

So i am going along here fine, right ? running the computer and something crashed.
i did a reboot and the OS would only partially load and then hang with a kernel panic.

i thought " no problem, i have re-installed my OS a dozen times, here we go again. "

tried several distros of linux, but could not get past the part of install where a partition is formatted. It always hangs.
i installed the HD as a slave in another box. i had three partitions set up on this drive. / , /swap, and /stuff (where i keep stuff).

So, i tried to mkfs ext3 /dev/hdb1 and it errored out with bad blocks....

i can mount the other partition (/stuff) just fine, and all the data there seems to be intact. But what do i do to be able to use that partition again. the drive is only 6 months old.

any suggestions ?

thanks

[Morphine Drip]

try running the 'badblocks' command. man badblocks for cli info

good luck.

[ruwach]

thanks mr. drip. i checked the man page and was advised to use bad blocks as a -c option to mkfs.
so that is what i did. seems ok now.
many thanks.
now another question.
i want to install an operating system on this partition. when the linux installer tries to format the disk, will it pick up and correct for the bad blocks?
thanks

[Morphine Drip]

most welcome. once the badblocks app has flagged/found the bad blocks the OS will not attempt to write to those blocks again.

[Satanic Atheist]

Just a quick note on dodgy drives after a forced reboot without a shutdown...

I have 2 hard drives in my main machine, one is a 30Gb boot drive that is permanently installed into the computer and is used purely for booting and starting the OS (Slackware 10). The second drive is a removable 120Gb drive so I can just rip it out and take it away if I go anywhere. This is the drive that has the problem - due to heat dissipation problems, the IDE connector works its way loose (I have a large tower case with only one case fan, and no fan on the CPU, just a massive heatsink). When the cable works its way loose, I have to forcibly shutdown the computer (pull the plug) and reconfirm the solid IDE connection to restart the machine.

Thankfully, to date, I have not lost any data (but see previous posts about near-misses!) and have to regularly start up in mode 3 (no X) to test the drive with e2fsck several times to fix the damage. Often, I have to use the -f function to override the apparent OKness report from the journal. Eventually, the drive is safe to remount again.

I love removable technology, but sometimes it does get screwy...

Also, my original 30Gb (Quantum Fireball) died when something went pear-shaped and Linux reported Read_Drive_Not_Ready errors under /var/log/messages. Note, SMART was enabled and I have NEVER received a SMART report to indicate imminent drive failure.

Now, I''m about to slap the 120Gb into an IcyBox USB2.0 so hopefully I won't have trouble with the IDE connector any more.

James

[blobaugh]

SA, ever think about just using a live boot cd and running fsck? I've had my laptop drive freak out a couple times on me and that solved the problem. Had to run fsck a few times

[Satanic Atheist]

Yeah, I use Knoppix, Slackware-Live and DSL to fsck the drives. This was just one VERY bad drive (dual-boot Windows and Linux - both showed errors).

I eventually sold it on eBay to someone who badly wanted the chipset off it.

James