On Sun, Apr 12, 1998 at 02:55:29PM -0700, George Bonser wrote: > > On Sun, Apr 12, 1998 at 04:09:50PM +0200, Marcus Brinkmann wrote: > > > I have now some bad blocks on my 2-year-old WD Caviar IDE drive. I'm not > > > overly concerned, because I have a brand new SeaGate, on where I install > > > hamm atm, but I wonder if linux can mark bad blacks as 'used', so that it > > > doesn't write on them anymore. Or how do you cope with bad blocks? > > > (I also get irq timeouts and drive resets, and then the system hangs. > > > Why?) > > Just a little note: > I installed one of these new DMA33/EIDE drives in a system that had been > running for over 6 months. At the same time, I moved from 2.0.29 kernel to > 2.0.33. The machine has not been up for more than two weeks since. I > started getting disk errors and IDE resets as soon as I installed that new > drive. Is that new Seagate that you mention one of the new large DMA33 > capable drives and did the problems with the WD start around the time you > installed the Seagate? Also which kernel are you running?
Interesting. The disk in our system is an ancient Quantum 1Gb, and actually I suspected the power supply might have been faulty before deploying the machine (but it took 15 months for anything to happen); occasionally I got disk resets even loading LILO. I suspect dry joints in the power supply (Quantums seem to be very power-sensitive, a friend of mine has had the same power supply problem with the same disk) but I haven't opened it to look yet. Hamish -- Hamish Moffatt, [EMAIL PROTECTED], [EMAIL PROTECTED], [EMAIL PROTECTED] Latest Debian packages at ftp://ftp.rising.com.au/pub/hamish. PGP#EFA6B9D5 CCs of replies from mailing lists are welcome. http://hamish.home.ml.org -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
