RHL 9 bad block check fails

From: Robert M. Riches Jr (spamtrap42_at_verizon.net)
Date: 01/06/04

  • Next message: Arif Ali: "Re: i'm not able 2 configure soundcard in RH8.0.." 
    Date: Tue, 06 Jan 2004 05:23:13 GMT

    Is there any evidence that the bad block check during RHL 9
    install is excessively pessimistic or reports problems when
    it shouldn't?

    While installing RHL 9, I enabled the bad block checks as I
    have for several installations on other machines. This is
    the first time I have seen a report of any bad blocks. The
    dialog box said the partition had bad blocks and installing
    on it was not recommended. This was the root and main
    partition.

    This machine has functioned flawlessly with RHL 8.0 for most
    of a year (except for a quirk shared by its twin with 'rpm
    -V <pkgname>' and the known rpm hang software bug). The
    disk is a Western Digital 60GB. The motherboard is an Iwill
    P4ES with the Intel 845E chipset, the CPU is a 1.8GHz
    Pentium 4 processor, and RAM is 512MB with ECC enabled. The
    only PCI cards are a Radeon 7500 PCI graphics card and the
    NIC. The power supply is 300 or 350W, which I would think
    ought to be adequate being as the identical power supply
    runs another machine with twice the RAM, 2.4GHz CPU, two
    hard disks, a CD ROM drive, a CD burner, and a few more PCI
    cards.

    One disk-related quirk is just to be different I left one
    cylinder of free space between each partition--it's a result
    of mild obsessive-compulsive and some experience with
    guardbanding.

    The other quirk is under RHL 8.0 this machine and its twin
    would sometimes report segmentation faults when running 'rpm
    -V <pkgname>' from a script on all ~1400 installed packages.
    All core files showed the program bombing at the same line
    of code with essentially the same stack trace every time. I
    reported a bug against rpm, but the responder wasn't
    inclined to want to try to fix the software.

    I ran into another round of 'rpm -V' segmentation
    violations, but several iterations of 'rpm --rebuilddb' seem
    to have fixed that. There are nothing relevant in
    /var/log/messages or the output of 'dmesg' about disk I/O
    problems.

    So, does anyone have experience with the bad block check (if
    anyone else runs it) being too pessimistic?

    Thanks.

    Robert Riches
    spamtrap42@verizon.net
    (Yes, that is one of my email addresses.)

  • Next message: Arif Ali: "Re: i'm not able 2 configure soundcard in RH8.0.."

    Relevant Pages