install just hangs... - Slackware

This is a discussion on install just hangs... - Slackware ; On 20 Nov 2007 23:38:11 GMT, Joost Kremers wrote: > Joost Kremers wrote: >> still, i'm hoping some of you may have a few more trouble-shooting tips to >> offer, a few things i can try to get slack 12 ...

+ Reply to Thread
Page 2 of 2 FirstFirst 1 2
Results 21 to 35 of 35

Thread: install just hangs...

  1. Re: install just hangs...

    On 20 Nov 2007 23:38:11 GMT, Joost Kremers wrote:
    > Joost Kremers wrote:
    >> still, i'm hoping some of you may have a few more trouble-shooting tips to
    >> offer, a few things i can try to get slack 12 to boot, or figure out why
    >> it's not booting...

    >
    > ok, here are the results of today's installation attempts:
    >
    > slackware 12.0: hugesmp.s stops booting when messages about ACPI appear. if
    > i boot the hugesmp.s kernel with acpi=off, booting stops when messages
    > about the scheduler appear.
    >
    >
    > anyone have a clue what might be wrong? or how i might go about
    > trouble-shooting this one? or even where i might ask next?
    >
    >


    Joost,

    Just a thought. I recently put together a computer with a Pentium
    Dual Core E2140 processor. Used a dual boot Win 2k and Slackware 12 OS.
    Started doing a little over clocking, mainly on Win 2k and changing
    stuff in the BIOS. Would go back and forth between the OS's and then one
    day tried to boot into Linux and just blew up during the booting. Was
    running just a little over clocked at the time so dropped back to
    standard settings as far as overclocking the processors. Still couldn't
    boot into Slack.
    I had set the ACPI in BIOS to ACPI v3.0. Changed back to v1.0 and away
    it went. Might be worth at least checking to see if you can try a couple
    of changes.



    --
    Ken P



  2. Re: install just hangs...


    Joost Kremers wrote :

    > if i boot 11.0 with the option nousb, i get to a login prompt. i've been
    > able to partition the hard drive, and i would probably be able to install
    > slackware, even 12.0, this way, but i fear that when i do, it won't boot
    > either...


    Worst case scenario: It wont boot Slackware.

    Situation right now: It wont boot Slackware.
    --
    Thomas O.

    This area is designed to become quite warm during normal operation.

  3. Re: install just hangs...

    Ken P wrote:
    > I had set the ACPI in BIOS to ACPI v3.0. Changed back to v1.0 and away
    > it went. Might be worth at least checking to see if you can try a couple
    > of changes.


    yeah, i tried changing all sorts of things in the BIOS setup, most of which
    i'd never heard of before... when i changed some power saving settings, the
    boot did indeed get past the acpi messages (though there were fewer such
    messages now), but it stopped again just a few messages later.

    when i have more time, i'll try a few more options. perhaps i'll hit the
    right combination at some point...

    but really, it almost looks as if the kernel is allowed to boot a few
    seconds, and then after some time is just killed off...

    --
    Joost Kremers joostkremers@yahoo.com
    Selbst in die Unterwelt dringt durch Spalten Licht
    EN:SiS(9)

  4. Re: install just hangs...

    -----BEGIN PGP SIGNED MESSAGE-----
    Hash: SHA1

    I've been following this and I've got to say I'm just as stumped as you
    are Joost.

    On 2007-11-20, Joost Kremers wrote:
    > slackware 12.0: hugesmp.s stops booting when messages about ACPI appear. if
    > i boot the hugesmp.s kernel with acpi=off, booting stops when messages
    > about the scheduler appear.
    >
    > this is consistent across boots, and with two different media. i have a
    > slack 12 dvd and a slack 12 cd, and they both show the same effect, booting
    > always stops at the same point.
    >
    > i've also booted the huge.s kernel, but didn't get any further with
    > it. (besides, there's a core 2 duo in that machine, so i think having smp
    > would be good.)


    Odd that it stops on ACPI, then later stops on scheduler if noacpi is
    set.

    > zenwalk 4.8: doesn't boot. zenwalk doesn't show boot messages, so i don't
    > know where it stops, i just know it does.


    zenwalk 4.8 uses the 2.6.22.x kernel, which is a later minor version
    than what Slackware-12.0 ships. I would see if there was some way to
    get boot messages displayed. If it stops at the same location(s), we
    can pretty much rule the kernel minor version out (unless of course
    this problem is present in both, and given 2.6.22's lengthy release
    time, that seems unlikely).

    > slackware 11.0: stops booting when the message "probing for usb devices"
    > appears. (it says "this may take a few minutes", but i've let it run for a
    > bit longer than that. nothing happens. note that the dvd should be fine,
    > i've used it to install slack 11 on another machine.)
    >
    > if i boot 11.0 with the option nousb, i get to a login prompt. i've been
    > able to partition the hard drive, and i would probably be able to install
    > slackware, even 12.0, this way, but i fear that when i do, it won't boot
    > either...


    Did you use the 2.4 kernel or the 2.6 kernel? This is very important
    need-to-know stuff! You know better than to omit that! :^) Still, I
    highly doubt this has anything to do with the kernel version.

    > so. i'll go out on a limb and say that it's *not* due to bad install
    > media. but other than that, it doesn't make any sense to me. the point
    > where boot messages stop is predictable, but different for each distro
    > version, and depends on boot options.


    I'm ruling out boot media too. There's just too much circumstantial
    evidence against it to pursue that for the time being. I haven't yet
    ruled out the kernel, but I'm betting against it. It sounds like your
    motherboard has a really flaky CMOS and just doesn't want to act right
    with either the kernel or the install media. I'm betting on the kernel.
    I've seen that even in brand-new machines (though usually only in cheap
    ones).[0] Can you provide use with specs such as the motherboard model
    and/or chipset?

    > anyone have a clue what might be wrong? or how i might go about
    > trouble-shooting this one? or even where i might ask next?


    I would boot Slackware-11.0, hopefully with the 2.6 kernel and "nousb",
    then attempt an install of Slackware-12.0. To save time, I'd just do a
    full install of the a/ diskset and install LILO. If it boots, then we
    know for a fact that it's the hardware not playing nice with your boot
    media and not the kernel.

    Another option would be to do a PXE install. Granted, I've never done
    one of these but from what I understand, it's not too hard. You'd need
    a dhcp server and a tftp server with all the right files, and of
    course, your hardware must support PXE booting.

    [0] Right after AMD released the Athlon64 processor, E-Machines was the
    first major manufacturer to sell desktops with this chip. The BIOS
    they used didn't support el torito boot media, so no recent Linux OS
    would boot from a CD. The last version of Slackware that would boot on
    it without using boot floppies was Slackware-8.0 which still used
    floppy emulation.

    - --
    It is better to hear the rebuke of the wise,
    Than for a man to hear the song of fools.
    Ecclesiastes 7:5
    -----BEGIN PGP SIGNATURE-----
    Version: GnuPG v1.4.7 (GNU/Linux)

    iD8DBQFHRJFurZS6hX/gvjoRAhtrAJ4lANEimE2YN9sEayfZZb0eTjXabACfcFAy
    uFKUISSGgFJiLs3V4ThyDrU=
    =wAM7
    -----END PGP SIGNATURE-----

  5. Re: install just hangs...

    +Alan Hicks+ wrote:
    > Odd that it stops on ACPI, then later stops on scheduler if noacpi is
    > set.


    i'll say...

    >> zenwalk 4.8: doesn't boot. zenwalk doesn't show boot messages, so i don't
    >> know where it stops, i just know it does.

    >
    > zenwalk 4.8 uses the 2.6.22.x kernel, which is a later minor version
    > than what Slackware-12.0 ships. I would see if there was some way to
    > get boot messages displayed. If it stops at the same location(s), we
    > can pretty much rule the kernel minor version out (unless of course
    > this problem is present in both, and given 2.6.22's lengthy release
    > time, that seems unlikely).


    there should be a way to show boot messages, but i haven't looked at it yet.

    [slackware 11]
    > Did you use the 2.4 kernel or the 2.6 kernel? This is very important
    > need-to-know stuff! You know better than to omit that! :^)


    yes... i didn't think of the fact that slack 11 came with two kernel
    series.

    originally, i booted the sata.i kernel, which is a 2.4 kernel. today, i
    also tried booting the test26.i kernel, and lo and behold, it stopped right
    after the ACPI messages appeared...

    in the BIOS setup, it's possible to indicate which USB devices should be
    made available after boot: none, just keyboard and mouse, or all. it was
    originally set to "none", which caused the usb keyboard to be unusable. so
    i set it to "all". however, when set to "none", and with a ps/2 keyboard
    plugged in, the 2.4 kernel boots without problems, and i've been able to
    install slack 11. i haven't done anything with it yet, because i had a
    class to teach. so i don't know yet if i can use usb in this setup at
    all... (under windows it's not a problem.)

    btw, this BIOS setting has no effect on the 2.6 kernel.

    > ones).[0] Can you provide use with specs such as the motherboard model
    > and/or chipset?


    yup. the machine is a fujitsu-siemens celsius w360. it's data spec can
    easily be found on the fujitsu-siemens site:

    Mainboard: D2587
    CPU: Core 2 Duo, E6550
    Chipset: Intel Q35

    ok, so i've done a bit of googling, and it seems this chipset is pretty
    new, but it should work under linux.

    > I would boot Slackware-11.0, hopefully with the 2.6 kernel and "nousb",
    > then attempt an install of Slackware-12.0. To save time, I'd just do a
    > full install of the a/ diskset and install LILO. If it boots, then we
    > know for a fact that it's the hardware not playing nice with your boot
    > media and not the kernel.


    well, it doesn't boot. the 2.6 kernel at least. so that suggests it *is*
    the kernel not playing nice with the hardware...

    > Another option would be to do a PXE install. Granted, I've never done
    > one of these but from what I understand, it's not too hard. You'd need
    > a dhcp server and a tftp server with all the right files, and of
    > course, your hardware must support PXE booting.


    it does. (it's listed in the BIOS boot menu as the fourth option.) but it
    requires another machine of course, which is not impossible but at least a
    bit cumbersome.

    i think i have two options now: fiddle with the BIOS settings and hope i
    find a combination that allows me to boot, or create an install cd with a
    custom kernel. i've already been doing the first, but perhaps with a bit
    more fiddling i stumble upon the right combo. the second, well... :-/

    i'd be really great to get slack running on this machine. right now, it's
    running vista, and in spite of all the reports about vista's hardware
    requirements, this thing flies. i've never seen programs come up so
    fast. click on the firefox link and bam, it's there. click on a pdf link
    and bam, the acroread plugin opens up instantly. on my old machine, this
    could take up to a minute...


    --
    Joost Kremers joostkremers@yahoo.com
    Selbst in die Unterwelt dringt durch Spalten Licht
    EN:SiS(9)

  6. Re: install just hangs...

    -----BEGIN PGP SIGNED MESSAGE-----
    Hash: SHA1

    On 2007-11-22, Joost Kremers wrote:
    > originally, i booted the sata.i kernel, which is a 2.4 kernel. today, i
    > also tried booting the test26.i kernel, and lo and behold, it stopped right
    > after the ACPI messages appeared...


    Now we're getting somewhere...

    > ok, so i've done a bit of googling, and it seems this chipset is pretty
    > new, but it should work under linux.


    Definitely a buggy chipset IMHO.

    > well, it doesn't boot. the 2.6 kernel at least. so that suggests it *is*
    > the kernel not playing nice with the hardware...


    Since the 2.4 kernel wrks and the 2.6 kernel doesn't, we just gotta
    figure out what in the 2.6 kernel isn't doing right with the BIOS. You
    check for any BIOS updates yet?

    > i think i have two options now: fiddle with the BIOS settings and hope i
    > find a combination that allows me to boot, or create an install cd with a
    > custom kernel. i've already been doing the first, but perhaps with a bit
    > more fiddling i stumble upon the right combo. the second, well... :-/


    I wouldn't keep fiddling with the BIOS too much longer. I'd load 11.0
    on it with the 2.4 kernel, then start compiling 2.6 kernels and see
    what works. Once you find a winning combination, make yourself a
    custom installation CD and voila.

    - --
    It is better to hear the rebuke of the wise,
    Than for a man to hear the song of fools.
    Ecclesiastes 7:5
    -----BEGIN PGP SIGNATURE-----
    Version: GnuPG v1.4.7 (GNU/Linux)

    iD8DBQFHRw/8rZS6hX/gvjoRAqIuAKC7sRSSyUGNbAUCAvKZA5B59hmgbgCfZ2G/
    JvrIdVk0irIg9ZpqmS/4uCU=
    =RVSa
    -----END PGP SIGNATURE-----

  7. Re: install just hangs...

    +Alan Hicks+ wrote:
    > On 2007-11-22, Joost Kremers wrote:
    >> originally, i booted the sata.i kernel, which is a 2.4 kernel. today, i
    >> also tried booting the test26.i kernel, and lo and behold, it stopped right
    >> after the ACPI messages appeared...

    >
    > Now we're getting somewhere...


    not necessarily very far, though. ;-) remember that if i pass acpi=off to
    the kernel, it still doesn't manage to boot...

    >> ok, so i've done a bit of googling, and it seems this chipset is pretty
    >> new, but it should work under linux.

    >
    > Definitely a buggy chipset IMHO.


    as in: this particular computer is broken, or this chipset in general is
    flakey?

    > Since the 2.4 kernel wrks and the 2.6 kernel doesn't, we just gotta
    > figure out what in the 2.6 kernel isn't doing right with the BIOS. You
    > check for any BIOS updates yet?


    that's a good point. turns out there is one. i'll see if i can apply it.

    > I wouldn't keep fiddling with the BIOS too much longer. I'd load 11.0
    > on it with the 2.4 kernel, then start compiling 2.6 kernels and see
    > what works. Once you find a winning combination, make yourself a
    > custom installation CD and voila.


    yeah, that's what i figured i'd do. though it all honesty, i haven't got a
    single clue what kernel options to try...


    --
    Joost Kremers joostkremers@yahoo.com
    Selbst in die Unterwelt dringt durch Spalten Licht
    EN:SiS(9)

  8. Re: install just hangs...

    On 2007-11-23, Joost Kremers wrote:
    > +Alan Hicks+ wrote:
    >> On 2007-11-22, Joost Kremers wrote:

    >
    >> I wouldn't keep fiddling with the BIOS too much longer. I'd load 11.0
    >> on it with the 2.4 kernel, then start compiling 2.6 kernels and see
    >> what works. Once you find a winning combination, make yourself a
    >> custom installation CD and voila.

    >
    > yeah, that's what i figured i'd do. though it all honesty, i haven't got a
    > single clue what kernel options to try...



    Perhaps try making a -current ISO with the latest 2.6.23.8 kernel in
    it and see if it will boot. I haven't looked at the kernel ChangeLogs
    from 2.6.21.x to 2.6.23.x, but I'm guessing that there's a good chance
    of whatever was wrong (assuming it was a kernel problem) has been
    fixed since then.

    -RW

  9. Re: install just hangs...

    -----BEGIN PGP SIGNED MESSAGE-----
    Hash: SHA1

    On 2007-11-23, Joost Kremers wrote:
    > not necessarily very far, though. ;-) remember that if i pass acpi=off to
    > the kernel, it still doesn't manage to boot...


    Yes, but the 2.4 kernel does boot. That's something at least.

    >> Definitely a buggy chipset IMHO.

    >
    > as in: this particular computer is broken, or this chipset in general is
    > flakey?


    Not enough information to know for certain. I'll admit I've done no
    googling on the chipset information you've sent me. I simply realized
    it was one I knew nothing about and moved on since it seemed that 2.4
    worked. :^)

    >> I wouldn't keep fiddling with the BIOS too much longer. I'd load 11.0
    >> on it with the 2.4 kernel, then start compiling 2.6 kernels and see
    >> what works. Once you find a winning combination, make yourself a
    >> custom installation CD and voila.

    >
    > yeah, that's what i figured i'd do. though it all honesty, i haven't got a
    > single clue what kernel options to try...


    I'd remove all power management and USB options to start.

    - --
    It is better to hear the rebuke of the wise,
    Than for a man to hear the song of fools.
    Ecclesiastes 7:5
    -----BEGIN PGP SIGNATURE-----
    Version: GnuPG v1.4.7 (GNU/Linux)

    iD8DBQFHR3wqrZS6hX/gvjoRAt5oAJ9zD4++8BZl/Nu0uGwGZFMf9Du5dQCdHz5N
    FwaO7vOwbkgxqLVxllH13Jk=
    =KzdJ
    -----END PGP SIGNATURE-----

  10. Re: install just hangs...

    -----BEGIN PGP SIGNED MESSAGE-----
    Hash: SHA1

    Sorry this is followed-up to Robby, since it's almost 100% for Joost.
    :^)

    On 2007-11-24, Robby Workman wrote:
    > Perhaps try making a -current ISO with the latest 2.6.23.8 kernel in
    > it and see if it will boot. I haven't looked at the kernel ChangeLogs
    > from 2.6.21.x to 2.6.23.x, but I'm guessing that there's a good chance
    > of whatever was wrong (assuming it was a kernel problem) has been
    > fixed since then.


    I highly doubt this will work, since the 2.6 kernel shipped with 11.0
    behaved in the same way. What's interesting is that the 2.4 kernel
    _does_ work fine with this chip, meaning this bug may have been
    introduced a long time ago (if it is indeed a software bug in the
    kernel).

    Your post made me consider a rather off-the-wall possibility, but one
    that I think may warrant consideration. The 2.4 kernel works fine
    (perhaps lacking some chipset features), but no as-yet tested 2.6
    kernel does. If after testing the lastest kernel, the symptoms
    continue, I would compile every chipset into the kernel as a module,
    not statically compiled like Slackware kernels do. Then, if it will
    boot attempt to load these modules one at a time and check for erratic
    behavior. The kernel may be using an improper chipset driver. The 2.4
    kernel may be using a much more basic and generic driver that the
    motherboard can deal with, but the 2.6 kernel may load a more advanced
    (and improper) module.

    This could be a bug in the hardware, or a bug in the kernel. I've seen
    instances where the hardware reported itself incorrectly.[0] In any
    event, the fact that it works out of the box with 11.0's 2.4 kernel and
    fails with it's 2.6 kernel (and 12.0's 2.6 kernel) means that this is
    probably best sent to the LKML. I'd do a search for posts containing
    that motherboard chipset to the LKML and see if anyone is working on
    support. You may also want to look at using the -mm kernel as lots of
    new patches are always merged to it.

    [0] For example, many of the early Netgear FA312 NICs were shipped with
    a PCI ID that said they were FA311s.

    - --
    It is better to hear the rebuke of the wise,
    Than for a man to hear the song of fools.
    Ecclesiastes 7:5
    -----BEGIN PGP SIGNATURE-----
    Version: GnuPG v1.4.7 (GNU/Linux)

    iD8DBQFHR3+orZS6hX/gvjoRAlmSAKCazVM9RLzNdBLaxSHvz8yDV/Q+JACfUmam
    bQkrjTix8B2c/yJddeVrUKc=
    =TcGs
    -----END PGP SIGNATURE-----

  11. Re: install just hangs...

    +Alan Hicks+ says:

    pgp trash troll delete

    >Since the 2.4 kernel wrks and the 2.6 kernel doesn't, we just gotta
    >figure out what in the 2.6 kernel isn't doing right with the BIOS. You
    >check for any BIOS updates yet?


    Never mind the BIOS updates. The problem is pgp trash trolling. If
    you would dispense with the pgp trash trolling, things would start
    working very smoothly.

    cordially, as always,

    rm

  12. Re: install just hangs...

    Joost Kremers says:
    > The Coward brayed:


    >> Now we're getting somewhere...


    >not necessarily very far, though. ;-) remember that if i pass
    >acpi=off to the kernel, it still doesn't manage to boot...


    The problem lies entirely in the pgp trash trolling that others
    contribute to the thread. The presence of pgp trash will compromise
    any computer identified in a thread that is soiled by the presence
    of either The Coward Hicks, or any other pgp trash troll.

    Your only option is pgp trash troll delete.

    cordially, as always,

    rm

  13. Re: install just hangs...

    +Alan Hicks+ says:

    pgp trash troll delete

    >On 2007-11-23, Joost Kremers wrote:


    >> not necessarily very far, though. ;-) remember that if i pass acpi=off to
    >> the kernel, it still doesn't manage to boot...


    >Yes, but the 2.4 kernel does boot. That's something at least.


    Not only is the pgp trash trolling a likely offender, but the use of
    +lights+ around a Coward's name can bugger things up as well.

    cordially, as always,

    rm

  14. Re: install just hangs...

    +Alan Hicks+ says:

    pgp trash troll delete

    >I highly doubt this will work, since the 2.6 kernel shipped with 11.0
    >behaved in the same way.


    You're right. Only deleting the pgp trash trolling will work.

    cordially, as always,

    rm

  15. Re: install just hangs...

    Joost Kremers wrote:
    > when i try booting this new machine with the dvd, however, it loads the
    > kernel, starts booting and then at some point just stops.


    time for an update, if someone's still interested, and otherwise for
    google's sake. ;-)

    the relevant machine is a fujitsu-siemens celsius w360. in my search for a
    solution to my problem, i came across a guy who turned out to be the man
    at fujitsu-siemens responsible for making sure that their hardware would be
    able to run linux. he had already come across my problem and had found that
    there was a workaround: i needed to pass the option 'pci=nommconf' to the
    kernel, and then it would boot.

    although i haven't (yet) been able to find out what that option does, i'm
    happy to see slack 12 booting on this machine. from what the guy at
    fujitsu-siemens told me, i got the impression that this may be a BIOS bug,
    and it if is, i'm sure it'll get fixed with some BIOS update in the near
    future. if not, it's a kernel bug, and that will probably also get
    fixed. ;-)

    anyways, he also told me i'd need a very recent kernel, >2.6.24 (which,
    ATM, is still in -rc), to get the network card running, and i think i also
    need a more recent X.org than the one shipping with slack 12, so i'll be
    upgrading those. but from the looks of it, i should have a working
    slackware soon!

    thanks to those that offered suggestions in this thread.


    --
    Joost Kremers joostkremers@yahoo.com
    Selbst in die Unterwelt dringt durch Spalten Licht
    EN:SiS(9)

+ Reply to Thread
Page 2 of 2 FirstFirst 1 2