fork failed: not enough space during execute command when updating to .22 - SGI

This is a discussion on fork failed: not enough space during execute command when updating to .22 - SGI ; Hello during updating my origin 200 from .21 to .22 i receive a lot of similars errors like eoe.sw.base: ( cd $rbase/dev;find . -type d -print|xargs chmod g-w 2>&- ; exit 0 ) ERROR: An error occurred while: Upgrading selected ...

+ Reply to Thread
Results 1 to 3 of 3

Thread: fork failed: not enough space during execute command when updating to .22

  1. fork failed: not enough space during execute command when updating to .22

    Hello

    during updating my origin 200 from .21 to .22 i receive a lot of similars
    errors like

    eoe.sw.base: ( cd $rbase/dev;find . -type d -print|xargs chmod g-w 2>&- ;
    exit 0 )
    ERROR: An error occurred while:
    Upgrading selected eoe.sw subsystems
    Unable to execute "( cd $rbase/dev;find . -type d -print|xargs chmod g-w
    2>&- ; exit 0 )": fork failed: Not enough space (subsystem eo
    e.sw.base 1289434520)


    par shows the following

    0mS[ 3] : END-select() = 1
    0mS[ 3] : read(0, "1", 1) = 1
    0mS[ 3] : read(0, "\n", 1) = 1
    0mS[ 3] : ioctl(1, TIOCGWINSZ, {row=41, col=135,
    xpixel=0, ypixel=0}) OK
    0mS[ 3] : write(1, "eoe.sw.base: ( cd $rbase/dev;f",
    83) = 83
    1mS[ 3] : time() = 1068915037
    1mS[ 3] : write(3, "11/15/2003 17:50 eoe.sw.base: ",
    100) = 100
    1mS[ 3] : pipe() = 5, 6
    1mS[ 3] : pipe() = 7, 8
    1mS[ 3] : sproc(0xfa75110, PR_SADDR|PR_BLOCK|PR_NOLIBC,
    0x7ffb7190) errno = 12 (Not enough space)
    2mS[ 3] : close(7) OK
    2mS[ 3] : close(8) OK
    2mS[ 3] : close(5) OK
    2mS[ 3] : close(6) OK
    2mS[ 3] : time() = 1068915037
    2mS[ 3] : write(3, "11/15/2003 17:50 ERROR: An err",
    246) = 246


    [origin]:~ $ systune | grep -i RLIMIT_STACK
    rlimit_stack_max = 1288870912 (0x4cd29c00) ll
    rlimit_stack_cur = 1000870912 (0x3ba81400) ll

    any hints whats to do now?

    regards
    Joerg








  2. Re: fork failed: not enough space during execute command when updating to .22

    In article ,
    Joerg Behrens wrote:
    :during updating my origin 200 from .21 to .22 i receive a lot of similars
    :errors like

    :eoe.sw.base: ( cd $rbase/dev;find . -type d -print|xargs chmod g-w 2>&- ;
    :exit 0 )
    :ERROR: An error occurred while:
    : Upgrading selected eoe.sw subsystems
    :Unable to execute "( cd $rbase/dev;find . -type d -print|xargs chmod g-w
    :2>&- ; exit 0 )": fork failed: Not enough space (subsystem eo
    :e.sw.base 1289434520)

    Are you doing this in the miniroot or "live" ?

    The problem looks to me to be one of running out of swap space
    while virtual swap (vswap) is turned off. I would -tend- to expect that
    to be associated with miniroot installs, which would normally have
    vswap off... but of course you might have vswap off in normal
    operations.

    I would suggest trying again with vswap turned on and set to a fair size.
    The fork() operations are wanting to duplicate the entire 'inst' address
    space momentarily, but they give up nearly all of it as soon as the
    exec() of the 'sh' takes place.
    --
    History is a pile of debris -- Laurie Anderson

  3. Re: fork failed: not enough space during execute command when updating to .22

    "Walter Roberson" schrieb im Newsbeitrag
    news:bp6b49$30r$1@canopus.cc.umanitoba.ca...
    > In article ,
    > Joerg Behrens wrote:
    > :during updating my origin 200 from .21 to .22 i receive a lot of similars
    > :errors like
    >
    > :eoe.sw.base: ( cd $rbase/dev;find . -type d -print|xargs chmod g-w 2>&- ;
    > :exit 0 )
    > :ERROR: An error occurred while:
    > : Upgrading selected eoe.sw subsystems
    > :Unable to execute "( cd $rbase/dev;find . -type d -print|xargs chmod g-w
    > :2>&- ; exit 0 )": fork failed: Not enough space (subsystem eo
    > :e.sw.base 1289434520)
    >
    > Are you doing this in the miniroot or "live" ?


    I dont know exactly what you mean. But i dont think that there is the
    miniroot invoked. I'm going to install the .22 overlays from 'live' from
    disk.

    > The problem looks to me to be one of running out of swap space
    > while virtual swap (vswap) is turned off. I would -tend- to expect that
    > to be associated with miniroot installs, which would normally have
    > vswap off... but of course you might have vswap off in normal
    > operations.


    Yeah your'e right. vswap was disabled. But its disabled on all the other
    origins 200 here too.
    Ok.. i turned it on. Increase the 80000 default up to 320000 and start it
    with /etc/init.d/swap stop | start.

    [origin]:~ $ swap -l
    lswap path dev pri swaplo blocks free maxswap vswap
    2 /usr/swap
    0,414 4 0 0 0 0 804800
    3 /.swap.virtual
    0,414 2 0 0 0 0 320000
    1 /dev/swap
    0,419 0 0 4096000 4096000 4096000 0
    [origin]:~ $ swap -s
    total: 0.00k allocated + 200.31m add'l reserved = 200.31m bytes used, 6.26g
    bytes available

    4 CPUs: 95,6% idle, 3,0% usr, 1,4% ker, 0,0% wait, 0,0% xbrk, 0,1% intr
    Memory: 4096M max, 3962M avail, 2859M free, 2000M swap, 2000M free swap

    > I would suggest trying again with vswap turned on and set to a fair size.
    > The fork() operations are wanting to duplicate the entire 'inst' address
    > space momentarily, but they give up nearly all of it as soon as the
    > exec() of the 'sh' takes place.


    Now i tried inst -f ./ again and go to reinstall the already installed
    products with
    keep *
    install same
    go

    But the problem still exists. It starts at 41% of the installation. All
    prior executes went fine.

    regards
    Joerg Behrens



+ Reply to Thread