osr6 not killing child procs when osr5 did - SCO

This is a discussion on osr6 not killing child procs when osr5 did - SCO ; This problem is new to OpenServer 6.0 (with MP2). The exact same windows clients/client terminal emulation software has worked fine with prior SCO releases (OSR 5.x), all that was upgraded was the server OS from 5.06 to 6.0 mp2 What ...

+ Reply to Thread
Results 1 to 6 of 6

Thread: osr6 not killing child procs when osr5 did

  1. osr6 not killing child procs when osr5 did

    This problem is new to OpenServer 6.0 (with MP2).
    The exact same windows clients/client terminal emulation software has
    worked fine with prior SCO releases (OSR 5.x), all that was upgraded
    was the server OS from 5.06 to 6.0 mp2

    What happens is the users open up with their windows XP terminal emulation
    software (attachmate, hummingbird, i was also able to duplicate with stock
    hyperterm and anzio) windows - run some applications in the windows, then
    close the windows via the windows "X" close-app. This shuts the login and
    correctly updates utmp/wtmp (correctly being relative as the entire thing
    is useless after a reboot due to other OSR6 bug) - but never kills the
    other (child) processes run within that login session. The ttyp* is
    never released, and eventually they run out of licenses.

    I have verified that it happens whether coming in via telnetd or sshd.

    Unfortunately, we've no control over the department causing this headache,
    we can't swap out the client software, can't train/teach them other methods
    as they are used to this for their windows apps and to be fair - all their
    habits worked just perfectly until SCO server OS was updated.

    --- - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - ---
    -Joe Chasan- Magnatech Business Systems, Inc.
    joe@magnatechonline.com Hicksville, NY - USA
    http://www.MagnatechOnline.com Tel.(516) 931-4444/Fax.(516) 931-1264

  2. Re: osr6 not killing child procs when osr5 did

    What shell are the users using? You might be looking at a process
    handling issue which is specific to each shell.

    --R Robert


    Joe Chasan wrote:
    > This problem is new to OpenServer 6.0 (with MP2).
    > The exact same windows clients/client terminal emulation software has
    > worked fine with prior SCO releases (OSR 5.x), all that was upgraded
    > was the server OS from 5.06 to 6.0 mp2
    >
    > What happens is the users open up with their windows XP terminal emulation
    > software (attachmate, hummingbird, i was also able to duplicate with stock
    > hyperterm and anzio) windows - run some applications in the windows, then
    > close the windows via the windows "X" close-app. This shuts the login and
    > correctly updates utmp/wtmp (correctly being relative as the entire thing
    > is useless after a reboot due to other OSR6 bug) - but never kills the
    > other (child) processes run within that login session. The ttyp* is
    > never released, and eventually they run out of licenses.
    >
    > I have verified that it happens whether coming in via telnetd or sshd.
    >
    > Unfortunately, we've no control over the department causing this headache,
    > we can't swap out the client software, can't train/teach them other methods
    > as they are used to this for their windows apps and to be fair - all their
    > habits worked just perfectly until SCO server OS was updated.
    >
    > --- - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - ---
    > -Joe Chasan- Magnatech Business Systems, Inc.
    > joe@magnatechonline.com Hicksville, NY - USA
    > http://www.MagnatechOnline.com Tel.(516) 931-4444/Fax.(516) 931-1264



  3. Re: osr6 not killing child procs when osr5 did

    On Thu, Jul 27, 2006 at 11:56:31AM -0700, ThreeStar wrote:
    > What shell are the users using? You might be looking at a process
    > handling issue which is specific to each shell.


    bourne shell, and i have traced it to trap command being passed
    differently it seems from child processes on osr6 as compared to
    osr5. in any event, by undoing trap command it would appear i've
    solved the issue.

    thanks.

    -joe

    > Joe Chasan wrote:
    > > This problem is new to OpenServer 6.0 (with MP2).
    > > The exact same windows clients/client terminal emulation software has
    > > worked fine with prior SCO releases (OSR 5.x), all that was upgraded
    > > was the server OS from 5.06 to 6.0 mp2
    > >
    > > What happens is the users open up with their windows XP terminal emulation
    > > software (attachmate, hummingbird, i was also able to duplicate with stock
    > > hyperterm and anzio) windows - run some applications in the windows, then
    > > close the windows via the windows "X" close-app. This shuts the login and
    > > correctly updates utmp/wtmp (correctly being relative as the entire thing
    > > is useless after a reboot due to other OSR6 bug) - but never kills the
    > > other (child) processes run within that login session. The ttyp* is
    > > never released, and eventually they run out of licenses.
    > >
    > > I have verified that it happens whether coming in via telnetd or sshd.
    > >
    > > Unfortunately, we've no control over the department causing this headache,
    > > we can't swap out the client software, can't train/teach them other methods
    > > as they are used to this for their windows apps and to be fair - all their
    > > habits worked just perfectly until SCO server OS was updated.
    > >
    > > --- - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - ---
    > > -Joe Chasan- Magnatech Business Systems, Inc.
    > > joe@magnatechonline.com Hicksville, NY - USA
    > > http://www.MagnatechOnline.com Tel.(516) 931-4444/Fax.(516) 931-1264

    >

    --- - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - ---
    -Joe Chasan- Magnatech Business Systems, Inc.
    joe@magnatechonline.com Hicksville, NY - USA
    http://www.MagnatechOnline.com Tel.(516) 931-4444/Fax.(516) 931-1264

  4. Re: osr6 not killing child procs when osr5 did



    Joe Chasan wrote:
    > On Thu, Jul 27, 2006 at 11:56:31AM -0700, ThreeStar wrote:
    > > What shell are the users using? You might be looking at a process
    > > handling issue which is specific to each shell.

    >
    > bourne shell, and i have traced it to trap command being passed
    > differently it seems from child processes on osr6 as compared to
    > osr5. in any event, by undoing trap command it would appear i've
    > solved the issue.
    >
    > thanks.
    >
    > -joe
    >
    > > Joe Chasan wrote:
    > > > This problem is new to OpenServer 6.0 (with MP2).
    > > > The exact same windows clients/client terminal emulation software has
    > > > worked fine with prior SCO releases (OSR 5.x), all that was upgraded
    > > > was the server OS from 5.06 to 6.0 mp2
    > > >
    > > > What happens is the users open up with their windows XP terminal emulation
    > > > software (attachmate, hummingbird, i was also able to duplicate with stock
    > > > hyperterm and anzio) windows - run some applications in the windows, then
    > > > close the windows via the windows "X" close-app. This shuts the login and
    > > > correctly updates utmp/wtmp (correctly being relative as the entire thing
    > > > is useless after a reboot due to other OSR6 bug) - but never kills the
    > > > other (child) processes run within that login session. The ttyp* is
    > > > never released, and eventually they run out of licenses.
    > > >
    > > > I have verified that it happens whether coming in via telnetd or sshd.
    > > >
    > > > Unfortunately, we've no control over the department causing this headache,
    > > > we can't swap out the client software, can't train/teach them other methods
    > > > as they are used to this for their windows apps and to be fair - all their
    > > > habits worked just perfectly until SCO server OS was updated.
    > > >
    > > > --- - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - ---
    > > > -Joe Chasan- Magnatech Business Systems, Inc.
    > > > joe@magnatechonline.com Hicksville, NY - USA
    > > > http://www.MagnatechOnline.com Tel.(516) 931-4444/Fax.(516) 931-1264

    > >

    > --- - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - ---
    > -Joe Chasan- Magnatech Business Systems, Inc.
    > joe@magnatechonline.com Hicksville, NY - USA
    > http://www.MagnatechOnline.com Tel.(516) 931-4444/Fax.(516) 931-1264


    Hi Joe,
    I've been trying to solve the same issue.
    How did you undo the trap command, if you don't mind my stupid
    question.

    Thanks,
    Don.


  5. Re: osr6 not killing child procs when osr5 did

    On Thu, Jul 27, 2006 at 01:36:28PM -0700, foz_67 wrote:
    >
    >
    > Joe Chasan wrote:
    > > On Thu, Jul 27, 2006 at 11:56:31AM -0700, ThreeStar wrote:
    > > > What shell are the users using? You might be looking at a process
    > > > handling issue which is specific to each shell.

    > >
    > > bourne shell, and i have traced it to trap command being passed
    > > differently it seems from child processes on osr6 as compared to
    > > osr5. in any event, by undoing trap command it would appear i've
    > > solved the issue.
    > >
    > > thanks.
    > >
    > > -joe
    > >
    > > > Joe Chasan wrote:
    > > > > This problem is new to OpenServer 6.0 (with MP2).
    > > > > The exact same windows clients/client terminal emulation software has
    > > > > worked fine with prior SCO releases (OSR 5.x), all that was upgraded
    > > > > was the server OS from 5.06 to 6.0 mp2
    > > > >
    > > > > What happens is the users open up with their windows XP terminal emulation
    > > > > software (attachmate, hummingbird, i was also able to duplicate with stock
    > > > > hyperterm and anzio) windows - run some applications in the windows, then
    > > > > close the windows via the windows "X" close-app. This shuts the login and
    > > > > correctly updates utmp/wtmp (correctly being relative as the entire thing
    > > > > is useless after a reboot due to other OSR6 bug) - but never kills the
    > > > > other (child) processes run within that login session. The ttyp* is
    > > > > never released, and eventually they run out of licenses.
    > > > >
    > > > > I have verified that it happens whether coming in via telnetd or sshd.
    > > > >
    > > > > Unfortunately, we've no control over the department causing this headache,
    > > > > we can't swap out the client software, can't train/teach them other methods
    > > > > as they are used to this for their windows apps and to be fair - all their
    > > > > habits worked just perfectly until SCO server OS was updated.
    > > > >
    > > > > --- - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - ---
    > > > > -Joe Chasan- Magnatech Business Systems, Inc.
    > > > > joe@magnatechonline.com Hicksville, NY - USA
    > > > > http://www.MagnatechOnline.com Tel.(516) 931-4444/Fax.(516) 931-1264
    > > >

    > > --- - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - ---
    > > -Joe Chasan- Magnatech Business Systems, Inc.
    > > joe@magnatechonline.com Hicksville, NY - USA
    > > http://www.MagnatechOnline.com Tel.(516) 931-4444/Fax.(516) 931-1264

    >
    > Hi Joe,
    > I've been trying to solve the same issue.
    > How did you undo the trap command, if you don't mind my stupid
    > question.
    >
    > Thanks,
    > Don.


    you would only have to undo traps you intentionally set with your
    own trap command to begin with. if you weren't setting it, this
    may not help you.

    a common example is:
    trap : 2 #disables key
    trap 2 #re-enables break key

    you can combine more than one on a line.
    to display current traps, just type "trap" with no arguments

    when i trap everything, i may have something like:
    trap "some command here" 0 1 2 3 4 5 6 9 15
    and when undoing it will just be:
    trap 0 1 2 3 4 5 6 9 15

    --- - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - ---
    -Joe Chasan- Magnatech Business Systems, Inc.
    joe@magnatechonline.com Hicksville, NY - USA
    http://www.MagnatechOnline.com Tel.(516) 931-4444/Fax.(516) 931-1264

  6. Re: osr6 not killing child procs when osr5 did

    Thanks for the detailed help!!
    Excellent!
    Don.


    Joe Chasan wrote:
    > On Thu, Jul 27, 2006 at 01:36:28PM -0700, foz_67 wrote:
    > >
    > >
    > > Joe Chasan wrote:
    > > > On Thu, Jul 27, 2006 at 11:56:31AM -0700, ThreeStar wrote:
    > > > > What shell are the users using? You might be looking at a process
    > > > > handling issue which is specific to each shell.
    > > >
    > > > bourne shell, and i have traced it to trap command being passed
    > > > differently it seems from child processes on osr6 as compared to
    > > > osr5. in any event, by undoing trap command it would appear i've
    > > > solved the issue.
    > > >
    > > > thanks.
    > > >
    > > > -joe
    > > >
    > > > > Joe Chasan wrote:
    > > > > > This problem is new to OpenServer 6.0 (with MP2).
    > > > > > The exact same windows clients/client terminal emulation software has
    > > > > > worked fine with prior SCO releases (OSR 5.x), all that was upgraded
    > > > > > was the server OS from 5.06 to 6.0 mp2
    > > > > >
    > > > > > What happens is the users open up with their windows XP terminal emulation
    > > > > > software (attachmate, hummingbird, i was also able to duplicate with stock
    > > > > > hyperterm and anzio) windows - run some applications in the windows, then
    > > > > > close the windows via the windows "X" close-app. This shuts the login and
    > > > > > correctly updates utmp/wtmp (correctly being relative as the entire thing
    > > > > > is useless after a reboot due to other OSR6 bug) - but never kills the
    > > > > > other (child) processes run within that login session. The ttyp* is
    > > > > > never released, and eventually they run out of licenses.
    > > > > >
    > > > > > I have verified that it happens whether coming in via telnetd or sshd.
    > > > > >
    > > > > > Unfortunately, we've no control over the department causing this headache,
    > > > > > we can't swap out the client software, can't train/teach them other methods
    > > > > > as they are used to this for their windows apps and to be fair - all their
    > > > > > habits worked just perfectly until SCO server OS was updated.
    > > > > >
    > > > > > --- - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - ---
    > > > > > -Joe Chasan- Magnatech Business Systems, Inc.
    > > > > > joe@magnatechonline.com Hicksville, NY - USA
    > > > > > http://www.MagnatechOnline.com Tel.(516) 931-4444/Fax.(516) 931-1264
    > > > >
    > > > --- - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - ---
    > > > -Joe Chasan- Magnatech Business Systems, Inc.
    > > > joe@magnatechonline.com Hicksville, NY - USA
    > > > http://www.MagnatechOnline.com Tel.(516) 931-4444/Fax.(516) 931-1264

    > >
    > > Hi Joe,
    > > I've been trying to solve the same issue.
    > > How did you undo the trap command, if you don't mind my stupid
    > > question.
    > >
    > > Thanks,
    > > Don.

    >
    > you would only have to undo traps you intentionally set with your
    > own trap command to begin with. if you weren't setting it, this
    > may not help you.
    >
    > a common example is:
    > trap : 2 #disables key
    > trap 2 #re-enables break key
    >
    > you can combine more than one on a line.
    > to display current traps, just type "trap" with no arguments
    >
    > when i trap everything, i may have something like:
    > trap "some command here" 0 1 2 3 4 5 6 9 15
    > and when undoing it will just be:
    > trap 0 1 2 3 4 5 6 9 15
    >
    > --- - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - ---
    > -Joe Chasan- Magnatech Business Systems, Inc.
    > joe@magnatechonline.com Hicksville, NY - USA
    > http://www.MagnatechOnline.com Tel.(516) 931-4444/Fax.(516) 931-1264



+ Reply to Thread