Date: Thu, 20 Jan 2005 01:33:03 +0900 From: Rob <spamrefuse@yahoo.com> To: Rob <spamrefuse@yahoo.com>, freebsd-stable@freebsd.org Subject: Re: User's cron job creates zombie process on 5.3 Message-ID: <41EE8BBF.1010408@yahoo.com>
next in thread | raw e-mail | index | archive | help
Raymond Wiker wrote:
> Rob writes:
> > Raymond Wiker wrote:
> > > Peter Jeremy writes:
> > > > On Wed, 2005-Jan-19 09:16:59 +0900, Rob Lahaye wrote:
> > > > > tunnel="-L 55110:localhost:110 pop3.univ.net"
> > > > > tunnel_up=`pgrep -f -- "${tunnel}"`
> > > > > [ "${tunnel_up}" = "" ] && /usr/bin/ssh -N -f ${tunnel}
> > > >
> > > > >It works beautifully, but why does this also generate one zombie process:
> > > > > USER PID %CPU %MEM VSZ RSS TT STAT STARTED TIME COMMAND
> > > > > rob 655 0.0 0.0 0 0 ?? Z Sat02PM 0:00.01 <defunct>
> > > >
> > > > You get a zombie when a process has exited and the parent hasn't issued
> > > > a wait(2) (or SIG_IGN'd SIGCHLD). Have a look at what the parent process
> > > > is and that might give you an idea as to what is going wrong.
> > >
> > > Ancient Perl did not collect for children started via the
> > > backtick operator - is this a possible issue for /bin/sh as well? It
> > > should be harmless to call wait just after the use of the backtick
> > > operator above; does that change anything? I.e:
> > >
> > > tunnel_up=`pgrep -f -- "${tunnel}"`; wait
> > >
> > > To see the parent pid, add "-O ppid" to the arguments to ps;
> > > e.g,
> > >
> > > ps axww -O ppid
> >
> > Adding the "wait" here does not help at all.
> >
> > When I verify the parent process I have this:
> >
> > PID PPID TT STAT TIME COMMAND USER %CPU %MEM VSZ RSS STARTED
> > 423 417 ?? I 0:00.00 cron: running jo root 0.0 0.2 1360 1148 11:38PM
> > 425 423 ?? Z 0:00.00 <defunct> lahaye 0.0 0.0 0 0 11:38PM
> >
> >
> > So PID 423 "cron: running job (cron)" is the parent of my <defunct> zombie.
> >
> > After playing with commenting out lines, I found out that the ssh call is the
> > reason for the zombie: "/usr/bin/ssh -N -f -L 55110:localhost:110 pop3.univ.net"
> >
> > Then I tried exec : "exec /usr/bin/ssh -N -f ..."
> > And I also tried & : "/usr/bin/ssh -N -f ... &"
> > Both to no avail.
> >
> > What else can I try, and why is this ssh command causing a zombie process when
> > called from cron?
>
> Hmm... Maybe this will work?
>
> /bin/sh -c "/usr/bin/ssh -n -f ${tunnel} &"
>
> --- the effect of this should (hopefully) be that init becomes the
> parent of the zombie process.
No, makes no difference.
Peculiar of the problem is: I use this construct to keep a ssh-tunnel alive.
What else is better, than having a cron-script check whether the tunnel is
still active, and if not, re-establish the ssh-tunnel.
Strange, that such an obvious contruct ends up with a zombie process.
Another interesting detail: as soon as I kill the tunnel, created by the
cron-script, then the zombie process also disappears. Does that give a clue?
Any more suggestions how to tackle this?
Thsnks,
Rob.
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?41EE8BBF.1010408>