On 27/03/2008, Eddy Petrișor <[EMAIL PROTECTED]> wrote: > On 27/03/2008, Bastian Blank <[EMAIL PROTECTED]> wrote: > > On Wed, Mar 26, 2008 at 09:18:59PM +0100, Frans Pop wrote: > > > In Debian Installer we're seeing a regression in pidof. For certain > > > processes it no longer returns the PID. > > > > > > # ps ax | grep [d]ebian-installer > > > 452 root 1408 S /bin/sh /sbin/debian-installer > > > 5594 root 1964 S /bin/sh /sbin/debian-installer > /bin/network-console-m > > > 5641 root 1964 S /bin/sh /sbin/debian-installer > /bin/network-console-m > > > # pidof debian-installer > > > # > > > > > > This matches the behaviour of the sysvinit pidof: > > | $ pidof sh > > | 8402 > > | $ pidof test.sh > > | $ ps x | grep test.sh > > | 8402 pts/4 S+ 0:00 /bin/sh ./test.sh > > > current sysvinit does also this:
Errr, what I meant was that pgrep (if available in busybox - can't test, recently lenny qemu refuses to start d-i images) could replace pidof. And, of course, pgrep is part of procps. > 0 [EMAIL PROTECTED] ~ $ ps axu | grep kits_req_server > eddy 23013 0.0 0.1 4520 720 pts/7 S 10:51 0:00 > /bin/sh ./kits_req_server.sh > eddy 23015 0.0 0.0 4520 312 pts/7 S 10:51 0:00 > /bin/sh ./kits_req_server.sh > eddy 25990 0.0 0.1 3716 728 pts/7 S+ 11:31 0:00 grep > --colour kits_req_server > 0 [EMAIL PROTECTED] ~ $ pgrep kits_req_server.sh > 1 [EMAIL PROTECTED] ~ $ pgrep kits_req_server > 23013 > 23015 -- Regards, EddyP ============================================= "Imagination is more important than knowledge" A.Einstein
