[Condor-devel] Stopping condor fails with stale PID file [PATCH]

[Michael Hanke <michael.hanke@xxxxxxxxx>]

Hi,

on a system with a stale PID file (no condor_master running, e.g. after
a crash of the master) the init script's stop action fails, because it
waits for a non-existing process to end. This behavior can cause, for
example, a Debian package upgrade to fail. The attached patch addresses
this problem. I'd be glad if you could have a look at it and let me know
whether there are undesired side-effect of such a fix.

Thanks in advance,

Michael


-- 
Michael Hanke
http://mih.voxindeserto.de

From: Michael Hanke <michael.hanke@xxxxxxxxx>
Subject: Prevent init script failure on stop with stale PID file

If there is a stale condor PID file on a system the init script would wait till
timeout for a non-existing process to end. This effectively prevent a Debian
package from being uninstallable (and hence also upgradable), because the init
script will always fail until the PID file is removed.

This patch validates the PID read from the file and ignores it if no running
process for this PID can be found.

diff --git a/src/condor_examples/condor.boot.rpm b/src/condor_examples/condor.boot.rpm
index 2c58661..ef8d864 100755
--- a/src/condor_examples/condor.boot.rpm
+++ b/src/condor_examples/condor.boot.rpm
@@ -518,7 +518,14 @@ condor_master_pids() {
   masterpid=
   if [ -f "$PIDFILE" -a -r "$PIDFILE" ] ; then
     masterpid=`cat "$PIDFILE"` 2>/dev/null
-    FORCE_PIDFILE="yes"
+    # validate PID
+    `$PS | grep "^$masterpid "` > /dev/null
+    if [ $? -eq 0 ] ; then
+      FORCE_PIDFILE="yes"
+    else
+      masterpid=""
+      # maybe remove the stale PID file here
+    fi
   fi
   if [ "$FORCE_PIDFILE" = "yes" ] ; then
     MASTER_PIDS="$masterpid"