8:14 a.m.
Greetings,
While debugging some problems with thin startup, I made some minor adjustments
to the initscript to improve readability and status reporting.
GIT: [PATCH 1/2] Whitespace cleanup
GIT: [PATCH 2/2] Minor cleanup to initscript
Comments encouraged+welcome.
Thanks,
James
8:14 a.m.
New subject: [PATCH 1/2] Whitespace cleanup
---
conf/aeolus-conductor | 32 ++++++++++++++++----------------
1 files changed, 16 insertions(+), 16 deletions(-)
diff --git a/conf/aeolus-conductor b/conf/aeolus-conductor
index 0099c61..cf74892 100755
--- a/conf/aeolus-conductor
+++ b/conf/aeolus-conductor
@@ -39,24 +39,24 @@ start() {
fi
$THIN_PROG start -c "$CONDUCTOR_DIR" -l "$THIN_LOG" -P
"$THIN_PID" \
- -a $ADDR -e $RAILS_ENV --user $CONDUCTOR_USER -g $CONDUCTOR_GROUP -d --prefix=$PREFIX \
- --rackup "$CONDUCTOR_DIR/config.ru"
+ -a $ADDR -e $RAILS_ENV --user $CONDUCTOR_USER -g $CONDUCTOR_GROUP -d
--prefix=$PREFIX \
+ --rackup "$CONDUCTOR_DIR/config.ru"
RETVAL=$?
if [ $RETVAL -eq 0 ] ; then
# wait until we can contact the server
# this bit is based on what the mysql init script does
TIMEOUT="$STARTTIMEOUT"
- while [ $TIMEOUT -gt 0 ]; do
+ while [ $TIMEOUT -gt 0 ]; do
/usr/bin/curl --silent http://$ADDR:3000$PREFIX >& /dev/null
RETVAL=$?
if [ $RETVAL -eq 0 ] ; then
touch $THIN_LOCKFILE
- echo_success
- echo
+ echo_success
+ echo
exit 0
fi
sleep 1
- let TIMEOUT=${TIMEOUT}-1
+ let TIMEOUT=${TIMEOUT}-1
done
fi
@@ -72,24 +72,24 @@ stop() {
case "$1" in
start)
- start
- ;;
+ start
+ ;;
stop)
- stop
- ;;
+ stop
+ ;;
restart)
- stop
- start
- ;;
+ stop
+ start
+ ;;
reload)
;;
force-reload)
restart
;;
status)
- status -p $THIN_PID
- RETVAL=$?
- ;;
+ status -p $THIN_PID
+ RETVAL=$?
+ ;;
*)
echo "Usage: aeolus-conductor
{start|stop|restart|reload|force-reload|status}"
exit 1
--
1.7.6.4
8:14 a.m.
New subject: [PATCH 2/2] Minor cleanup to initscript
* Reference THIN_PROG instead of aeolus-conductor
* LOCKFILE was not defined
* Ensure removal of PID and LOCKFILE after stop()
* Include THIN_PROG when checking for status()
---
conf/aeolus-conductor | 25 +++++++++++++++----------
1 files changed, 15 insertions(+), 10 deletions(-)
diff --git a/conf/aeolus-conductor b/conf/aeolus-conductor
index cf74892..a3872a7 100755
--- a/conf/aeolus-conductor
+++ b/conf/aeolus-conductor
@@ -19,6 +19,7 @@ CONDUCTOR_GROUP="${CONDUCTOR_GROUP:-aeolus}"
THIN_PROG="${THIN_PROG:-thin}"
THIN_PID="${THIN_PID:-/var/run/aeolus-conductor/thin.pid}"
THIN_LOG="${THIN_LOG:-/var/log/aeolus-conductor/thin.log}"
+THIN_LOCKFILE="${THIN_LOCKFILE:-/var/lock/subsys/thin}"
PREFIX="${PREFIX:-/conductor}"
ADDR="${ADDR:-127.0.0.1}"
@@ -30,11 +31,12 @@ STARTTIMEOUT=60
. /etc/init.d/functions
start() {
- echo -n "Starting aeolus-conductor: "
- if [ -f "$THIN_PID" ] && checkpid `cat $THIN_PID` ; then
+ echo -n "Starting ${THIN_PROG}: "
+ RUNNING_THIN=$(pidofproc -p ${THIN_PID} ${THIN_PROG})
+ if [ -n "$RUNNING_THIN" ]; then
echo_failure
echo
- echo "Thin is already running"
+ echo "${THIN_PROG} is already running"
exit 1
fi
@@ -50,9 +52,9 @@ start() {
/usr/bin/curl --silent http://$ADDR:3000$PREFIX >& /dev/null
RETVAL=$?
if [ $RETVAL -eq 0 ] ; then
- touch $THIN_LOCKFILE
- echo_success
- echo
+ touch ${THIN_LOCKFILE}
+ echo_success
+ echo
exit 0
fi
sleep 1
@@ -66,8 +68,11 @@ start() {
}
stop() {
- echo -n "Shutting down aeolus-conductor: "
- killproc -p $THIN_PID thin
+ echo -n "Shutting down ${THIN_PROG}: "
+ killproc -p ${THIN_PID} ${THIN_PROG}
+ RETVAL=$?
+ echo
+ [ $RETVAL = 0 ] && rm -f ${THIN_LOCKFILE} ${THIN_PID}
}
case "$1" in
@@ -87,11 +92,11 @@ case "$1" in
restart
;;
status)
- status -p $THIN_PID
+ status -p ${THIN_PID} ${THIN_PROG}
RETVAL=$?
;;
*)
- echo "Usage: aeolus-conductor
{start|stop|restart|reload|force-reload|status}"
+ echo "Usage: $(basename $0)
{start|stop|restart|reload|force-reload|status}"
exit 1
;;
esac
--
1.7.6.4
8:21 a.m.
On Mon, Oct 31, 2011 at 09:14:51AM -0400, James Laska wrote:
Greetings,
While debugging some problems with thin startup, I made some minor adjustments
to the initscript to improve readability and status reporting.
> GIT: [PATCH 1/2] Whitespace cleanup
> GIT: [PATCH 2/2] Minor cleanup to initscript
Comments encouraged+welcome.
Thanks,
James
Nice James. I don't suppose you'd be interested in auditing the rest
of the init scripts as well?
--Hugh
--
== Hugh Brock, hbrock(a)redhat.com ==
== Engineering Manager, Cloud BU ==
== Aeolus Project: Manage virtual infrastructure across clouds. ==
== http://aeolusproject.org ==
"I know that you believe you understand what you think I said, but I’m
not sure you realize that what you heard is not what I meant."
--Robert McCloskey
8:29 a.m.
On Mon, 2011-10-31 at 09:21 -0400, Hugh Brock wrote:
On Mon, Oct 31, 2011 at 09:14:51AM -0400, James Laska wrote:
> Greetings,
>
> While debugging some problems with thin startup, I made some minor adjustments
> to the initscript to improve readability and status reporting.
>
> > GIT: [PATCH 1/2] Whitespace cleanup
> > GIT: [PATCH 2/2] Minor cleanup to initscript
>
> Comments encouraged+welcome.
>
> Thanks,
> James
Nice James. I don't suppose you'd be interested in auditing the rest
of the init scripts as well?
Sure. If these changes make the cut, I'll look at the following ...
/etc/rc.d/init.d/aeolus-conductor
/etc/rc.d/init.d/conductor-dbomatic
/etc/rc.d/init.d/aeolus-configserver
/etc/rc.d/init.d/deltacloud-core
/etc/rc.d/init.d/imagefactory
/etc/rc.d/init.d/iwhd
Feel free to adjust the above list as needed.
Thanks,
James
4569
days inactive
4569
days old
aeolus-devel@lists.fedorahosted.org
4 comments
2 participants
participants (2)
-
Hugh O. Brock -
James Laska