Para modificar la estética del titulo de cada video deberemos modificar el siguiente archivo openmcu-ru/conference.h
/// Video Mixer Configurator - Begin /// #define VMPC_CONFIGURATION_NAME "layouts.conf" #define VMPC_DEFAULT_ID "undefined" #define VMPC_DEFAULT_FW 704 #define VMPC_DEFAULT_FH 576 #define VMPC_DEFAULT_POSX 0 #define VMPC_DEFAULT_POSY 0 #define VMPC_DEFAULT_WIDTH VMPC_DEFAULT_FW/2 #define VMPC_DEFAULT_HEIGHT VMPC_DEFAULT_FH/2 #define VMPC_DEFAULT_MODE_MASK 0 #define VMPC_DEFAULT_BORDER 1 #define VMPC_DEFAULT_VIDNUM 0 #define VMPC_DEFALUT_SCALE_MODE 1 #define VMPC_DEFAULT_REALLOCATE_ON_DISCONNECT 1 #define VMPC_DEFAULT_NEW_FROM_BEGIN 1 #define VMPC_DEFAULT_MOCKUP_WIDTH 388 #define VMPC_DEFAULT_MOCKUP_HEIGHT 218 #ifdef USE_FREETYPE # define VMPC_DEFAULT_LABEL_MASK 89 //# define VMPC_DEFAULT_LABEL_COLOR 0x0ffffff # define VMPC_DEFAULT_LABEL_BGCOLOR 0xA85D //RGB 0x115599 # define VMPC_DEFAULT_FONTFILE "Russo_One.ttf" # define VMPC_DEFAULT_FONTSIZE "1/16" # define VMPC_DEFAULT_BORDER_LEFT "5/80" # define VMPC_DEFAULT_BORDER_RIGHT "5/80" # define VMPC_DEFAULT_BORDER_TOP "1/200" # define VMPC_DEFAULT_BORDER_BOTTOM "1/100" # define VMPC_DEFAULT_H_PAD "1/16" # define VMPC_DEFAULT_V_PAD "1/24" # define VMPC_DEFAULT_SHADOW_L "1/200" # define VMPC_DEFAULT_SHADOW_R "1/80" # define VMPC_DEFAULT_SHADOW_T "1/150" # define VMPC_DEFAULT_SHADOW_B "1/65" # define VMPC_DEFAULT_CUT_BEFORE_BRACKET 1 # define VMPC_DEFAULT_MINIMUM_WIDTH_FOR_LABEL "1/5" #endif
I threw this service monitoring script together because occasionally the OpenMCU-ru process in the developmental versions can crash out on my Ubuntu Server from time to time, so I wanted it to auto-restart. The script could easily be modified and used for any other service you like, but it was written specifically for OpenMCU-ru and Debian based operating systems.
There are two pieces to this puzzle (which makes for an awfully pathetic and simplistic puzzle), the “/etc/init.d/mcu-monitor” startup script to run the monitoring script as a daemon and the “/usr/local/bin/mcu-monitor.sh” monitoring script itself, both are fairly simple and straight forward.
As usual, to run this service at startup you can issue:
update-rc.d mcu-monitor defaults
#!/bin/bash # OpenMCU-ru process monitor startup script # Author: Nathan Thomas # Date: 04/14/2015 # ### BEGIN INIT INFO # Provides: mcu-monitor # Required-Start: $network $syslog $named $local_fs $remote_fs # Required-Stop: $network $syslog $named $local_fs $remote_fs # Default-Start: 2 3 4 5 # Default-Stop: 0 1 6 # X-Interactive: true # Short-Description: Start/stop OpenMCU-ru process monitor daemon ### END INIT INFO PROG=mcu-monitor DESC="Auto restarts the OpenMCU-ru service upon failure" HOME_DIR=/usr/local/bin DAEMON=${HOME_DIR}/${PROG}.sh PIDFILE=/var/run/${PROG}.pid pscheck() { # return number of processes echo "$( ps aux | grep "${PROG}.sh" | grep -cv "grep" 2>&1 )" return 0 } pidsmatch() { # do the curtains match the drapes? if [ -e $PIDFILE ] ; then # if pidfile exists if [ "$( pscheck )" -eq "1" ] ; then # if num procs=1 if [ "$( ps aux | grep "${PROG}.sh" | grep -v "grep" | awk '{ print $2 }' )" -eq "$( cat ${PIDFILE} )" ] ; then # compare pids echo "yes" return 0 fi elif [ "$( pscheck )" -gt "1" ] ; then return 1 # Hmm...multiple procs, we've got a problem else echo "no" return 0 fi else return 1 fi } rmpid() { # say bye bye to mr piddlesworth if [ -e $PIDFILE ] ; then rm -f $PIDFILE 2>&1 if [ "$?" -ne "0" ] ; then return 1 fi fi return 0 } killpids() { # if procCount>=1, squash em if [ "$( pscheck )" -ge "1" ] ; then kill $( pgrep "${PROG}.sh" 2>&1 ) 2>&1 if [ "$?" -ne "0" ] ; then return 1 fi fi return 0 } start() { rmpid # dump the pidfile killpids # not kids # Start the service echo -n $"Starting ${PROG} service: " start-stop-daemon --start --chuid root --pidfile $PIDFILE --chdir $HOME_DIR --background --make-pidfile --exec $DAEMON || return 1 2>&1 echo $"OK" return 0 } stop() { echo -n $"Stopping ${PROG} service: " start-stop-daemon --stop --pidfile $PIDFILE || return 1 2>&1 rmpid # dump the pidfile if [ "$?" -eq "0" -a "$( pscheck )" -ne "0" ] ; then # if pidfile is gone and somehow procs still alive killpids # not kids if [ "$?" -ne "0" ] ; then echo $"FAILED" return 1 fi fi echo $"OK" return 0 } restart() { stop && start } status() { if [ "$( pscheck )" -eq "1" -a "$( pidsmatch )" == "yes" ] ; then echo $"${PROG} (pid $( cat $PIDFILE )) is running" elif [ "$( pscheck )" -eq "1" -a "$( pidsmatch )" == "no" ] ; then echo $"Strange, ${PROG} is running, but PID doesn't match pidfile, fixing..." restart elif [ "$( pscheck )" -gt "1" ] ; then echo $"Multiple instances of ${PROG} running, fixing..." restart else echo $"${PROG} is stopped" fi return 0 } case "$1" in 'start') start ;; 'stop') stop ;; 'restart') restart ;; 'status') status ;; *) echo $"Usage: $0 {start|stop|restart|status}" exit 1 ;; esac exit $?
#!/bin/bash # Auto restart OpenMCU-ru when it crashes # Author: Nathan Thomas # Date: 04/14/2015 while true do unset var var=$( ps aux | grep "openmcu-ru" | grep -cv "grep" 2>&1 ) if [ "${var}" -eq "1" ] ; then # It's aliiiiiiive...do nothing (bitchslap) : elif [ "${var}" -gt "1" ] ; then # Hmm, multiple instances but why? Comment out if your debugging OpenMCU # and there's a legitimate reason to have multiple processes running simultaneously # Otherwise, Squish bug! kill $( pgrep "openmcu-ru" 2&>1 ) 2&>1 /usr/bin/service openmcu-ru start 2>&1 else # Balls, it's dead /usr/bin/service openmcu-ru start 2>&1 fi sleep .5 done