It almost certainly has something to do with your process failing to daemonize properly. Ruby processes struggle too, check http://stackoverflow.com/a/688448/119669 it has to do with parts of the process that is forked (your python daemon in this case) still inheriting some resources which are attached to the Capistrano session.
This unfortunately falls outside stuff I can help you with reasonably or remotely. You *might* have some success learning enough strace to see your process, and how it behaves when Cap disconnects. Lee Hambley http://lee.hambley.name/ +49 (0) 170 298 5667 On 19 May 2015 at 22:23, niristotle okram <nirish.ok...@gmail.com> wrote: > I apologize, i made a mistake in reporting the issue. :( > > The original report about " xyz.pid doesn't exist, when it actually does." > is not an ERROR. Its working as design. The service was manually stopped > before trying to stop again. And that is why it threw that error. The issue > here is, When the capistrano starts the service as a part of the task > "deploy:restart_app" > after the deploy, the service doesn't start up fine. Checking the status > "service xyz status" after the deploy returns "Process dead but pid > exists". > > This behaviors of service start failure is only seen post the cap deploy > and cannot be reproduced manually. > > > > > > > > > > On Tuesday, May 19, 2015 at 12:10:50 AM UTC-7, Lee Hambley wrote: >> >> Sorry, I can't see anything wrong with it. :-\ >> >> Lee Hambley >> http://lee.hambley.name/ >> +49 (0) 170 298 5667 >> >> On 19 May 2015 at 02:26, niristotle okram <nirish...@gmail.com> wrote: >> >>> hi Lee, >>> >>> here is the full /etc/init.d/ script http://pastebin.com/02G5tpgH >>> >>> >>> So, i placed a task to stop the app, then deploy & then start the >>> service/app. The task have the below commands to check >>> >>> whoami -----> ** [out :: server4] root >>> ll /var/run/ ---> this shows the xyz.pid file >>> >>> So i see the service stops just fine. >>> >>> ** [out :: server1] Stopping >>> >>> ** [out :: server1] Ok >>> >>> >>> >>> And the service also starts just fins >>> >>> >>> ** [out :: server1] Starting xyz... >>> >>> ** [out :: server1] Ok >>> >>> >>> >>> But on checking manually -- "service xyz status", i get this " Process >>> dead but pidfile exist". I can stop and start the service just fine >>> manually. >>> >>> >>> >>> >>> >>> >>> >>> >>> >>> >>> >>> >>> On Monday, May 18, 2015 at 12:24:50 PM UTC-7, Lee Hambley wrote: >>>> >>>> You wrote that the /etc/init.d/xyz is done by "sudo" so the deploy user >>>> apparently has access to password-less sudo (at least for some actions), it >>>> would appear that the file is not visible to `root`. Which I don't believe >>>> or expect. >>>> >>>> You included a part of the /etc/init.d/xyz, but didn't include the full >>>> thing for some reason, so I can't see what the value of $PIDFILE >>>> should be in this case (*please*, adhere to the list guidelines and >>>> paste long files in an external service, and link them), nor state where >>>> you got the template. >>>> >>>> I also don't understand the logic behind setting shell: 'bash' on the >>>> run() lines that interface with the init script. >>>> >>>> Your task: >>>> >>>> task :stop_app, :roles => :web do >>>> >>>> run "sudo /etc/init.d/xyz stop", :shell => :bash >>>> >>>> end >>>> >>>> >>>> I might suggest you extend that (or make a similar one, >>>> "debug_initd_stuff") that does something like: >>>> >>>> task :debug_initd_stuff, :roles => :web do >>>> >>>> run "sudo whoam" >>>> >>>> run "sudo ls -l /etc/init.d" >>>> >>>> run "sudo ls -l /var/run" >>>> >>>> end >>>> >>>> >>>> You might also want to run the init.d script through shellcheck.net, >>>> since there are quite a few violations and bad practices already in sight >>>> there, shellcheck might help you iron some of them out. (That said, >>>> honestly the problem is probably something much simpler.) >>>> >>>> >>>> Lee Hambley >>>> http://lee.hambley.name/ >>>> +49 (0) 170 298 5667 >>>> >>>> On 18 May 2015 at 21:14, niristotle okram <nirish...@gmail.com> wrote: >>>> >>>>> Versions: >>>>> >>>>> - Ruby 2.1.1 >>>>> - Capistrano 2 >>>>> - Rake / Rails / etc >>>>> >>>>> Platform: >>>>> >>>>> - Working on.... RHEL 6 >>>>> - Deploying to... RHEL 6 >>>>> >>>>> >>>>> A part of the Deploy.rb: >>>>> >>>>> #before "deploy", "deploy:stop_app" >>>>> >>>>> #after "deploy", "deploy:start_app" >>>>> >>>>> after "deploy", "deploy:restart_app" >>>>> >>>>> namespace :deploy do >>>>> >>>>> task :update_code, :roles => :web, :except => { :no_release => >>>>> true } do >>>>> >>>>> on_rollback { puts "DO NOT WANT TO ROLL BACK?" } >>>>> >>>>> strategy.deploy! >>>>> >>>>> finalize_update >>>>> >>>>> end >>>>> >>>>> >>>>> >>>>> task :stop_app, :roles => :web do >>>>> >>>>> run "sudo /etc/init.d/xyz stop", :shell => :bash >>>>> >>>>> end >>>>> >>>>> >>>>> >>>>> task :start_app, :roles => :web do >>>>> >>>>> run "sudo /etc/init.d/xyz start", :shell => :bash >>>>> >>>>> end >>>>> >>>>> >>>>> >>>>> >>>>> >>>>> task :restart_app, :roles => :web do >>>>> >>>>> run "sudo /etc/init.d/xyz restart", :shell => :bash >>>>> >>>>> end >>>>> >>>>> end >>>>> >>>>> >>>>> >>>>> >>>>> >>>>> >>>>> >>>>> >>>>> >>>>> I have the parameter in the 'deploy.rb', >>>>> >>>>> *set :user, 'a_user'* >>>>> >>>>> Q: Which user performs the task to restart the service (xyz) after the >>>>> deployment of app (xyz)? I am getting the errors saying the xyz.pid >>>>> doesn't >>>>> exist, when it actually does. This is a part of the shell script while >>>>> stopping the service. >>>>> >>>>> >>>>> A part of the /etc/init.d/xyz >>>>> >>>>> case "$1" in >>>>> start) >>>>> printf "%-50s" "Starting $DAEMON_NAME..." >>>>> cd $DIR >>>>> [ -d $LOGPATH ] || mkdir $LOGPATH >>>>> [ -f $LOGFILE ] || su $DAEMON_USER -c 'touch $LOGFILE' >>>>> PID=`$PYTHON $DAEMON $DAEMON_OPTS > $LOGFILE 2>&1 & echo $!` >>>>> #echo "Saving PID" $PID " to " $PIDFILE >>>>> if [ -z $PID ]; then >>>>> printf "%s\n" "Fail" >>>>> else >>>>> echo $PID > $PIDFILE >>>>> printf "%s\n" "Ok" >>>>> fi >>>>> ;; >>>>> status) >>>>> printf "%-50s" "Checking $DAEMON_NAME..." >>>>> if [ -f $PIDFILE ]; then >>>>> PID=`cat $PIDFILE` >>>>> if [ -z "`ps axf | grep ${PID} | grep -v grep`" ]; then >>>>> printf "%s\n" "Process dead but pidfile exists" >>>>> else >>>>> echo "Running" >>>>> fi >>>>> else >>>>> printf "%s\n" "Service not running" >>>>> fi >>>>> ;; >>>>> stop) >>>>> printf "%-50s" "Stopping $DAEMONNAME" >>>>> PID=`cat $PIDFILE` >>>>> cd $DIR >>>>> if [ -f $PIDFILE ]; then >>>>> kill -HUP $PID >>>>> printf "%s\n" "Ok" >>>>> rm -f $PIDFILE >>>>> else >>>>> printf "%s\n" "pidfile not found" >>>>> fi >>>>> ;; >>>>> >>>>> restart) >>>>> $0 stop >>>>> $0 start >>>>> ;; >>>>> >>>>> *) >>>>> echo "Usage: $0 {status|start|stop|restart}" >>>>> exit 1 >>>>> esac >>>>> >>>>> >>>>> >>>>> Capistrano log >>>>> >>>>> * executing `deploy:restart_app' >>>>> >>>>> * executing multiple commands in parallel >>>>> >>>>> -> "else" :: "sudo /etc/init.d/xyz restart" >>>>> >>>>> -> "else" :: "sudo /etc/init.d/xyz restart" >>>>> >>>>> -> "else" :: "sudo /etc/init.d/xyz restart" >>>>> >>>>> -> "else" :: "sudo /etc/init.d/xyz restart" >>>>> >>>>> servers: ["server1", "server2", "server3", "server4"] >>>>> >>>>> [server1] executing command >>>>> >>>>> [server2] executing command >>>>> >>>>> [server3] executing command >>>>> >>>>> [server4] executing command >>>>> >>>>> ** [out :: server1] Stopping >>>>> >>>>> ** [out :: server1] cat: /var/run/xyz.pid: No such file or directory >>>>> >>>>> ** [out :: server1] pidfile not found >>>>> >>>>> ** [out :: server1] Starting xyz... >>>>> >>>>> ** [out :: server2] Stopping >>>>> >>>>> ** [out :: server2] cat: /var/run/xyz.pid: No such file or directory >>>>> >>>>> ** [out :: server2] pidfile not found >>>>> >>>>> ** [out :: server2] Starting xyz... >>>>> >>>>> ** [out :: server2] Ok >>>>> >>>>> ** [out :: server1] Ok >>>>> >>>>> ** [out :: server3] Stopping >>>>> >>>>> ** [out :: server3] cat: /var/run/xyz.pid: No such file or directory >>>>> >>>>> ** [out :: server3] pidfile not found >>>>> >>>>> ** [out :: server4] Stopping >>>>> >>>>> ** [out :: server4] Ok >>>>> >>>>> ** [out :: server3] Starting xyz... >>>>> >>>>> ** [out :: server3] Ok >>>>> >>>>> ** [out :: server4] Starting xyz... >>>>> >>>>> ** [out :: server4] Ok >>>>> >>>>> command finished in 659ms >>>>> >>>>> Finished: SUCCESS >>>>> >>>>> >>>>> >>>>> >>>>> I can cat the file as the deploy user just fine. >>>>> >>>>> >>>>> >>>>> >>>>> -- >>>>> You received this message because you are subscribed to the Google >>>>> Groups "Capistrano" group. >>>>> To unsubscribe from this group and stop receiving emails from it, send >>>>> an email to capistrano+...@googlegroups.com. >>>>> To view this discussion on the web, visit >>>>> https://groups.google.com/d/msgid/capistrano/56a2a2dd-fd26-4b14-a2da-0d7af37f8354%40googlegroups.com >>>>> <https://groups.google.com/d/msgid/capistrano/56a2a2dd-fd26-4b14-a2da-0d7af37f8354%40googlegroups.com?utm_medium=email&utm_source=footer> >>>>> . >>>>> For more options, visit https://groups.google.com/d/optout. >>>>> >>>> >>>> -- >>> You received this message because you are subscribed to the Google >>> Groups "Capistrano" group. >>> To unsubscribe from this group and stop receiving emails from it, send >>> an email to capistrano+...@googlegroups.com. >>> To view this discussion on the web, visit >>> https://groups.google.com/d/msgid/capistrano/d47d5020-1915-4194-be85-b72e157b0c23%40googlegroups.com >>> <https://groups.google.com/d/msgid/capistrano/d47d5020-1915-4194-be85-b72e157b0c23%40googlegroups.com?utm_medium=email&utm_source=footer> >>> . >>> >>> For more options, visit https://groups.google.com/d/optout. >>> >> >> -- > You received this message because you are subscribed to the Google Groups > "Capistrano" group. > To unsubscribe from this group and stop receiving emails from it, send an > email to capistrano+unsubscr...@googlegroups.com. > To view this discussion on the web, visit > https://groups.google.com/d/msgid/capistrano/bdf4409b-4f59-42f2-be66-2bb4f895dbfe%40googlegroups.com > <https://groups.google.com/d/msgid/capistrano/bdf4409b-4f59-42f2-be66-2bb4f895dbfe%40googlegroups.com?utm_medium=email&utm_source=footer> > . > > For more options, visit https://groups.google.com/d/optout. > -- You received this message because you are subscribed to the Google Groups "Capistrano" group. To unsubscribe from this group and stop receiving emails from it, send an email to capistrano+unsubscr...@googlegroups.com. To view this discussion on the web, visit https://groups.google.com/d/msgid/capistrano/CAN_%2BVLXjwR1QFzupAak1khCVvskxaVPmk1PhNRA6pzbjwZC5EQ%40mail.gmail.com. For more options, visit https://groups.google.com/d/optout.
