08-05-2005
SSH starting nohup'd process - not exiting
I'm trying to ssh into a remote server, run a script which may or may not start a nohup'd background process and then exit leaving the process running on the remote server. I'm looping through a number of servers to do this but the script hangs as soon as it comes to a server where the remote process needs to be started. It seems that it's a limitation of ksh that you get a report of background processes running when you exit from the shell.
code:
if [ ${running} -eq 0 -o ${newdate} -eq 0 ];then
nohup ${bin_base}/${stream}/bin/xacct_viewlog Gatherer 500 >> ${stream_home}/${logfilename} &
echo " I have restarted with process ID: $!"
echo $! > ${pid_file}
echo $currentdate > ${datefile}
echo " Exiting having restarted ..."
fi
so far so good, this works fine when run as a shell script from the command line on the box.
When I run it from a remote host using the following script:
package_list="server1 server2 server3"
kick_me ()
{
set -x
server=$1
for stream in $stream_list
do
ssh -q $server -l xacct /var/tmp/oas_xacct_log_flatten.sh $stream
done
}
for server in $package_list
do
case "$server" in
medG1oat|medG1)
stream_list="stream1 stream2 stream3"
kick_me $server
;;
esac
done
I have simplified this script but it loops through several clients and several streams on each client running the shell script on the remote client to kick off the background process. The problem is that when it gets to the first process it actually has to start (the script checks to see if an instance of itself is already running and if it is already running it goes onto the next in the loop) then it kicks off the process and sits there. Looking through the man pages for ksh it seems that the shell won't detach while there are running jobs despite the job being kicked off as nohup. I have tried changing the code on the client playing with and without nohup etc and no joy.
Any ideas?
I leave this contract at 4pm to emigrate to Australia so this is mainly for the benefit of my boss (he's actually a good bloke) but it grinds my privates that i cannot solve this.
10 More Discussions You Might Find Interesting
1. Shell Programming and Scripting
Hello all,
How do I start a background process and save the process id to a file on my system. For example %wait 5 & will execute and print the process id. I can't figure out how to get it to a file. I've tried: > filename 0>filename 1>filename.
Any assistance is most appreciated.
Thanks,
Jim... (10 Replies)
Discussion started by: jleavitt
10 Replies
2. UNIX for Dummies Questions & Answers
Hi,
I run a binary application with GUI accessibility. To launch and close the application i follow the following steps:
1.Log into a console session. Export display to the local workstation.
2.Launch X windows app ( eXceed )
3.From terminal session go to the my application directory and... (2 Replies)
Discussion started by: shantaputi
2 Replies
3. Shell Programming and Scripting
I want to print the pid of a nohup process to a file so later I can use the list of pid's in that file to stop the background processes again.
I use ksh on AIXv5.3:
nohup /start/script.ksh 1>/dev/null 2>&1
print $$ > .pid
nohup /start/script2.ksh 1>/dev/null 2>&1
print $$ >> .pid
But... (2 Replies)
Discussion started by: rein
2 Replies
4. Shell Programming and Scripting
Hello,
I need to run a command on remote Linux using the ssh command from my local machine. I am able to execute the command on remote machine using ssh but it's behaving strangely.
The command is supposed to start a daemon process on remote linux box and the control should return back to me... (5 Replies)
Discussion started by: nitinshukla
5 Replies
5. Shell Programming and Scripting
Hey all, I need to launch a script from within 2 other scripts that can run independently of the two parent scripts... Im having a hard time doing this, if anyone knows how please let me know.
More detail.
ScriptA (bash), ScriptB (ksh), ScriptC (bash)
ScriptA, launches ScriptB
ScirptB,... (7 Replies)
Discussion started by: trey85stang
7 Replies
6. Shell Programming and Scripting
I have a script I'm creating to spawn netcat listeners on a remote server for copying files at high speeds. The issue I'm running into is that after running the "nohup nc -l -p 12345 | tar -xvf - &" commands I can't get the remote shell to terminate. I'm not sure if this is working as intended or... (2 Replies)
Discussion started by: headcr4sh
2 Replies
7. Shell Programming and Scripting
I have a fairly simple script like so:
nohup java -jar program.jar >> log 2>&1 &
echo $! > pidfile
This works great when using ssh server "script.sh" but not when using ssh -t server "script.sh"
The solution I came up with was to call sleep after recording the child pid. My guess at the... (2 Replies)
Discussion started by: croachrose
2 Replies
8. AIX
I could not able to kill two process which is running in the required port for me.Can any body help me to kill the exiting process.
- 27000908 - - - <exiting>
- 30998528 - - - <exiting> (8 Replies)
Discussion started by: sasikanta
8 Replies
9. Shell Programming and Scripting
Hi
I have a script which we are calling as below from another script but it does not exit out and just hang there.
main.ksh
#!/bin.ksh
echo "test1"
test.ksh | tee /tmp/abc.txt
echo "test2"
it prints test1 and then just hangs on test.ksh | tee /tmp/abc.txt
if i check the messages from... (3 Replies)
Discussion started by: reldb
3 Replies
10. Shell Programming and Scripting
Hi,
i have a shellscript, where in i need to connect to different server start these three jobs and exit from the server and start the same three jobs on local server.
ssh user@remotehost 'bash -s' << EOF
${GETT_HOME}/bin/start1 &
sleep 10
${GETT_HOME}/bin/start2 &
sleep 10... (1 Reply)
Discussion started by: swapnabhargav
1 Replies
XINIT(1) General Commands Manual XINIT(1)
NAME
xinit - X Window System initializer
SYNOPSIS
xinit [ [ client ] options ... ] [ -- [ server ] [ display ] options ... ]
DESCRIPTION
The xinit program is used to start the X Window System server and a first client program on systems that cannot start X directly from
/etc/init or in environments that use multiple window systems. When this first client exits, xinit will kill the X server and then termi-
nate.
If no specific client program is given on the command line, xinit will look for a file in the user's home directory called .xinitrc to run
as a shell script to start up client programs. If no such file exists, xinit will use the following as a default:
xterm -geometry +1+1 -n login -display :0
If no specific server program is given on the command line, xinit will look for a file in the user's home directory called .xserverrc to
run as a shell script to start up the server. If no such file exists, xinit will use the following as a default:
X :0
Note that this assumes that there is a program named X in the current search path. However, servers are usually named Xdisplaytype where
displaytype is the type of graphics display which is driven by this server. The site administrator should, therefore, make a link to the
appropriate type of server on the machine, or create a shell script that runs xinit with the appropriate server.
Note, when using a .xserverrc script be sure to ``exec'' the real X server. Failing to do this can make the X server slow to start and
exit. For example:
exec Xdisplaytype
An important point is that programs which are run by .xinitrc should be run in the background if they do not exit right away, so that they
don't prevent other programs from starting up. However, the last long-lived program started (usually a window manager or terminal emula-
tor) should be left in the foreground so that the script won't exit (which indicates that the user is done and that xinit should exit).
An alternate client and/or server may be specified on the command line. The desired client program and its arguments should be given as
the first command line arguments to xinit. To specify a particular server command line, append a double dash (--) to the xinit command
line (after any client and arguments) followed by the desired server command.
Both the client program name and the server program name must begin with a slash (/) or a period (.). Otherwise, they are treated as an
arguments to be appended to their respective startup lines. This makes it possible to add arguments (for example, foreground and back-
ground colors) without having to retype the whole command line.
If an explicit server name is not given and the first argument following the double dash (--) is a colon followed by a digit, xinit will
use that number as the display number instead of zero. All remaining arguments are appended to the server command line.
EXAMPLES
Below are several examples of how command line arguments in xinit are used.
xinit This will start up a server named X and run the user's .xinitrc, if it exists, or else start an xterm.
xinit -- /usr/local/bin/Xqdss :1
This is how one could start a specific type of server on an alternate display.
xinit -geometry =80x65+10+10 -fn 8x13 -j -fg white -bg navy
This will start up a server named X, and will append the given arguments to the default xterm command. It will ignore .xinitrc.
xinit -e widgets -- ./Xsun -l -c
This will use the command .Xsun -l -c to start the server and will append the arguments -e widgets to the default xterm command.
xinit /usr/ucb/rsh fasthost cpupig -display ws:1 -- :1 -a 2 -t 5
This will start a server named X on display 1 with the arguments -a 2 -t 5. It will then start a remote shell on the machine
fasthost in which it will run the command cpupig, telling it to display back on the local workstation.
Below is a sample .xinitrc that starts a clock, several terminals, and leaves the window manager running as the ``last'' application.
Assuming that the window manager has been configured properly, the user then chooses the ``Exit'' menu item to shut down X.
xrdb -load $HOME/.Xresources
xsetroot -solid gray &
xclock -g 50x50-0+0 -bw 0 &
xload -g 50x50-50+0 -bw 0 &
xterm -g 80x24+0+0 &
xterm -g 80x24+0-0 &
twm
Sites that want to create a common startup environment could simply create a default .xinitrc that references a site-wide startup file:
#!/bin/sh
. /usr/local/lib/site.xinitrc
Another approach is to write a script that starts xinit with a specific shell script. Such scripts are usually named x11, xstart, or
startx and are a convenient way to provide a simple interface for novice users:
#!/bin/sh
xinit /usr/local/lib/site.xinitrc -- /usr/local/bin/X -br
ENVIRONMENT VARIABLES
DISPLAY This variable gets set to the name of the display to which clients should connect.
XINITRC This variable specifies an init file containing shell commands to start up the initial windows. By default, .xinitrc in the
home directory will be used.
FILES
.xinitrc default client script
xterm client to run if .xinitrc does not exist
.xserverrc default server script
X server to run if .xserverrc does not exist
SEE ALSO
X(7), startx(1), Xserver(1), xterm(1)
AUTHOR
Bob Scheifler, MIT Laboratory for Computer Science
X Version 11 xinit 1.1.1 XINIT(1)