Hello,
If a process exits normally the PID file will be removed, you are
making the kernel kill the process by using `kill -9`.
You should not be using kill -9 as a norm. You should shutdown the
application according to the documentation provided with the app.
If there is no specify way then a standard `kill` command should do not
the trick.
If PID files are left behind usually this is because the process was
shutdown correctly. This is not a problem, most applications will just
over write the PID file the next time they start up.
How ever some apps will complain and then you would be required to
remove the PID file.
If you pay closer attention to the files in /var/run and
/var/lock/subsys you will notice that all the files in /var/run/ are
about 5k and contain a number, which is the PID.
The files located in /var/lock/subsys will be empty :) because there are
there as place holders / locks so to speak. When an application is
running it can us this directory as for file locking. For example If I
ran passwd and it needed to lock the file until I was done (which it
does not) then it would touch a file called passwd in /var/lock/subsys
and if another user tried to run passwd it would not let me make changes
because the lock file exist.
Michael
Stephen Cartwright wrote:
But why are the files in /var/lock/subsys on RH and (RH based distros)
created given that there is already pid files in /var/run/ which can
be used to tell whether or not daemon is already running. The lock
files in /var/lock/subsys don't seem to lock a resource, they just
seem to indicate that the process is running... but then why not just
use the pid file?
Also what happens if the process dies unexpectedly? How does it work
with pid files? Is the creation and removal of the pid file done by
the OS? Is /var/run checked for old pid files? I used "kill -9" to
kill a process and the pid file was still there.
Thank you!
On 6/27/05, Robert Lewko <[EMAIL PROTECTED]> wrote:
On June 27, 2005 10:47 am, Stephen Cartwright wrote:
What exactly is the difference between a pid file and a subsystem lock
file? Why do you need both of them?
Thanks!
A *.pid file is a file that keeps the PID "process id" and that is to record
which process to send signals to for various reasons, ie. killing the process
(thats not the only reason, but the most common one).
What a lock file is for is to signal that a particular resource is in use
_______________________________________________
clug-talk mailing list
[email protected]
http://clug.ca/mailman/listinfo/clug-talk_clug.ca
Mailing List Guidelines (http://clug.ca/ml_guidelines.php)
**Please remove these lines when replying
