David C. Rankin wrote: > You can look at: > > Quick-and-dirty way to ensure only one instance of a shell script is > running at a time > https://stackoverflow.com/questions/185451/quick-and-dirty-way-to-ensure-only-one-instance-of-a-shell-script-is-running-at > > Whether you are starting the application from the shell or calling exec > from within an application, the process will be the same. > > That is the basic reason for the Linux /run directory. At its simplest, > form, to ensure only one instance is running you simply check whether a > pidfile exists in some directory, and if not start the app and write the > pid of the application to the pidfile. This provides the benefit of being > able to simply read the pidfile to get the pid of the current running > instance. > Yes I looked at /run. All of these scripting examples are not appropriate for the use case I have - in fact I find them ugly - I do not want to write pid files or lock files anywhere. TDE works very well with KUniqueApplication, when forked. The problem is after it is forked, I loose control over it from the application that started the process. I am also thinking of adding a dbus method stop() as the application is already present in dbus(). To me it looks the easiest way to control it. --------------------------------------------------------------------- To unsubscribe, e-mail: trinity-devel-unsubscribe@xxxxxxxxxxxxxxxxxxxxxxxxxx For additional commands, e-mail: trinity-devel-help@xxxxxxxxxxxxxxxxxxxxxxxxxx Read list messages on the web archive: http://trinity-devel.pearsoncomputing.net/ Please remember not to top-post: http://trinity.pearsoncomputing.net/mailing_lists/#top-posting
