Am Freitag, 1. April 2005 10:58 schrieb Philipp Kern: > Original submitter wrote: > > licq fails to start when its PID file (~/.licq/licq.pid) exists, and a > > process with that PID exists, even if that process isn't licq. I > > believe a > > simple check that the process has the right name (and owner!!) is in > > order. > > System tools like ``start-stop-daemon'' check all those properties on a > pid before even considering that it is already running, so this might > be valid.
I know that on Linux systems you can find out the process behind a PID by using the proc file system. The bug submitter has, however, requested that licq handle that, and that would necessarily have to be done in some platform-independent fashion. Add to that that licq is a user process and therefore might not have full access to the proc file system. Therefore, I think that this issue cannot be reasonably solved. I had requested that the submitter come up with an idea to approach this problem, but since there was no response, I assume that he has given up. > Package maintainer wrote: > > Since no solution was proposed for this bug, I close it. You can > > reopen > > it if you have additional information. > > Couldn't you just submit it upstream as a bug and tag it ``upstream''? > Or ``help''. Sure, the problem is easy to fix for the experienced and > with a reboot. But the average user does not know why licq won't start > up. I understand that the situation is undesirable, but to my knowledge it cannot be solved. It won't help if I just shove the problem upstream. I invite you, if you have a better idea, to forward this bug and reopen and tag it appropriately. -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
