Not sure, I'm not an admin by trade, but I was actually referring to the fact that some admins assume that user Perl processes that don't die, are doing so unintentionally.
I'm sure there may also be security issues with the user/socket coding as well. --------------------- -Tom Kinzer Long Beach, California -----Original Message----- From: drieux [mailto:[EMAIL PROTECTED] Sent: Wednesday, November 26, 2003 2:12 PM To: begin begin Subject: Sockets and Daemonizing - was Re: Count the no of times a script is called On Wednesday, Nov 26, 2003, at 13:28 US/Pacific, Tom Kinzer wrote: > drieux suggests: > >> The alternative of course would be to have >> the 'perl code' start up as a daemon that >> handled requests on a socket, this way you >> would save the 'start up' overhead of >> invoking a new process each time through. > > Make sure your sys admin is OK with this one. Good Point! p0: My UberGeek just smacked me upside the head that Fortran does not have a native Socket Library and that the standard trick would be to compile up some c code, a la <http://people.web.psi.ch/rohrer_u/sample2.htm> p1: You will forgive the bad habits of system programming, where 'hey kids, why not just spin up a daemon that will dish up services on a socket...' is just a part of solving the overall throughput issues. The idea had been to save on the overhead of repeated compile, open counter file, update counter file, do the required perl bit sequence each time the fortran wanted to make a call to it. p2: which are the specific issues worth underlining that a sys admin would have with a 'mere user' doing socket level coding? ciao drieux --- -- To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED] -- To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED]
