I'd be using asych CFC's on a gateway for all my logging if I had a copy of Enterprise...
As it is, I'm using WDDX and the application scope... granted, it's not all that efficient in terms of memory use, but it's not a terribly high-traffic system and at the moment it's got horsepower to spare. I use the WDDX file to write the application-scope structure to disk... I use a struct to keep track of hits, visitors, and IPs. Right now, I just checked the page and I'm up to 1,000 visitors and 14,000+ page views over the course of the last 3 months. It's a portal for a labor organization, so their usage is limited to members... but, that means that 1000 out of 1200 users has accessed the site! That's some pretty deep penetration. ;) But... this little WDDX/script widget actually writes the cache to disk on every request... it keeps it in RAM so it only has to read it from disk if a failure occurs. I debated doing this on every hit... but for cryin out loud... think about Apache logs, or any other OS or server log. Writing a small amount of text to disk is a rediculously small cost... But there's about 3 reasons I want to get my hands on CFMX7 for that application... application.cfc would make a lot of this easier... esp things like onApplicationEnd for flushing the RAM-cached counter and poll to disk... Enterprise would give me a gateway for async CFCs (which, incidentally, is one of the target uses for Tartan... it's a fantastic interface for async CFC gateways)... and the reports/printable documents would make those portions a lot simpler too. Anyway... I've been using Barney's suggestion for about 8 months and it's fab. I'd BE using your idea if I had a copy of Enterprise to do it with. Laterz, J On Wed, 16 Mar 2005 19:06:06 -0500, Michael Dinowitz <[EMAIL PROTECTED]> wrote: > > > But a crash of CF/Machine will still result in loss of data > > > > Does that happen often? I can't remember the last time it's happend > > to me (at least a couple years). We rarely cycle down CF (maybe 6-8 > > times a year), but that's of more concern than random crashes, and > > Application.cfc addresses the issue. There's always a chance for data > > loss, just the window is larger with my example. > I'm paranoid and always assume something will go wrong. > > > > scheduler in CF is based on a CFHTTP call which is basically a double > > > overhead. > > > > Yeah, that's true. You could dispatch the request from cron, if it > > matters much. However, under load one extra request per minute is > > insignificant, and if you're not under load then it's also > > insignificant. Doesn't seem like much of a concern. > I just don't like running extra processes if I can help it and while I have > had great success with the scheduler and scheduled tasks from outside CF, > others have not. Still, it's a valid and viable operation. Want to write up > a paper on if for Fusion Authority? :) > -- Continuum Media Group LLC Burnsville, MN 55337 http://www.web-relevant.com http://cfobjective.neo.servequake.com ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~| Find out how CFTicket can increase your company's customer support efficiency by 100% http://www.houseoffusion.com/banners/view.cfm?bannerid=49 Message: http://www.houseoffusion.com/lists.cfm/link=i:4:199060 Archives: http://www.houseoffusion.com/cf_lists/threads.cfm/4 Subscription: http://www.houseoffusion.com/lists.cfm/link=s:4 Unsubscribe: http://www.houseoffusion.com/cf_lists/unsubscribe.cfm?user=89.70.4 Donations & Support: http://www.houseoffusion.com/tiny.cfm/54
