I tried running that code on my CF 4.5.1sp2 box and all I could get was an error. Invalid format mask.
Then I ran the code below in CFMX (a trial ver I downloaded 3-4 days ago, no patches), which reruns itself once per second. Its at about 400 iterations at this point and still going. cpu usage looks to avg roughly 20%, although all I'm doing is eyeballing taskman, with the only memory increase being from my browser, which is slowly creeping up. <cfparam name="url.iteration" default="1"> <cfset url.iteration=url.iteration+1> <html><head><title></title> <CFOUTPUT><meta http-equiv="REFRESH" content="1; url=#cgi.script_name#?iteration=#url.iteration#"></CFOUTPUT> </head><body> <CFSET col1 = 123456789> <CFOUTPUT>#numberformat(replace(decimalformat(col1),",","","ALL"),"_.______")#</cfoutput> </body></html> --------------------------------------- Matt Robertson, MSB Designs, Inc. http://mysecretbase.com - Retail http://foohbar.org - ColdFusion Tools --------------------------------------- ---------- Original Message ---------------------------------- from: "Haggerty, Mike" <[EMAIL PROTECTED]> Reply-To: [EMAIL PROTECTED] date: Thu, 12 Dec 2002 14:15:08 -0500 >To everyone on this list; > >Big thanks for the help in reference to the 'C++ Pure Virtual Function Call' >thread that began on this list about two weeks ago. > >The solution was a ultimately a tricky one; I ended up upgrading the CF4.5.0 >box to CF5 and found I was able to eliminate 99% of the problems. But >performance was still sub-optimal, and there was one page that was popping >up on my error logs intermittently. > >I spent some time today going through that page line by line, and discovered >the following innoculous-looking lines of code (that the guy who owned this >project before me wrote): > ><CFSET col1 = 123456789> ><CFOUTPUT>#numberformat(replace(decimalformat(col1), ",", "", "ALL"), >"_.______")#</cfoutput> > >They kill CF server dead. After being loaded about 10 - 20 times, the code >will eat up all available memory and resources on the server (warning: >evaluate this statement at your own risk). > >I think the problem is with the numberformat being passed a string, but I am >not really sure. Anyone have any guesses as to why this code is so lethal >yet will not throw an error on its own? > >M > > > ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~| Archives: http://www.houseoffusion.com/cf_lists/index.cfm?forumid=4 Subscription: http://www.houseoffusion.com/cf_lists/index.cfm?method=subscribe&forumid=4 FAQ: http://www.thenetprofits.co.uk/coldfusion/faq Signup for the Fusion Authority news alert and keep up with the latest news in ColdFusion and related topics. http://www.fusionauthority.com/signup.cfm
