CFDump Issues

[kodemonki]

Today I'm trying to cfdump a parsed XML file, last week I was trying to CFDump a query response.  Today's error message:

The request has exceeded the allowable time limit Tag: cfoutput

Last week's message:

The request has exceeded the allowable time limit Tag: cfquery.

In both instances, when I remove the cfdump tag and just output as normal, it works fine and very fast. This is on my test server running CF8 on Windows Server 2003.  On my production server (CF8/Windows Server 2008) I get an error about java heap space.    CFDump does work on small data sets (like "select 1 from dual").  Any suggestions on how I can improve CFDump's performance?  Thanks!

[Adam Cameron.]

How much data are you trying to dump out?  Maybe try to reduce the amount you're dumping.

What's your request timeout?  Maybe disable it on dev.

How much memory is allocated to your heap?

Do you have debugging switched on?  Perhaps switch if off unless you need it.  It adds rather a large overhead onto every request, and it's just a waste of space most of the time.

--
Adam

[kodemonki]

I'm trying to dump a 12MB xml file.

The maximum java heap size is 1GB.

I increased the timeout to 1000000000000 for another test run, so we'll see how that goes.  After this next run is done I'll disable debugging and let you know how that goes.

[cfjedimaster]

12 megs of XML will be a large amount of HTML when done.

[Steve Sommers]

My first recommendation would be to not dump the entire xml -- I fail to see effenciency of visually sifting through a 12MB xlm. If you still need it, try adding requesttimeout=5000 to you URL query parameters.

[Reed_Powell-ttnmOb · Accepted solution]

The usual reason is that you're dealing with a new xml format and want to get a quick idea of how it is structured.  12MB isn't really all that much data for an XML file, and the nice thing about using CFDUMP is that you can see exactly what parts of it are structures and which are arrays.  An alternative:  save it to a file and use IE to look at it. 

-reed

[recursivecodes]

In that case I'd have to say just dump the xml string to the browser before parsing it...  Most browsers have built in XML pretty printing nowadays...

[12Robots]

It's not that 12mb is that large a file, but for CFDump to parse and display something that size requires a LOT of HTML and a lot of resources to process/build.  Dumps are slow, and the bigger they are, the slower they get.

[kodemonki]

I ended up finding a .csv of the data I wanted and just imported it that way.  IE kept crashing when trying to show the data, but Firefox worked fine.

Thanks for all your help!

[ilssac]

You might also want to use an alternative output to the <cfdump...> tag.

If you tell the <cfdump...> tag to just output "text" you don't get all the HTML, CSS and DHTML JavaScript that really blows up the HTML version.

Of course you also don't get all the pretty colored, expanding and collapsing boxes either.