Looking for any intel on known issues.
- did not want to drag anyone into debugging for us but wanted to mention the item in case it strikes a chord
Copy link to clipboard
Where are you seeing/tracking the delay?
Browser dev tools confirm and quantify the unusually slow cfm page rendering. Other pages outside of cf are unaffected by this significant delay. Oddly this periodically renders as actual timeout errors.
If this doesn't ring a bell for any known issues...we will keep digging.
(If we find the reason we will post here.)
To start with, you're comparing two different things. ColdFusion processes CFM pages, but does not process HTML pages.
There could be any number of reasons to explain the delay, but as @BKBK has pointed out, you are comparing apples and SUVs. HTML pages are sent directly to the browser for parsing; CFML pages build HTML on the fly and hands that to the web server to send to the browser.
Is this a new install of CF Server? If not, how long has the server been operating, and how recently has the delay started becoming noticeable? Have you tried altering configuration (how much memory is allowed to the server, etc.)? How much memory does the physical server have? CPU speed and number of cores? RAM type? The list goes on and on. Please help us help you by providing the necessary information for us to analyze.
^ _ ^
Hi Wolfshade & bkbk
- this is a new install.
- just mentioning html page response as a point of reference
- other cf servers we run have only slightly slower response time when comparing cfm to htm
(normally we would see 60-80ms on a cfm which is 10x performance drop)
- i was truly just checking for 'known issues'
- if nothing comes to mind we will keep debugging/tuning or reinstall
I would run this test for a quick estimate of the CFM page processing time:
<!--- At top of page --->
<cfset pageStart = getTickCount()>
Contents of CFM page
<!--- At bottom of page --->
<cfset timeInterval = getTickCount()-pageStart>
<cflog text="Duration: #timeInterval# milliseconds" file="pageProcessingTest" >
BKBK - thanks for the reminder to isolate actual processing time
- we zeroed in on a 0ms processing
- found our test was grabbing gunk in application.cfm (which we over looked)
Operator Error. Doh.
Application.cfm - good find, @cffiddler. Thanks for sharing it.