I've been running a vfp site for many years. Recently I began having trouble with certain requests taking a long time to complete. I have tried several approaches to determining the cause. The odd thing is that I can't recreate the behaviour.
For example, I can log requests noting the time in and time out using fw_enter and fw_exit. However, whenever I issue the request myself the execution time is normal. I have tried executing the request from machines off the network through vpns and proxies around the world with the same result. No matter what I do I cannot recreate the problem.
It has nothing to do with how many channels are active at the time of the request. And there is no apparent correlation with any other server activity at the time, near as I can tell. The issue seems to be solely with requests coming from crawlers. I have yet to have a user experience the issue.
Nevertheless these requests can hog resources tying up foxweb channels for 30-45 seconds before exiting normally.
The problems don't seem to correlate to any particular type of page request. Defrag etc has no impact.
What other factors can cause a request to tie up a channel while eventually exiting normally? I guess I'm going to migrate the intalation to a different server to see if the problem carries... but that's a lot of work just for a hit and miss and even if it works it won't tell me the cause.
Any ideas? (I know it's kind of a vague issue)