Batch processing issue 10.02

[Dandello2000]

I have a lot of Perl created pages on my websites and I test them using Wampserver on a WinXP machine before uploading them to the host server.
In CSE version 10.0120 and previous versions, I've had no problem with CSE validating pages with Perl derived includes (in SHTML pages) in batch mode on localhost.
But CSE 10.0203 gives me 500 errors (premature end of headers) on many, if not all, of these pages while in batch mode - these errors DO NOT appear when these same pages are checked in regular mode.
Any ideas?

[Albert Wiersch]

Sorry, I haven't heard of this issue before.

If you open one of the pages using File->Open from the Web, then does it work?

Is there a way I could reproduce this on my end?

It's also possible there is problem with your server/scripts, that CSE HTML Validator is now finding.

[Dandello2000]

If I use File->Open from the Web, all the Perl-driven pages work. And the problem seems to be oddly intermittent in batch mode. Out of ten pages, eight will have problems on one run, then four will have problems on the next run and a different four will have issues the third run and only one or so will fail on the fourth run.

If this was a Perl script issue, these pages would be failing all the time.

I'm wondering if it's an issue of CSE not always giving enough time for the scripts to finish loading. (Some of these are fairly hefty databases.)

If this is the case, where do I change the length of time for CSE to wait for the server response while in batch mode?

[Albert Wiersch]

If this is the case, where do I change the length of time for CSE to wait for the server response while in batch mode?

Hello,

You can slow down the Batch Wizard in the Limits page in the Batch Wizard Options. Please try setting the "Wait time before each URL request" to 1, or lower the "Max concurrent HTTP requests" and let me know if that helps.

Thanks!

[Dandello2000]

Setting the wait time to 1 seems to have done the trick.

[Albert Wiersch]

Setting the wait time to 1 seems to have done the trick.

Great!