Markus Wichitill wrote:
Stas Bekman wrote:

It seemed completely consistent to me. "nmake test" with mod_include always leaks, "nmake test" without mod_include and partial tests don't leak. Of course the long runtime of full tests has limited the number of tests I could run.


But does the test fail or not? If it doesn't then t/SMOKE won't help here.


No, it didn't ever actually fail.

t\perl\ithreads.........................Scalars leaked: 1
Scalars leaked: 1
t\perl\ithreads.........................ok

And since the FATAL switch is already used, I guess those warnings can't be easily turned into failures?

No. Normally those happen after the handler has been run already. How come do you get them on the client side? Is the client leaking those? (doesn't make sense). Or is it something specific to win32, where server logs end up on the client side? Do you have those leak logs in error_log too?


--
__________________________________________________________________
Stas Bekman            JAm_pH ------> Just Another mod_perl Hacker
http://stason.org/     mod_perl Guide ---> http://perl.apache.org
mailto:[EMAIL PROTECTED] http://use.perl.org http://apacheweek.com
http://modperlbook.org http://apache.org   http://ticketmaster.com

---------------------------------------------------------------------
To unsubscribe, e-mail: [EMAIL PROTECTED]
For additional commands, e-mail: [EMAIL PROTECTED]



Reply via email to