SEGV in $bb->cleanup

2004-08-26 Thread Philippe M. Chiasson
I am trying to track down a strange segv in $bb->cleanup and it seems quite tricky to figure out what is the cause of it. Happens reliably for me on one system but not on another identical (well, almost) system. Can people here try this test patch on their systems and report back on success/failure

Re: time for 1.99_15?

2004-08-26 Thread Markus Wichitill
Stas Bekman wrote: Markus Wichitill wrote: t\perl\ithreads.Attempt to free temp prematurely: SV 0x6cc4a14, Perl interpreter: 0x6c90cfc at C:\Dev\debug\modperl-2.0\t\response/TestPerl/ithreads.pm line 44. request has failed (the response code was: 500) Seems to have crashe

Re: time for 1.99_15?

2004-08-26 Thread Stas Bekman
Markus Wichitill wrote: Markus Wichitill wrote: t\perl\ithreads.Scalars leaked: 1 Scalars leaked: 1 t\perl\ithreads.ok Just now I got this instead: t\perl\ithreads.Attempt to free temp prematurely: SV 0x6cc4a14, Perl interpre

Re: Apache lying on OpenBSD about socket blocking state

2004-08-26 Thread Stas Bekman
Ken Simpson wrote: Okay, I've figured out why Socket::opt_set is unable to make a socket blocking on OpenBSD. The APR library -- I can only guess in an attempt to be more efficient -- caches the state of each socket object so that it can avoid making unnecessary system calls. Before even trying to

Apache lying on OpenBSD about socket blocking state

2004-08-26 Thread Ken Simpson
Okay, I've figured out why Socket::opt_set is unable to make a socket blocking on OpenBSD. The APR library -- I can only guess in an attempt to be more efficient -- caches the state of each socket object so that it can avoid making unnecessary system calls. Before even trying to set the blocking