https://bz.apache.org/bugzilla/show_bug.cgi?id=66316
--- Comment #4 from Seann Herdejurgen <se...@herdejurgen.com> --- I have a production environment and a QA environment. In production, DBM data is stored in a tmpfs file system instead of hark disk. With a tmpfs file system, I am unable to reproduce deadlock using mod_perl's PerlResponseHandler ModPerl::Registry. I have been running ModPerl::Registry using DBM data on tmpfs file system in production for two days without a deadlock. If I implement storing DBM data in tmpfs file system in QA I do not see deadlock. If I revert to hard disk then I can reproduce deadlock within a couple of minutes. This implies to me that whatever Perl is waiting for in the deadlock situation is related to the speed of the storage subsystem. -- You are receiving this mail because: You are the assignee for the bug. --------------------------------------------------------------------- To unsubscribe, e-mail: bugs-unsubscr...@httpd.apache.org For additional commands, e-mail: bugs-h...@httpd.apache.org
