In my case I'm only interested in SLES11, where it seems to exhibit the behavior I'm looking for.. I'm just trying to make sure building the apr runtime to use flock doesn't have some negative/unforeseen impact on the modules I'm using. It sounds like Apache itself doesn't use apr_file_lock from the responses. So it's down to the modules that I'm using in my config. I can't see why (given no NFS), there would be a negative effect quite frankly,
Greg From: thomas bonfort Sent: Thursday, April 04, 2013 3:13 AM To: Greg Lazar Cc: dev@apr.apache.org Subject: Re: Building the apr runtime library appropriately for the worker MPM On 3 April 2013 21:03, Greg Lazar <gregla...@hotmail.com> wrote: The apr_file_lock routine, in the version of apr runtime library that I've currently built uses fnctl which does not appear to be thread safe. I need to use the worker MPM, so is it recommended that I build the apr runtime library to use flock instead. NFS file systems aren't a concern in this case. I didn't know how extensively the apr_file_lock was used through the rest of apache and if I'd be introducing any regressions by switching the apr runtime library to use flock. Thank you. Hi Greg, I had brought this question up on the modules-dev mailing list a few months ago: http://mail-archives.apache.org/mod_mbox/httpd-modules-dev/201109.mbox/%3CCAOM3y2i3osynN3Rnuvoqyhm1NHBgZTxxnt%2BT0y7nPz9tEpu6AA%40mail.gmail.com%3E Note that it would seem flock falls back to fcntl on some platforms, so using flock will not fix the issue on those platforms. regards, thomas
