Flex failure during apache 1.3.28 make

2003-07-21 Thread Boyle Owen
Greetings, I'm trying to compile the new 2.8.15 with apache 1.3.28 but hit a problem when make tries to run flex on the file src/modules/ssl/ssl_expr_scan.l. I'm running Solaris 8 on a Sparc and flex is version 2.4.7. Up until now, I've always regarded flex as one of those mysterious little

Re: Flex failure during apache 1.3.28 make

2003-07-21 Thread Darin Holloway
I've found that the Sun versions of certain utilities (sed in my case, tar for lots of people) have problems with code that works fine with GNU utilities, my advice would be to install the gnu version of most utilities in /usr/local/bin or in /opt/bin and either use them as the first in your

Re: Flex failure during apache 1.3.28 make

2003-07-21 Thread R. DuFresne
wasn't this an issue with a modssl version a year or two ago? something like the source files in the tarball not having the proper date stamps and as Mad's mentiones, required a touch of a few files to make flex more 'flexable'? Thanks, Ron DuFresne On Mon, 21 Jul 2003, Mads Toftum wrote:

Re: [ANNOUNCE] mod_ssl 2.8.15 for Apache 1.3.28

2003-07-21 Thread Igor Sysoev
On Fri, 18 Jul 2003, Ralf S. Engelschall wrote: As you've certainly recognized, Apache 1.3.28 was released. I've prepared the companion mod_ssl 2.8.15 which cleanly (without any conflicts) patches into its source tree. 2.8.15 does not fix EAPI shared pool bug introduced in 2.8.13 and

RE: Flex failure during apache 1.3.28 make - RESOLVED

2003-07-21 Thread Boyle Owen
Thanks all! Touching the .c files in src/modules/ssl let flex do its work and the make continued without a hitch. I repeated the build with a cleanly untarred distribution and it compiled smoothly. Looking back, my original attempt failed because I set the wrong path for EAPI_MM (I upgraded

Re: Flex failure during apache 1.3.28 make - RESOLVED

2003-07-21 Thread Mads Toftum
On Mon, Jul 21, 2003 at 05:14:53PM +0200, Boyle Owen wrote: Thanks all! Touching the .c files in src/modules/ssl let flex do its work and the make continued without a hitch. Well, to be precise, that's not what happened. Make checks the date of the .c file that is output from flex - if the