On Wed, Dec 02, 1998, [EMAIL PROTECTED] wrote:
> Full_Name: Jason Haar
> Version: 2.1.0 and 2.1.1
> OS: Linux (Redhat 5.1)
> Submission from: dirt.trimble.com (206.40.88.88)
>
> Apache 1.3.3 with mod_ssl. I configure as so:
>
> mod_ssl:
> ./configure --with-apache=../apache_1.3.3 --with-ssleay=/usr/local/ssl/
>
> apache:
> CFLAGS="-O6" ./configure "--suexec-caller=www" "--with-perl=/usr/bin/perl"
> "--enable-module=info" "--enable-module=expires" "--enable-module=ssl"
> "--enable-module=usertrack" "--enable-module=auth_dbm"
>
> All compiles fine, but when it runs, httpd will segfault when downloading
> via SSL files over a certain size (somewhere between 10 and 20Kb)
> - typically images.
>
> The same thing happens if the binary is run on a different Linux server,
>
> [Wed Dec 2 13:40:22 1998] [notice] httpd: child pid 913 exit signal
> Segmentation fault (11)
This sounds like the old SSLeay 0.8 problem. What is the
version of your SSLeay under /usr/local/ssl? Run
$ /usr/local/ssl/ssleay version
to find this out. When it's 0.8 this is your problem. You should really use
SSLeay 0.9, because 0.8 has such and other known problems.
> Also, if I go back to apache-1.3.2 with mod_ssl-2.0.11 on the same
> servers - it all works fine - so this is definitely a new problem...
Hmmm... that's interesting. But then we have to find out _WHERE_ it segfaults.
So please compile with OPTIM="-g -ggb3" and try to analyse it. For instance
when you get a core file try to load it into gdb and present us the output of
"bt". When this isn't possible then start httpd with
$ gdb /path/to/your/httpd
$ run -X
after the segfault occurs use
gdb> bt
Ralf S. Engelschall
[EMAIL PROTECTED]
www.engelschall.com
______________________________________________________________________
Apache Interface to SSLeay (mod_ssl) www.engelschall.com/sw/mod_ssl/
Official Support Mailing List [EMAIL PROTECTED]
Automated List Manager [EMAIL PROTECTED]
