Custom Apache Build System

host4leicester said:
Here is the error we are getting (same server as mq-james)


Make Complete
Installing php...
Installing PHP CLI binary: /usr/local/bin/
Installing PHP CLI man page: /usr/local/man/man1/
Installing PHP SAPI module: apache
[activating module `php4' in /etc/httpd/conf/httpd.conf]
cp libs/libphp4.so /usr/lib/apache/libphp4.so
chmod 755 /usr/lib/apache/libphp4.so
cp /etc/httpd/conf/httpd.conf /etc/httpd/conf/httpd.conf.bak
cp /etc/httpd/conf/httpd.conf.new /etc/httpd/conf/httpd.conf
rm /etc/httpd/conf/httpd.conf.new
Installing shared extensions: /usr/local/lib/php/extensions/no-debug-non-zts-20020429/
Installing PEAR environment: /usr/local/lib/php/
make[1]: *** [install-pear-installer] Segmentation fault
make: *** [install-pear] Error 2

*** The install has failed, do you want to try to install it again? (y,n):

A good way to find the source of this problem is to do the command
find . -name core -exec file {} \;
.That should file the core dump generated from the make. It should also tell you which program is actually causing the seg fault.

Also, something else you may want to do, is:
cd /usr/local/directadmin/customapache
rm -f configure.*
./build update
./build clean
./build all
That should rebuild everything from scratch. If you still get seg faults doing this, then there is something else wrong with your system.
 
host4leicester said:
there is nothing in that file.
is this wrong?

it said all files were succesfully installed...

If its not in the file that is not the problem... I have seen the pear error previously where the path specified for IMAP is incorrect.

Chris
 
cd /usr/local/directadmin/customapache
rm -f configure.*
./build update
./build clean
./build all

Try that as Tom mentioned, if you still get the error I think I may know where the problem stands :) - if you have custom configure lines back them up before deleting them and restore *after* a successful recompile. Just noticed your SSL problem which makes me think think the problem could be with OpenSSL

Chris
 
well it compiled succesfully and it restarted succesfully but its still acting like its not working (domains not loading...)

also last time I had to comment the ssl section in the config files for all users. this time the ssl section isnt even there...
 
host4leicester said:
well it compiled succesfully and it restarted succesfully but its still acting like its not working (domains not loading...)

Is BIND running ok? are you getting any errors? can you access through the IP/~user?

Chris
 
no and BIND is not showing up in the stats in DA (which im sure it usually is)

How do I test it or recompile it?
 
Working domains not loading is a result of the /etc/httpd/conf/httpd.conf not loading the domain config files. At the bottom, it should be including a file for each domain. Those files should be located in /usr/local/directadmin/data/users/${user}/httpd.conf
where the ${user} is the owner of the domain.
 
yes they are that is not the problem, I'm pretty sure its BIND... as before in previous post how do I find out?
 
service named status
service named start (if its off)
service named reload (if its on)

Chris
 
I also get the following error when restarting apache after enabling ssl.

Syntax error on line 1520 of /etc/httpd/conf/httpd.conf:
Invalid command 'SSLEngine', perhaps mis-spelled or defined by a module not included in the server configuration
 
I have reloaded named and the domains are still down (and IP's)
I am convinced its SSL, because I can't enable it without getting the errors above when restarting apache...

Any ideas?
 
Hello,

Note to anyone trying to restart apapache and gets SSLEngine errors, you *must* use:
Code:
service httpd restart
because apachectl doesn't pass any of the -DHAVE_SSL type commands when starting apache.(there are probably about 30 of them)

For any bind related issues, normally I find that boot scripts are the cause. Namely the "stop" part... I've been replacing any rndc commands with "killproc named".. and for restarts, I replace rndc commands with"stop" then "start". Basically, all the rndc commands were getting hung up and this ended up creating many copies of scripts trying to do things, but they never finish.
Type: "ps -ax | grep named" to see what's running. You should get at *most* 2 results, 1 being "[named]", the other being the program you just ran (doesn't always show up).

John
 
I always use the service command. We managed to fix the issue. Now to fix the MySQL issue onwards and upwards...
 
John,

A client of mine mailed you yesterday (dutch time :D) and asked you about a seg fault error with child processes, you recommended a recompile. It worked fine for a few hours, but then the httpd childs all died, and only one parent was left, but this one didn't do anything anymore.
I forced a restart this morning, but this really is getting irritating. Httpd keeps failing on us.

The ./configure.php contains :
#!/bin/sh
./configure \
--with-apxs \
--with-curl \
--with-curl-dir=/usr/local/lib \
--with-gd \
--with-gd-dir=/usr/local/lib \
--with-gettext \
--with-jpeg-dir=/usr/local/lib \
--with-kerberos \
--with-mcrypt \
--with-mysql \
--with-pear \
--with-png-dir=/usr/local/lib \
--with-xml \
--with-zlib \
--with-zlib-dir=/usr/local/lib \
--enable-bcmath \
--enable-calendar \
--enable-ftp \
--enable-magic-quotes \
--with-freetype-dir=/usr/local/lib \
--enable-sockets \
--enable-track-vars
It's standard, only the freetype thingy has been added.

The Apache httpd is standard.

I recompiled all modules without errors...

Can you think of any solution ?
 
is this custom Apache a necessary step for me? i installed the most recent version of DA, all softwares were up to date, so i do not recompile my apache, am i right?
 
If you installed the DA package the last few days you should ok, if you had DA already and just upgraded it recently then you may need to recompile some software.

Chris
 
Back
Top