I am on Debian Lenny and I had nginx 0.7.65 from lenny-backports installed. I needed the uwsgi module so I added the deb-src for lenny-backports and did apt-get source nginx. This grabbed the 0.7.67-3.
I followed this guide to a T: http://brandonkonkle.com/blog/2010/sep/14/django-uwsgi-and-nginx/ and then did dpkg -i nginx*.deb as per the instructions.
It doesn't seem like it replaced 0.7.65 because I get this:
(Reading database ... 27253 files and directories currently installed.)
Preparing to replace nginx 0.7.67-3~bpo50+1 (using nginx_0.7.67-3~bpo50+1_i386.deb) ...
Unpacking replacement nginx ...
Preparing to replace nginx-dbg 0.7.67-3~bpo50+1 (using nginx-dbg_0.7.67-3~bpo50+1_i386.deb) ...
Unpacking replacement nginx-dbg ...
Setting up nginx (0.7.67-3~bpo50+1) ...
Trying a soft restart
Processing triggers for man-db ...
Setting up nginx-dbg (0.7.67-3~bpo50+1) ...
And, if I try to /etc/init.d/nginx stop and start again, it wont obey. I think I may have caused a conflict of some sorts?
The dpkg -i was supposed to install 0.7.67, and I have 2 deb files after I did the dpkg-buildpackage: nginx_0.7.67-3~bpo50+1_i386.deb and nginx-dbg_0.7.67-3~bpo50+1_i386.deb
Where did I go wrong? Could someone explain why nginx doesnt obey at all now? I can't stop it or start it. If I go into aptitude it tells me that 0.7.67 is the installed version.
test test configuration
$ sudo nginx -t
try run nginx without /etc/init.d/nginx
$sudo nginx
Read log:
$sudo tail -n 100 /var/log/syslog
I recall an issue in debian's packaging a while back where they changed the pid location. It is also possible you changed a shared library that is causing the existing nginx to not find the code in the newly relocated address. You could do:
and see what the error is on the line after start-stop-daemon
Worst case, find the pid of nginx and do a kill (or if it is stuck, kill -9)
At that point, you should be able to restart nginx normally.