Categories
Nginx

Nginx 1.2.7

Today, Nginx 1.2.7 has been released with a few new features but also bugfixes merged from 1.3.x. See the full Changelog for more details.

As a consequence, Dotdeb’s packages of Nginx 1.2.7 are now available for Debian 6.0 “Squeeze” (amd64/i386) with the following changes :

  • Logs are now rotated on a weekly basis.
  • Passenger has been upgraded to the 3.0.19 version.
  • The chunkin-nginx-module, nginx-lua and nginx-upload-module modules got minor updates.

As usual, if you want to know which module has been included in each Nginx flavor, you just have to look at this document.

39 replies on “Nginx 1.2.7”

Did you add realip module to naxsi flavor?! Thank you very much! Last time I tried to use naxsi + vhosts by reverse proxy, but without this it wasn’t “nice” 🙂

websites went down tried “service nginx restart”
Restarting nginx: nginxnginx: [emerg] bind() to [::]:80 failed (98: Address already in use)
nginx: [emerg] bind() to [::]:80 failed (98: Address already in use)

First time it’s happened
using nginx-extras

I run a update on my Debian Box to the new version and find this strange behavior. I try to explain this bug.

First the Installation error:
http://pastebin.com/8f4qHAE0

After a second run of apt-installation there was still nginx 1.2.6 running.
I killed all PIDs of them manually.
As next step I tried to start nginx /etc/init.d/nginx start

Result: no output and nginx was not running.

As next step I checked my configuration manually:

root@nginx:~# /usr/sbin/nginx -t
nginx: the configuration file /etc/nginx/nginx.conf syntax is ok
nginx: [emerg] open() “/var/log/nginx/domain/subdomains/kernel/access.log” failed (2: No such file or directory)
nginx: configuration file /etc/nginx/nginx.conf test failed

The rc-script was modified and don’t show any error messages if they occur.

I compare init scrips of both version (1.2.7 and 1.2.6) Version 1.2.6 shows my mistake and version 1.2.7 is silent.

After installing, I cannot reload config files and cannot update any server blocks. Added a new server block, issue the reload and not working. Removing an existent server block is the same.

if pid variable in /etc/nginx/nginx.conf is somewhat different than : pid /var/run/nginx.pid .., (eg. more spaces between “pid” and “/var/run/nginx.pid”) then /etc/init.d/nginx fails with message : nginx: invalid option: “/var/run/nginx.pid”. problem is in statement : PID=$(awk -F'[ ;]’ ‘/[^#]pid/ {print $2}’ /etc/nginx/nginx.conf).
result of awk is different in that case and init.d nginx script fails… dont know how you evaluate this…, just mentioning…
thx,

Hello Guillaume,

I was wondering if you could tell me all of the configure options if i want to compile nginx from source to be like your nginx-full package?

I LOVE your work…. your packages are so helpful. But i recently switched to wheezy… and none of your packages work, even if i have a newer version of their dependencies.

I cannot move back to squeeze though… so i would like to try re-creating your nginx-full package.

Any help is greatly appreciated.

Thank you in advance.

This is excellent…. thank you so much
Guillaume.

Do you have the same thing for php?

I would like to use the latest php5-fpm, php5-apc, php5-cli, and php5-geoip.

But wheezy does not have any packages available that use the latest versions.

Thank you in advance!

Guillaume,
Please forgive my ignorance… but i am a novice with php, and i do not know how i would use those git repos to install the php components i need… in the same way that your packages provide.

I could not get them to work under any circumstances.

The packages said that they need very specific dependencies that are not available in wheezy…. allow me to paste what i am getting when trying to install dotdeb packages… if you do not mind. And btw… i sincerely appreciate your help in any form.

root@black~ $ aptitude install php5-fpm
The following NEW packages will be installed:
php5-fpm{b}
0 packages upgraded, 1 newly installed, 0 to remove and 0 not upgraded.
Need to get 2,724 kB of archives. After unpacking 8,548 kB will be used.
The following packages have unmet dependencies:
php5-fpm : Depends: libdb4.8 which is a virtual package.
Depends: libssl0.9.8 (>= 0.9.8m-1) which is a virtual package.
Depends: php5-common (= 5.4.11-1~dotdeb.0) but it is not going to be installed.
The following actions will resolve these dependencies:

Keep the following packages at their current version:
1) php5-fpm [Not Installed]

Accept this solution? [Y/n/q/?]

Not sure if I’m doing something completely wrong, but I setup dotdeb, installed nginx 1.2.7 from that, and it would not start. I wound up removing the IPv6 line from the default server block, which then allowed me to load the default page via my browser. However, the second I added php5-fpm or a new config file, nothing works (even after I removed both). nginx isn’t listening on port 80, it doesn’t even look like any processes are running, yet running “service nginx status” say it’s running.. has anyone else been having problems like this? I’ve tried installing it multiple times on multiple servers with the same result.

@Andrew — don’t know if this will help, but after much trial and error, I ended up purging all things nginx, installing nginx-full, then installing nginx-extras, then deleted the default config. Then it started. I’m sure much of that was unnecessary, but I was at the point of installing Apache… I’ve been down all day.

Has anybody encountered this Nginx error upon trying to start or restart Nginx?

Starting nginx: nginxnginx: invalid option: “5”

I’ve tried rebooting 3x and did ‘nginx remove’ and then ‘sudo apt-get install nginx-full’ and still get this same error. I did install beanstalkd and memcached recently, but they are up and running, so I don’t think it’s related to that. I think it’s just an nginx issue after this last upgrade.

I got it working, though by guess and check so I hope I didn’t screw the pooch on it.

I opened /etc/init.d/nginx found start() and stop() which both had “5” in it. After trying to change to the value to 4, I still had the error, but this time with “invalid option: 4”. So I deleted the number. And it works! Here’s the code from /etc/init.d/nginx in case it helps future googlers. http://pastebin.com/BLHzKtfR

This upgrade broke ssl for me. If I run “openssl s_client -connect 127.0.0.1:443”, I always get the error:
CONNECTED(00000003)
write:errno=104.
I reverted to version 1.2.6 of nginx-passenger from the archives and everything works.
Do I have to change my nginx configuration to make it work with 1.2.7??

Nginx not write to the log error. Just do not print it out on the console. Then install nginx from the official repository. Got the console error.

nginx: [emerg] could not build the server_names_hash, you should increase either server_names_hash_max_size: 512 or server_names_hash_bucket_size: 64

Other errors not display too. FIX please.

Comments are closed.