How to change default Nginx setting on Homestead Laravel Virtualbox VM - laravel

I merely followed the default Laravel Homestead setup here using VirtualBox. Working great.
But I need an additional Nginx rewrite setting in my apps vhost file on the VM, something like;
location / {
if ($request_method !~ "(POST)"){
rewrite ....
}
try_files $uri $uri/ /index.php?$query_string;
}
I can add that manually to the vhost file on the VM, but it's removed every time I provision my box.
How do I make that setting automatically applied when provision my vagrant VM box.
Is there a after provision hook, so I can run a script, or is there another easy solution available?

Easiest way to achieve this is editing scripts/serve.sh file, which contains server block template in block variable.

Related

Custom nginx configuration for Laravel in AWS Elastic Beanstalk

Hi I am using AWS elastic beanstalk for my Laravel application
When I first uploaded my application it is not showing the login page, 404 error so I tried with app_url/index.php/login and it worked
This is how I figure out I need to add this to my Nginx configuration, I manually added this, ssh into ec2 instance, and worked
location / {
try_files $uri $uri/ /index.php?$query_string;
}
but every time I deploy my new code it replaces my manual code
I also read https://docs.aws.amazon.com/elasticbeanstalk/latest/dg/java-tomcat-proxy.html?icmpid=docs_elasticbeanstalk_console this doc and created this folder structure in my application root folder
.ebextensions
nginx
conf.d
elasticbeanstalk
my-server-conf.conf
Seems not working for me
Please let me know what should I put In my-server-conf.conf file and what should I do so it works for every deploy
You should be using .platform, not .ebextensions as explained in aws docs. For example:
.platform/nginx/conf.d/elasticbeanstalk/laravel.conf
location / {
try_files $uri $uri/ /index.php?$query_string;
}

AWS Elasticbeanstalk overriding Nginx config using .platform is not working

I am deploying my Laravel application to AWS ElasticBeanstalk. I have deployed it. Now, I am trying to override "/etc/nginx/conf.d/elasticbeanstalk/php.conf" file using .platform folder.
I created .platform/etc/nginx/conf.d/elasticbeanstalk/php.conf file right inside the project's root folder. Then I put in the configuration content.
Then I deploy my application executing "be deploy" command. But the Nginx config file is not overridden. What is wrong with my config and how can I get it working?
I tried using .ebextensions too creating a config file with the following content. The file is not just created.
files:
/etc/nginx/conf.d/elasticbeanstalk/laravel.conf:
mode: "000755"
owner: root
group: root
content: |
location / {
try_files $uri $uri/ /index.php?$query_string;
gzip_static on;
}
The nginx config file you are trying to set is used in Amazon Linux 1 (AL1).
For AL2, the nginx config files should be provided (aws docs) using:
.platform/nginx/conf.d/
or if you want to overwrite main nginx config file, use
.platform/nginx/nginx.conf
Thus, you can try the following file .platform/nginx/conf.d/laravel.conf with content of:
location / {
try_files $uri $uri/ /index.php?$query_string;
gzip_static on;
}
If it does not work, you can inspect nginx logs in /var/log or nginx config folder to check if the file is correctly placed in nginx config folder.
Solution I used
I used this and it worked.
.platform/nginx/conf.d/elasticbeanstalk/laravel.conf

Don't think I have php setup right in nginx virtual host configs

I'm trying to install magento php ecommerce on a virtual server that previously did not have php. When I put simple php pages into a directory they run (for example I have index.php run phpinfo). But when I try to run the magento setup scripts they bomb out without outputting an error. I have two other virtual servers through a different service (running apache) and when I drop the magento files into them the setup pages light up right away and start checking requirements. On my nginx server the only output is a small grey box, but no error output.
My best guess is that I have not configured nginx virtual hosts properly for php.
I am on php5.5 using fpm on Ubuntu 12.04. I don't run a default file in sites-enabled, just two vhost files. Here's the vhost in question (the other site is working fine, but it's python):
server {
listen 80;
listen [::]:80;
#
server_name magento.mydomain.com;
#
root /var/www/magento.mydomain.com/public_html;
index index.php index.html index.htm;
#
# location / {
# try_files $uri $uri/ /index.php =404;
# }
# location / {
# /index.php;
# }
location ~ \.php$ {
include /etc/nginx/fastcgi_params;
fastcgi_pass 127.0.0.1:8070;
fastcgi_index index.php;
fastcgi_param SCRIPT_FILENAME /var/www/magento.mydomain.com/public_html/$fastcgi_script_name;
}
}
You can see my commented out location / directives. I have tried those is various fashion. My hunch was that magento was calling php with directory names only (instead of somedir/index.php) so I was messing with those, but it seems like that would be handled by the index directive before the locations.
I'm pretty green with nginx. Does anyone see anything obvious?

Vagrant/Virtualbox Nginx Provisioning

I'm running Laravel/Homestead as my Vagrant box and have heard that Nginx has some weird caching issues. The way to solve the problem is to set "sendfille off" in the nginx config file.
How can I provision my Laravel/Homestead Vagrant box so that I don't have to ssh into the box and modify the setting?
In your host's homestead folder you can find homestead/scripts/serve.sh. There sendfile is already set to off as default for your app. I think this will set nginx sendfile option to off.
If you still want to edit the nginx.conf permanently, write a line of code into serve.sh to change the config file.
More info on serve.sh
The serve.sh script will run on --provision for any site configured in the Homestead.yaml.
It creates each site in /etc/nginx/sites-available/ and symlinks it to /etc/nginx/sites-enabled/. Nginx and php5-fpm get restarted after that.
In nginx.conf we can see at the end of the http{...}-block has:
include /etc/nginx/conf.d/*.conf;
include /etc/nginx/sites-enabled/*;
This includes all sites in /etc/nginx/sites-available/ where the server{...}-blocks are defined. For more information about nginx config look at the Nginx Admin Guide.

Wildcards in a Windows hosts file

I want to setup my local development machine so that any requests for *.local are redirected to localhost. The idea is that as I develop multiple sites, I can just add vhosts to Apache called site1.local, site2.local etc, and have them all resolve to localhost, while Apache serves a different site accordingly.
I am on Windows XP.
I tried adding
127.0.0.1 *.local
to my c:\windows\system32\drivers\etc\hosts file, also tried:
127.0.0.1 .local
Neither of which seem to work.
I know I can set them up on different port numbers, but that is a pain since it is hard to remember which port is which.
I don't want to have to setup a local DNS server or anything hard, any suggestions?
Acrylic DNS Proxy (free, open source) does the job. It creates a proxy DNS server (on your own computer) with its own hosts file. The hosts file accepts wildcards.
Download from the offical website
http://mayakron.altervista.org/support/browse.php?path=Acrylic&name=Home
Configuring Acrylic DNS Proxy
To configure Acrylic DNS Proxy, install it from the above link then go to:
Start
Programs
Acrylic DNS Proxy
Config
Edit Custom Hosts File (AcrylicHosts.txt)
Add the folowing lines on the end of the file:
127.0.0.1 *.localhost
127.0.0.1 *.local
127.0.0.1 *.lc
Restart the Acrylic DNS Proxy service:
Start
Programs
Acrilic DNS Proxy
Config
Restart Acrylic Service
You will also need to adjust your DNS setting in you network interface settings:
Start
Control Panel
Network and Internet
Network Connections
Local Area Connection Properties
TCP/IPv4
Set "Use the following DNS server address":
Preferred DNS Server: 127.0.0.1
If you then combine this answer with jeremyasnyder's answer (using VirtualDocumentRoot) you can then automatically setup domains/virtual hosts by simply creating a directory.
To answer your question, you cannot use wildcards in the hosts file under Windows.
However, if you want to only change the hosts file to make new sites work.... you can configure your Apache like this and you don't have to keep editing it's config:
http://postpostmodern.com/instructional/a-smarter-mamp/
Basically a quick summary based on my setup, add the following to your apache.conf file:
LoadModule vhost_alias_module modules/mod_vhost_alias.so
NameVirtualHost *:80
<Directory "/xampp/sites">
Options Indexes FollowSymLinks Includes ExecCGI
AllowOverride All
Order allow,deny
Allow from all
</Directory>
<VirtualHost *:80>
VirtualDocumentRoot c:/xampp/sites/%-1/%-2+/
</VirtualHost>
This allows me to add an entry like:
127.0.0.1 test.dev
and then make the directory, c:\xampp\sites\dev\test and place the necessary files in there and it just works.
The other option is to use <Directory> tags in apache.conf and reference the pages from http://localhost/project/.
I don't think that it is possible.
You anyway have to modify the apache virtualroot entries every time you add a new site and location, so it's not a big work to syncronise the new name to the Windows vhost file.
Update: please check the next answer and the comments on this answer. This answer is 6 years old and not correct anymore.
To add to the great suggestions already here, XIP.IO is a fantastic wildcard DNS server that's publicly available.
myproject.127.0.0.1.xip.io -- resolves to --> 127.0.0.1
other.project.127.0.0.1.xip.io -- resolves to --> 127.0.0.1
other.machine.10.0.0.1.xip.io -- resolves to --> 10.0.0.1
(The ability to specify non-loopback addresses is fantastic for testing sites on iOS devices where you cannot access a hosts file.)
If you combine this with some of the Apache configuration mentioned in other answers, you can potentially add VirtualHosts with zero setup.
Editing the hosts file is less of a pain when you run "ipconfig /flushdns" from the windows command prompt, instead of restarting your computer.
I found a posting about Using the Windows Hosts File that also says "No wildcards are allowed."
In the past, I have just added the additional entries to the hosts file, because (as previously said), it's not that much extra work when you already are editing the apache config file.
You could talk your network administrator into setting up a domain for you (say 'evilpuppetmaster.hell') and having the wildcard there so that everything (*.evilpuppetmaster.hell') resolves to your IP
We have this working using wildcard DNS in our local DNS server: add an A record something like *.local -> 127.0.0.1
I think that your network settings will need to have the chosen domain suffix in the domain suffix search list for machines on the network, so you might want to replace .local with your company's internal domain (e.g. .int) and then add a subdomain like .localhost.int to make it clear what it's for.
So *.localhost.int would resolve to 127.0.0.1 for everybody on the network, and config file settings for all developers would "just work" if endpoints hang off that subdomain e.g. site1.localhost.int, site2.localhost.int This is pretty much the scheme we have introduced.
dnsmasq also looks nice, but I have not tried it yet:
http://ihaveabackup.net/2012/06/28/using-wildcards-in-the-hosts-file/
I have written a simple dns proxy in Python. It will read wildcard entries in /etc/hosts. See here: http://code.google.com/p/marlon-tools/source/browse/tools/dnsproxy/dnsproxy.py
I have tested in Linux & Mac OS X, but not yet in Windows.
You may try AngryHosts, which provided a way to support wildcard and regular expression. Actually, it's a hosts file enhancement and management software.
More features can be seen # http://angryhosts.com/features/
I'm using DNSChef to do that.
https://thesprawl.org/projects/dnschef/
You have to download the app, in Linux or Mac you need python to run it. Windows have their own exe.
You must create a ini file with your dns entries, for example
[A]
*.google.com=192.0.2.1
*.local=127.0.0.1
*.devServer1.com=192.0.2.3
Then you must launch the dns application with admin privileges
sudo python dnschef.py --file myfile.ini -q
or in windows
runas dnschef.exe --file myfile.ini -q
Finally you need to setup as your only DNS your local host environment (network, interface, dns or similar or in linux /etc/resolv.conf).
That's it
I made this simple tool to take the place of hosts. Regular expressions are supported.
https://github.com/stackia/DNSAgent
A sample configuration:
[
{
"Pattern": "^.*$",
"NameServer": "8.8.8.8"
},
{
"Pattern": "^(.*\\.googlevideo\\.com)|((.*\\.)?(youtube|ytimg)\\.com)$",
"Address": "203.66.168.119"
},
{
"Pattern": "^.*\\.cn$",
"NameServer": "114.114.114.114"
},
{
"Pattern": "baidu.com$",
"Address": "127.0.0.1"
}
]
#petah and Acrylic DNS Proxy is the best answer, and at the end he references the ability to do multi-site using an Apache which #jeremyasnyder describes a little further down...
... however, in our case we're testing a multi-tenant hosting system and so most domains we want to test go to the same virtualhost, while a couple others are directed elsewhere.
So in our case, you simply use regex wildcards in the ServerAlias directive, like so...
ServerAlias *.foo.local
Here is the total configuration for those trying to accomplish the goal (wildcards in dev environment ie, XAMPP -- this example assumes all sites pointing to same codebase)
hosts file (add an entry)
file: %SystemRoot%\system32\drivers\etc\hosts
127.0.0.1 example.local
httpd.conf configuration (enable vhosts)
file: \XAMPP\etc\httpd.conf
# Virtual hosts
Include etc\extra\httpd-vhosts.conf
httpd-vhosts.conf configuration
file: XAMPP\etc\extra\httpd-vhosts.conf
<VirtualHost *:80>
ServerAdmin admin#example.local
DocumentRoot "\path_to_XAMPP\htdocs"
ServerName example.local
ServerAlias *.example.local
# SetEnv APP_ENVIRONMENT development
# ErrorLog "logs\example.local-error_log"
# CustomLog "logs\example.local-access_log" common
</VirtualHost>
restart apache
create pac file:
save as whatever.pac wherever you want to and then load the file in the browser's network>proxy>auto_configuration settings (reload if you alter this)
function FindProxyForURL(url, host) {
if (shExpMatch(host, "*example.local")) {
return "PROXY example.local";
}
return "DIRECT";
}
You can use echoipdns for this (https://github.com/zapty/echoipdns).
By running echoipdns local all requests for .local subdomains are redirected to 127.0.0.1, so any domain with xyz.local etc will resolve to 127.0.0.1. You can use any other suffix also just replace local with name you want.
Echoipdns is even more powerful, when you want to use your url from other machines in network you can still use it with zero configuration.
For e.g. If your machine ip address is 192.168.1.100 you could now use a domain name xyz.192-168-1-100.local which will always resolve to 192.168.1.100. This magic is done by the echoipdns by looking at the ip address in the second part of the domain name and returning the same ip address on DNS query. You will have to run the echoipdns on the machine from which you want to access the remote system.
echoipdns also can be setup as a standalone DNS proxy, so by just point to this DNS, you can now use all the above benefits without running a special command every time, and you can even use it from mobile devices.
So essentially this simplifies the wildcard domain based DNS development for local as well as team environment.
echoipdns works on Mac, Linux and Windows.
NOTE: I am author for echoipdns.
I could not find a prohibition in writing, but by convention, the Windows hosts file closely follows the UNIX hosts file, and you cannot put wildcard hostname references into that file.
If you read the man page, it says:
DESCRIPTION
The hosts file contains information regarding the known hosts on the net-
work. For each host a single line should be present with the following
information:
Internet address
Official host name
Aliases
Although it does say,
Host names may contain any printable character other than a field delim-
iter, newline, or comment character.
that is not true from a practical level.
Basically, the code that looks at the /etc/hosts file does not support a wildcard entry.
The workaround is to create all the entries in advance, maybe use a script to put a couple hundred entries at once.
While you can't add a wildcard like that, you could add the full list of sites that you need, at least for testing, that works well enough for me, in your hosts file, you just add:
127.0.0.1 site1.local
127.0.0.1 site2.local
127.0.0.1 site3.local
...
Configuration for nginx config auto subdomain with Acrylic DNS Proxy
auto.conf file for your nginx sites folder
server {
listen 80;
server_name ~^(?<branch>.*)\.example\.com;
root /var/www/html/$branch/public;
index index.html index.htm index.php;
charset utf-8;
location / {
try_files $uri $uri/ /index.php$is_args$args;
}
location = /favicon.ico { access_log off; log_not_found off; }
location = /robots.txt { access_log off; log_not_found off; }
error_log /var/log/nginx/$branch.error.log error;
sendfile off;
client_max_body_size 100m;
location ~ \.php$ {
try_files $uri /index.php =404;
fastcgi_pass php-fpm:9000;
fastcgi_index index.php;
fastcgi_buffers 16 16k;
fastcgi_buffer_size 32k;
fastcgi_param SCRIPT_FILENAME $document_root$fastcgi_script_name;
include fastcgi_params;
}
location ~ /\.ht {
deny all;
}
}
Add to Acrylic hosts file 127.0.0.1 example.com *.example.com and restart Acrylic service.
$branch - your subdomain name.
Set instead of root /var/www/html/$branch/public; your project path
This can be done using Pi-Hole, just edit the "/etc/hosts" and restart dns service.
nano /etc/hosts
pihole restartdns
Example:
127.0.1.1 raspberrypi
192.168.1.1 w1.dev.net
192.168.1.2 w2.dev.net
192.168.1.3 w3.dev.net
You can use a dynamic DNS client such as http://www.no-ip.com. Then, with an external DNS server CNAME *.mydomain.com to mydomain.no-ip.com.

Resources