Install wordpress on CentOS7 with nginx & caching

Lately I’ve become interested in the LEMP stack (as opposed to the LAMP stack.) As such I’ve decided to set up a wordpress site running CentOS 7, nginx, mariadb, php-fpm, zend-opcache, apc, and varnish. This writeup will borrow heavily from two of my other writeups, Install Wordpress on CentOS 7 with  SELinux and Speed up WordPress in CentOS7 using caching. This will be a mashup of those two with an nginx twist with guidance from digitalocean. Let’s begin.

Repositories

To install the required addons we will need to have the epel repository enabled:

yum -y install epel-release

nginx

Install necessary packages:

sudo yum -y install nginx
sudo systemctl enable nginx

Optional: symlink /usr/share/nginx/ to /var/www/ (for those of us who are used to apache)

sudo ln -s /usr/share/nginx/ /var/www

Open necessary firewall ports:

sudo firewall-cmd --add-service=http --permanent
sudo systemctl restart firewalld

start nginx:

sudo systemctl start nginx

Navigate to your new site to make sure it brings up the default page properly.

MariaDB

Install:

sudo yum -y install mariadb-server mariadb
sudo systemctl enable mariadb

Run initial mysql configuration to set database root password

sudo systemctl start mariadb
sudo mysql_secure_installation

Configure:

Create a wordpress database and user:

mysql -u root -p 
#enter your mysql root password here
create user wordpress;
create database wordpress;
GRANT ALL PRIVILEGES ON wordpress.* To 'wordpress'@'localhost' IDENTIFIED BY 'password';
quit;

php-fpm

Install:

sudo yum -y install php-fpm php-mysql php-pclzip
sudo systemctl enable php-fpm

Configure:

Uncomment cgi.fix_pathinfo and change value to 0:

sudo sed -i 's/\;\(cgi.fix_pathinfo=\)1/\10/g' /etc/php.ini

Modify the listen= parameter to listen to UNIX socket instead of TCP:

sudo sed -i 's/\(listen =\).*/\1 \/var\/run\/php-fpm\/php-fpm.sock/g' /etc/php-fpm.d/www.conf

Change listen.owner and listen.group to nobody:

sudo sed -i 's/\(listen.owner = \).*/\1nobody/g; s/\(listen.group = \).*/\1nobody/g' /etc/php-fpm.d/www.conf

Change running user & group from apache to nginx:

sudo sed -i 's/\(^user = \).*/\1nginx/g; s/\(^group = \).*/\1nginx/g' /etc/php-fpm.d/www.conf

Start php-fpm:

sudo systemctl start php-fpm

Caching

To speed up wordpress further we need to install a few bits of caching software. Accept defaults when prompted.

zend-opcache & apc:

Install necessary packages:

sudo yum -y install php-pecl-zendopcache php-pecl-apcu php-devel gcc
sudo pecl install apc

Add apc extension to php configuration:

sudo sh -c "echo '\

#Add apc extension 
extension=apc.so' >> /etc/php.ini"

Restart php-fpm:

sudo systemctl restart php-fpm

Varnish

Install:

sudo yum -y install varnish
sudo systemctl enable varnish

Configure nginx to listen on port 8080 instead of port 80:

sudo sed -i /etc/nginx/nginx.conf -e 's/listen.*80/&80 /'

Change varnish to listen on port 80 instead of port 6081:

sudo sed -i /etc/varnish/varnish.params -e 's/\(VARNISH_LISTEN_PORT=\).*/\180/g'

Optional: change varnish to cache files in memory with a limit of 256M (caching to memory is much faster than caching to disk)

sudo sed -i /etc/varnish/varnish.params -e 's/\(VARNISH_STORAGE=\).*/\1\"malloc,256M\"/g'

Add varnish configuration to work with caching wordpress sites:

Update 2/25/2018 added section to allow facebook to properly scrape varnish cached sites.

sudo sh -c 'echo "/* SET THE HOST AND PORT OF WORDPRESS
 * *********************************************************/
vcl 4.0;
import std;

backend default {
 .host = \"127.0.0.1\";
 .port = \"8080\";
 .first_byte_timeout = 60s;
 .connect_timeout = 300s;
}
 
# SET THE ALLOWED IP OF PURGE REQUESTS
# ##########################################################
acl purge {
 \"localhost\";
 \"127.0.0.1\";
}

#THE RECV FUNCTION
# ##########################################################
sub vcl_recv {

#Facebook workaround to allow proper shares
if (req.http.user-agent ~ \"facebookexternalhit\")
        {
        return(pipe);
        }

# set realIP by trimming CloudFlare IP which will be used for various checks
set req.http.X-Actual-IP = regsub(req.http.X-Forwarded-For, \"[, ].*$\", \"\"); 

 # FORWARD THE IP OF THE REQUEST
 if (req.restarts == 0) {
 if (req.http.x-forwarded-for) {
 set req.http.X-Forwarded-For =
 req.http.X-Forwarded-For + \", \" + client.ip;
 } else {
 set req.http.X-Forwarded-For = client.ip;
 }
 }

 # Purge request check sections for hash_always_miss, purge and ban
 # BLOCK IF NOT IP is not in purge acl
 # ##########################################################

 # Enable smart refreshing using hash_always_miss
if (req.http.Cache-Control ~ \"no-cache\") {
 if (client.ip ~ purge) {
 set req.hash_always_miss = true;
 }
}

if (req.method == \"PURGE\") {
 if (!client.ip ~ purge) {
 return(synth(405,\"Not allowed.\"));
 }

 ban (\"req.url ~ \"+req.url);
 return(synth(200,\"Purged.\"));

}

if (req.method == \"BAN\") {
 # Same ACL check as above:
 if (!client.ip ~ purge) {
 return(synth(403, \"Not allowed.\"));
 }
 ban(\"req.http.host == \" + req.http.host +
 \" && req.url == \" + req.url);

 # Throw a synthetic page so the
 # request wont go to the backend.
 return(synth(200, \"Ban added\"));
}

# Unset cloudflare cookies
# Remove has_js and CloudFlare/Google Analytics __* cookies.
 set req.http.Cookie = regsuball(req.http.Cookie, \"(^|;\s*)(_[_a-z]+|has_js)=[^;]*\", \"\");
 # Remove a \";\" prefix, if present.
 set req.http.Cookie = regsub(req.http.Cookie, \"^;\s*\", \"\");

 # For Testing: If you want to test with Varnish passing (not caching) uncomment
 # return( pass );

# DO NOT CACHE RSS FEED
 if (req.url ~ \"/feed(/)?\") {
 return ( pass ); 
}

#Pass wp-cron

if (req.url ~ \"wp-cron\.php.*\") {
 return ( pass );
}

## Do not cache search results, comment these 3 lines if you do want to cache them

if (req.url ~ \"/\?s\=\") {
 return ( pass ); 
}

# CLEAN UP THE ENCODING HEADER.
 # SET TO GZIP, DEFLATE, OR REMOVE ENTIRELY. WITH VARY ACCEPT-ENCODING
 # VARNISH WILL CREATE SEPARATE CACHES FOR EACH
 # DO NOT ACCEPT-ENCODING IMAGES, ZIPPED FILES, AUDIO, ETC.
 # ##########################################################
 if (req.http.Accept-Encoding) {
 if (req.url ~ \"\.(jpg|png|gif|gz|tgz|bz2|tbz|mp3|ogg)$\") {
 # No point in compressing these
 unset req.http.Accept-Encoding;
 } elsif (req.http.Accept-Encoding ~ \"gzip\") {
 set req.http.Accept-Encoding = \"gzip\";
 } elsif (req.http.Accept-Encoding ~ \"deflate\") {
 set req.http.Accept-Encoding = \"deflate\";
 } else {
 # unknown algorithm
 unset req.http.Accept-Encoding;
 }
 }

 # PIPE ALL NON-STANDARD REQUESTS
 # ##########################################################
 if (req.method != \"GET\" &&
 req.method != \"HEAD\" &&
 req.method != \"PUT\" && 
 req.method != \"POST\" &&
 req.method != \"TRACE\" &&
 req.method != \"OPTIONS\" &&
 req.method != \"DELETE\") {
 return (pipe);
 }
 
 # ONLY CACHE GET AND HEAD REQUESTS
 # ##########################################################
 if (req.method != \"GET\" && req.method != \"HEAD\") {
 return (pass);
 }
 
 # OPTIONAL: DO NOT CACHE LOGGED IN USERS (THIS OCCURS IN FETCH TOO, EITHER
 # COMMENT OR UNCOMMENT BOTH
 # ##########################################################
 if ( req.http.cookie ~ \"wordpress_logged_in|resetpass\" ) {
 return( pass );
 }

 #fix CloudFlare Mixed Content with Flexible SSL
 if (req.http.X-Forwarded-Proto) {
 return(hash);
 }

 # IF THE REQUEST IS NOT FOR A PREVIEW, WP-ADMIN OR WP-LOGIN
 # THEN UNSET THE COOKIES
 # ##########################################################
 if (!(req.url ~ \"wp-(login|admin)\") 
 && !(req.url ~ \"&preview=true\" ) 
 ){
 unset req.http.cookie;
 }

 # IF BASIC AUTH IS ON THEN DO NOT CACHE
 # ##########################################################
 if (req.http.Authorization || req.http.Cookie) {
 return (pass);
 }
 
 # IF YOU GET HERE THEN THIS REQUEST SHOULD BE CACHED
 # ##########################################################
 return (hash);
 # This is for phpmyadmin
if (req.http.Host == \"pmadomain.com\") {
return (pass);
}
}

sub vcl_hash {

if (req.http.X-Forwarded-Proto) {
 hash_data(req.http.X-Forwarded-Proto);
 }
}


# HIT FUNCTION
# ##########################################################
sub vcl_hit {
 return (deliver);
}

# MISS FUNCTION
# ##########################################################
sub vcl_miss {
 return (fetch);
}

# FETCH FUNCTION
# ##########################################################
sub vcl_backend_response {
 # I SET THE VARY TO ACCEPT-ENCODING, THIS OVERRIDES W3TC 
 # TENDANCY TO SET VARY USER-AGENT. YOU MAY OR MAY NOT WANT
 # TO DO THIS
 # ##########################################################
 set beresp.http.Vary = \"Accept-Encoding\";

 # IF NOT WP-ADMIN THEN UNSET COOKIES AND SET THE AMOUNT OF 
 # TIME THIS PAGE WILL STAY CACHED (TTL), add other locations or subdomains you do not want to cache here in case they set cookies
 # ##########################################################
 if (!(bereq.url ~ \"wp-(login|admin)\") && !bereq.http.cookie ~ \"wordpress_logged_in|resetpass\" ) {
 unset beresp.http.set-cookie;
 set beresp.ttl = 1w;
 set beresp.grace =3d;
 }

 if (beresp.ttl <= 0s ||
 beresp.http.Set-Cookie ||
 beresp.http.Vary == \"*\") {
 set beresp.ttl = 120 s;
 # set beresp.ttl = 120s;
 set beresp.uncacheable = true;
 return (deliver);
 }

 return (deliver);
}

# DELIVER FUNCTION
# ##########################################################
sub vcl_deliver {
 # IF THIS PAGE IS ALREADY CACHED THEN RETURN A HIT TEXT 
 # IN THE HEADER (GREAT FOR DEBUGGING)
 # ##########################################################
 if (obj.hits > 0) {
 set resp.http.X-Cache = \"HIT\";
 # IF THIS IS A MISS RETURN THAT IN THE HEADER
 # ##########################################################
 } else {
 set resp.http.X-Cache = \"MISS\";
 }
}" > /etc/varnish/default.vcl'

Restart varnish & nginx:

sudo systemctl restart nginx varnish

Logging

By default varnish does not log its traffic. This means that your apache log will only log things varnish does not cache. We have to configure varnish to log traffic so you don’t lose insight into who is visiting your site.

Update 2/14/2017:  I’ve discovered a better way to do this. The old way is still included below, but you really should use this other way.

New way:

CentOS ships with some systemd scripts for you. You can use them out of the box by simply issuing

systemctl start varnishncsa
systemctl enable varnishncsa

If you are behind a reverse proxy then you will want to tweak the varnishncsa output a bit to reflect x-forwarded-for header values (thanks to this github discussion for the guidance.) Accomplish this by appending a modified log output format string to /lib/systemd/system/varnishncsa.service:

sudo sed -i /lib/systemd/system/varnishncsa.service -e "s/ExecStart.*/& -F \'%%{X-Forwarded-For}i %%l %%u %%t \"%%r\" %%s %%b \"%%{Referer}i\" \"%%{User-agent}i\"\' /g"

Lastly, reload systemd configuration, enable, and start the varnishncsa service:

sudo systemctl daemon-reload
sudo systemctl enable varnishncsa
sudo systemctl start varnishncsa

 


Old way:

First, enable rc.local

sudo chmod +x /etc/rc.local
sudo systemctl enable rc-local

Next, add this entry to the rc.local file:

sudo sh -c 'echo "varnishncsa -a -w /var/log/varnish/access.log -D -P /var/run/varnishncsa.pid" >> /etc/rc.local'

If your varnish server is behind a reverse proxy (like a web application firewall) then modify the above code slightly (thanks to this site for the information on how to do so)

sudo sh -c "echo varnishncsa -a -F \'%{X-Forwarded-For}i %l %u %t \"%r\" %s %b \"%{Referer}i\" \"%{User-agent}i\"\' -w /var/log/varnish/access.log -D -P /var/run/varnishncsa.pid >> /etc/rc.local"

Once configuration is in place, start the rc-local service

sudo systemctl start rc-local

WordPress

Download, extract, and set permissions for your wordpress installation (this assumes your wordpress site is the only site on the server)

wget https://wordpress.org/latest.zip
sudo unzip latest.zip -d /usr/share/nginx/html/
sudo chown nginx:nginx -R /usr/share/nginx/html/

Configure nginx

Follow best practice for nginx by creating a new configuration file for your wordpress site. In this example I’ve created a file wordpress.conf inside the /etc/nginx/conf.d directory.

Create the file:

sudo vim /etc/nginx/conf.d/wordpress.conf

Insert the following, making sure to modify the server_name directive:

server {
    listen 8080;
    listen [::]:8080;
    server_name wordpress;
    root /usr/share/nginx/html/wordpress;
    port_in_redirect off;


    location / {
        index index.php;
        try_files $uri $uri/ /index.php?q=$uri&$args;
    }

    location ~ \.php$ {
        fastcgi_pass unix:/var/run/php-fpm/php-fpm.sock;
        fastcgi_index index.php;
        fastcgi_param SCRIPT_FILENAME $document_root$fastcgi_script_name;
        include fastcgi_params;
    }

    error_page 404 /404.html;     
    location = /40x.html {
     }

    error_page 500 502 503 504 /50x.html;
    location = /50x.html {
    }
}

Restart nginx

sudo systemctl restart nginx

Configure upload directory

If you want users to upload content, then you will want to assign the http_sys_rw_content_t selinux security context for the wp-uploads directory (create it if it doesn’t exist)

sudo mkdir /usr/share/nginx/html/wordpress/wp-content/uploads
sudo chown nginx:nginx /usr/share/nginx/html/wordpress/wp-content/uploads
sudo semanage fcontext -a -t httpd_sys_rw_content_t "/usr/share/nginx/html/wordpress/wp-content/uploads(/.*)?"
sudo restorecon -Rv /usr/share/nginx/html/wordpress/wp-content/uploads

Run the wizard

In order for the wizard to run properly we need to temporarily give the wordpress directory httpd_sys_rw_content_t selinux context

sudo chcon -t httpd_sys_rw_content_t /usr/share/nginx/html/wordpress/

Now navigate to your new website in a browser and follow the wizard, which will create a wp-config.php file inside the wordpress directory.

W3 Total Cache

Once your wordpress site is set up you will want to configure it to communicate with varnish. This will make sure the cache is always up to date when changes are made.

Install the W3 Total Cache wordpress plugin and configure it as follows:

opcode-database-object

Opcode cache: Opcode:Zend Opcache

Database cache: Check enable, select Opcode: Alternative PHP Cache (APC / APCu)

Object cache: Check enable, select Opcode: Alternative PHP Cache (APC / APCu)

Fragment cache: Opcode: Alternative PHP cache (APC / APCu)

proxy

Reverse Proxy: Check “Enable reverse proxy caching via varnish”
Specify 127.0.0.1 in the varnish servers box. Click save all settings.

Wrapping up

Once your site is properly set up, restore the original security context for the wordpress directory:

sudo restorecon -v /usr/share/nginx/html/wordpress/

Lastly restart nginx and varnish:

sudo systemctl restart nginx varnish

Success! Everything is working within the proper SELinux contexts and caching configuration.

Troubleshooting

403 forbidden

I received this error after setting everything up. After some digging I came across this site which explained what could be happening.

For me this meant that nginx couldn’t find an index file and was trying to default to a directory listing, which is not allowed by default. This is fixed by inserting a proper directive to find index files, in my case, index.php. Make sure you have “index index.php” in your nginx.conf inside the location / block:

    location / {
        index index.php;
    }

 Accessing wp-admin redirects you to port 8080, times out

If you find going to /wp-admin redirects you to a wrong port and times out, it’s because nginx is forwarding the portnumber. We want to turn that off (thanks to this site for the help.) Add this to your nginx.conf:

  port_in_redirect off;

Append users to powerbroker open RequireMembershipOf

The title isn’t very descriptive. I recently came across a need to script adding users & groups to the “RequireMembershipOf” directive of PowerBroker Open. PowerBroker is a handy tool that really facilitates joining a Linux machine to a Windows domain. It has a lot of configurable options but the one I was interested in was RequireMembershipOf – which as you might expect requires that the person signing into the Linux machine be a member of that list.

The problem with RequireMembershipOf is, as far as I can tell, it has no append function. It has an add function which frustratingly erases everything that was there before and includes only what you added onto the list. I needed a way to append a member to the already existing RequireMembershipOf list. My solution involves the usage of bash, sed, and a lot of regex. It boils down to two lines of code:

#take output of show require membership of, remove words multistring & local policy, replace spaces with carat (pbis space representation) and put results into variable (which automatically puts results onto a single line)

add=$(/opt/pbis/bin/config --show RequireMembershipOf | sed 's/\(multistring\)\|\(local policy\)//g' | sed 's/ /^/g')

#run RequireMembershipOf command with previous output and any added users

sudo /opt/pbis/bin/config RequireMembershipOf "$add" "<USER_OR_GROUP_TO_ADD>"

That did the trick.

Change ZFS based NFS SR address in Xenserver

I recently acquired a shiny new set of SSDs to host my VMs. The problem is I needed to create a new ZFS array to accommodate them. I needed to figure out a way to migrate my VMs to the new array and then instruct Xenserver to use the new array instead of the old one.

Fortunately with a bit of research I learned this is fairly painless. Thanks to this discussion on citrix forums that got me pointed in the right direction. To change the server / IP address of an existing NFS storage repository in Xenserver you must do the following:

  • Shut down affected VMs
  • Shutdown any VMs using NFS SRs
  • Copy the NFS SRs (the directories containing the .vhd files) to the new NFS server
  • xe pbd-unplug uuid=<uuid of pbd pointing to the NFS SR>
  • xe pbd-destroy uuid=<uuid of pbd pointing to the NFS SR>
  • xe pbd-create host-uuid=<uuid of Xen Host> sr-uuid=<uuid of the NFS SR> device-config-server=<New NFS server name> device-config-serverpath=<NFS Share Name>
  • xe pbd-plug uuid=<uuid of the pbd created above>
  • Reboot the VMs using NFS SRs

In my case since my VMs were on an existing ZFS volume with snapshots I wanted to preserve, I used ZFS send and receive to transfer data from my old array to my SSD array. Bonus: I was able to do this while the VMs were still running to ensure minimal downtime. My ZFS copy procedure was as follows:

  • Create recursive snapshot of my VM dataset
    zfs snapshot -r storage/VMs@migrate
  • Start the initial data transfer (this took quite some time to finish)
    zfs send -R storage/VMs@migrate | zfs recv ssd/VMs
  • Do another incremental snapshot and transfer after initial huge transfer is complete (this took much less time to do)
    zfs snapshot -r storage/VMs@migrate2
    zfs send -R -i storage/VMs@migrate storage/VMs@migrate2 | zfs recv ssd/VMs
  • Shutdown all affected VMs and do one more ZFS snapshot & transfer to ensure consistent data:
    zfs snapshot -r storage/VMs@migrate3
    zfs send -R -i storage/VMs@migrate2 storage/VMs@migrate3 | zfs recv ssd/VMs

In the above examples my source dataset was storage/VMs and the destination dataset was ssd/VMs.

Once the data was all transferred to the new location it was time to tell Xenserver about it. I had enough VMs that it was worth my time to write a little script to do it. It’s quick and dirty but it did the job. Behold:

#!/bin/bash
#Author: Nicholas Jeppson
#A simple script to change a xenserver NFS storage repository address to a new location
#Modify NFS_SERVER, NFS_PATH and/or NFS_VERSION to match your environment. 
#Run this script on each xenserver host in your pool. Empty output means the transfer was successful.
#This script takes one argument - the name of the SR to be transferred.

SR_NAME="$1"

NFS_SERVER=10.0.0.1
NFS_PATH=/mnt/ssd/VMs/$SR_NAME
NFS_VERSION=4

#Use sed and awk to grab necessary UUIDs
HOST_UUID=$(xe host-list|egrep -B3 `hostname`$ | grep uuid | awk '{print $5}')
PBD_UUID=$(xe pbd-list|grep -A4 -B4 $SR_NAME | grep -B2 $HOST_UUID |grep -w '^uuid ( RO)' | awk '{print $5}')
SR_UUID=$(xe pbd-list|grep -A4 -B4 $SR_NAME | grep -A2 $HOST_UUID | grep 'sr-uuid' | awk '{print $4}')

#Unplug & destroy old NFS location, create new NFS location
xe pbd-unplug uuid=$PBD_UUID
xe pbd-destroy uuid=$PBD_UUID
NEW_PBD_UUID=$(xe pbd-create host-uuid=$HOST_UUID sr-uuid=$SR_UUID device-config-server=$NFS_SERVER device-config-serverpath=$NFS_PATH device-config-nfsversion=$NFS_VERSION)
xe pbd-plug uuid=$NEW_PBD_UUID

Download the script here (right click / save as)

You can run this script in a simple for loop with something like this:

for SR in <list of SR names separated by a space>; do bash <name of script saved from above> $SR; done

If you named the above script nfs-migrate.sh, and you had three SRs to change (blog1, blog2, blog3) then it would be:

for SR in blog1 blog2 blog3; do bash nfs-migrate.sh $SR; done

After I migrated the data and ran that script, my VMs booted up using the new SSD array. Success.