Phusion Passenger Tips and Troubleshooting Ideas

Some tips and troubleshooting ideas for Phusion Passenger

Phusion Passenger manages resources for rails applications – spawning new instances as needed, etc..

  • [bash]passenger-status[/bash] – provide the status of passenger rails processes

Configuring Phusion Passenger

Add lines to /etc/apache2/apache2.conf to change the default settings

  • [bash]PassengerMaxPoolSize 10[/bash] – maximum number of total rails application instances, the default is 6
  • PassengerMaxInstancesPerApp 5 – sets the maximum pool size for any 1 rails application to 10 instances (default is no limit).
  • [bash]PassengerUseGlobalQueue ON[/bash] – sets globaly queing on, it is off by default. You want globaly queuing on if your requests have large differences in response times (slow and fast responses).

Related: Passenger documentation

Troubleshooting

If you try [bash]sudo passenger-status[/bash] and get something like
*** ERROR: Cannot query status for Passenger instance 2280:
Connection refused – /tmp/passenger.2280/info/status.socket
Restarting (not reloading) apache [bash]sudo /etc/init.d/apache2 restart[/bash] may fix the problem.

System Monitoring Tools for VPS

Tools for monitoring performance and troubleshooting Ubuntu VPS web servers

  • Munin – graphs of system resources over time. Very nice. Can be a bit difficult to setup.
  • top – system stats
  • iotop – like top, but for io stats. Install [bash]sudo apt-get install iotop[/bash] Useful setup [bash]iotop -b -o -d 30 -t[/bash] -b (batch – so you can keep a running tally of results) -o (only those processes with io) -d (delay and seconds – how often to print out stats) -t (include time in printout)
  • vmstat – stats on memory, io, swap, cpu and system. Example: [bash]vmstat 10[/bash] (prints out stats every ten seconds.
  • iostat

Error logs

  • [bash]sudo nano /var/log/apache2/error.log[/bash]

Apache web server access log statistics

  • Webalizer – [bash]sudo apt-get install webalizer[/bash] GeoIP is required for webalizer [bash]sudo apt-get install geoip-bin[/bash] detailed instructions

Linux/Ubuntu File and Directory Permissions

Linux (and therefore Ubuntu) has file permissions on each file and directory for the owner, group and everyone else. Those permissions determine if the file can be viewed, executed or edited.

Only the owner of a file or directory (or a privileged user, root for example) may change its mode.

Ownership of a file

To change the ownership of the file or directory: chown new_owner_username directory

[bash]chown john public_html[/bash]

to change the ownership of directory (and all the files and folders in the directory) and also the group: chown -R new_owner_username:new_groupname directory

[bash]chown -R john:developers public_html[/bash]

to change the ownership of all the files in the current directory and also the group: chown -R new_owner_username:new_groupname *

[bash]chown john:developers *[/bash]

File permissions

The easiest way to set Linux file permissions is using a 3 digit sequence. The first digit designates owner permission; the second, the group permission; and the third, everyone else’s permission.

Read = 4
Write = 2
Execute = 1

The digit is the sum of those. So if you want to grant only read permission you use 4; read and execute 5; read, write and execute = 7.

[bash]chmod 775 index.html[/bash]

That will set the permissions on index.html so the owner, and a user in the group specified can read, write and execute the file and everyone else can read and execute.

[bash]chmod -R 755 public_html[/bash]

That will set the permissions on files and directories (recursively through all subdirectories) so the owner can read, write and execute; members of the group and everyone else can read and execute (but not write).

[bash]ls – l[/bash]

That will give you a list of files and directories, in a directory, with the owner and group settings and the permissions for all 3 (those 2 and everyone else), which will look something like:

[bash]-rw-r–r– 1 root developers 397 2008-05-25 20:33 index.html
-rw-r–r– 1 mary developers 9177 2010-05-02 22:18 unix_file_permissions.html
…[/bash]

The lines start with the permissions for the owner, group and then everyone else. There are 9 total characters, 3 for each. Taking the top line above:

rw-r--r--
rw-  (means the owner has read and write permission but not execute)
r--  (means the group has only read permission)
r--  (means everyone else has only read permission)

The next column tells you the number of hard links to the file or directory. Then column tells you the owner, then the group. Then the byte size of the file, the date it was last change and then the file name.

root
means the username of this file is named root

developers
group (means those users in the group named developers have the group permissions indicated)

Related: Ubuntu command line interface syntax examples

Using the Host File in Ubuntu

You can use the host file to have your computer route to whatever addresses you desire (instead of using your nameserver). For example, by putting

[bash]sudo nano /etc/hosts[/bash]

Then add a line to the file with the ip address and the name you will use.
[bash]204.11.50.136 wastetime[/bash]

One useful way to use this is to test out a website on a new host prior to changing the nameserver to point to the new host. In this case, if you want to make sure your host file is being read you can ping wastetime and if it is working it will show the results for a ping to 204.11.50.136

Code to Display text only on WordPress Blog Home Page

Cut and past php code to determine if it is the wordpress blog home page (and only the home page itself).

[php]<?php if (is_home() && !is_paged()): ?>[/php]

Example, to print john if it is a “homepage” templated page and jill otherwise:

[php]<?php if (is_home(); ?>

<?php
if($name == ‘john’){
echo ‘Hi John’;
}else if($name == ‘jill’){
echo ‘Hi Jill’;
}else{
echo ‘Hi’;
}
?>[/php]

Using scp (secure copy) to Copy Files Between Computers

Copy a file from your local computer to a remote host using secure copy, scp (which uses ssh for data transfer and provides the same security as using ssh).

scp [filename] [username]@remotehost:[location]

[bash]scp file_to_copy.txt [email protected]:/some/remote/directory[/bash]

copy a directory to your home computer from the remote computer.

[bash]scp -r directory_to_copy [email protected]:[/bash]

copy a directory from a remote server to the current directory on your computer.

scp -r folder [username]@remotehost:[location] .

[bash]scp -r [email protected]:/some/remote/directory .[/bash]

If you don’t have automated keys setup you will be asked for the password for that user.

An example for copying a MySQL database. Including the : without a location puts the file in the home folder.

[bash]mysqldump database_name -uroot > database_dump.sql
scp database_dump.sql [email protected]:[/bash]

Then ssh into the remote server and open the mysql prompt
[bash]mysql -uroot -p
mysql> create database database_name;
mysql> exit[/bash]

Then run the mysqldump file

[bash]mysql database_name -uuser -p < database_dump.sql[/bash]

Remember to create the database user on the new machine (this has to match what is in the wp-config.php file).

Ubuntu/Linux cli syntax

Command line interface syntax for various actions

Make new directory, make [directory_name]

[bash]sudo mkdir new_directory[/bash]

Remove (file or directory)

rm [name]
[bash]rm -i file_to_remove.txt[/bash]
Using -i prompts you to confirm the deletion.
Remove directory and all of its contents without having to confirm. Obviously be careful.
[/bash]sudo rm -r directory_to_remove[/bash]

Move a file

mv [name] [new_location]
[bash]mv file.txt new_sub_directory/file.txt[/bash]

Rename a file (similar to moving)
mv [name] [new_name]
[bash]mv file.txt new_file_name.txt[/bash]

Copy a fold

cp -r [folder]/* [new_location]
[bash]cp -r folder/* /some_place/else/[/bash]

Keep ssh sessions live

If you want to stop your SSH sessions from being shut down you can add the following line to /etc/ssh/ssh_config on your local machine.
[bash]sudo gedit /etc/ssh/ssh_config[/bash]
Then add:
[bash]ServerAliveInterval 30[/bash]
This does remove the security feature of closing your session in case you leave your computer, but you may decide to take that risk. This sends a SSH package every 30 seconds so your server doesn’t close the connection due to inactivity.

Edit DNS Name Server

Set the DNS name server to use for your machine.

[bash]sudo nano /etc/resolv.conf[/bash]

Then you can use for example (8.8.8.8 is Google public DNS, 208.67.222.222 is OpenDNS):
[bash]nameserver 208.67.222.222
nameserver 8.8.8.8[/bash]

[bash]sudo lshw[/bash] will provide infor on your system, CPU types, 32 v 64 bit…

Checklist: Moving WordPress site to a New Host

Checklist for moving an existing WordPress site to a new web host

A checklist for moving an existing WordPress site to a new VPS web host, when you have full admin rights over the server.

  1. Set DNS TTL’s down to 5 minutes (a few days prior to the move). This will allow the nameserver update to propagate more quickly.
  2. Set up new domain on the new host (Checklist for setting up a new domain on your VPS)
  3. Install WordPress on new host
  4. Copy old content directory to new WordPress host
  5. Copy old database to new host (how to use secure copy to copy database to new server)
  6. Update wp-config on new host
  7. Enable mod_rewrite in Apache on the new server. From the command line:
  8. a2enmod rewrite

  9. Restart apache
    /etc/init.d/apache2 restart
  10. Test that everything works (you can change your host file to test things out easily)
  11. Update registrar to point domain to the nameservers for the new host

Related: create a new database and run .sql fileDisplay text based on if it is the WordPress blog home page

If WordPress is up to date you could also just copy over everything for steps 1 and 2. I am using this for several sites I have had for years, I figure starting with a clean install of WordPress is a good idea, but it is not necessary.

MySQL cli Syntax

Some MySQL cli syntax examples

How to create a MySQL database and import tables, data… from a sql file.

Login to the MySQL command line interface

[bash]mysql -u<username> -p[/bash]

[sql]mysql -uroot -p
mysql> create database some_database;
mysql> use some_database;
mysql> source some_database.sql;[/sql]

The third line runs the named sql file for the database you names.

Create a new user

[sql]mysql -uroot -p
mysql> USE ‘some_database’;
mysql> CREATE USER ‘new_user’@’localhost’ IDENTIFIED BY ‘user_password’;
Query OK, 0 rows affected (0.00 sec)
mysql> GRANT SELECT,INSERT,UPDATE,DELETE ON some_database.* TO ‘new_user’@’localhost’;
[/sql]

For some reason MySQL chooses to say 0 rows affected, when it succeeds and ads a row for the user. Anyway if you get that message, it worked.

Change password for a user

[sql]mysql -uroot -p
mysql> SET PASSWORD FOR ‘username’@’localhost’ = PASSWORD(‘new_password’);[/sql]

Restart mysql: [bash]sudo /etc/init.d/mysql restart[/bash]