Knowledge Base

Confluence - where, what, how

Accessing Confluence

Once you have installed Confluence using the Deploy4Me service you can work with it using the URL from the passwords. Commonly the URL will be:

http://ec2-....compute.amazonaws.com

If the cloud hosting didn't provide a public DNS name then you can use IP address to access your server and check it:

http://54.....120

If you have defined the custom DNS alias during the deployment, then use it like:

http://my-alias-to-confluence.com

Note that DNS record should point to the server public IP address. To prevent problems with DNS you may use Elastic IP, assign it to your instance and point your DNS record to it.

I see an error 502 or 503. What to do?

Deploy4Me installs Confluence behind the Apache proxy. On start of the new server it will take some time for Confluence to start accepting the requests. So, just refresh a page couple of times and you will see the Confluence welcome screen. If it is not helping then contact Deploy4Me support.

Where Confluence is installed?

Deploy4Me would set Confluence home to

/var/atlassian/application-data/confluence

Confluence installation files are placed to

/opt/atlassian/confluence

Confluence is installed behind the proxy. Apache is used as a proxy server. The configuration of Apache server can be found at (depending on system)

$(which apache2)/sites-available/Confluence.conf
$(which httpd)/sites-available/Confluence.conf

Confluence internally uses Tomcat as an application server. The configuration of Tomcat server can be found at

/opt/atlassian/confluence/conf/server.xml

How to stop/start/restart Confluence?

Confluence is deployed as a service, so use one of the Linux commands:

service confluence stop
service confluence start
service confluence restart

Confluence is not available after restart of the server on AWS. What to do?

Commonly the problem not appears but sometimes Amazon Web Services EC2 do not preserve the public DNS names for servers after restart. This means that you would rather reconfigure Nginx to use a new DNS name generated by AWS or configure a permanent DNS name for your server using Elastic IP address. To configure a permanent DNS name here is the step by step instruction:

  1. Allocate an Elastic IP via AWS EC2 console
  2. Associate your address with the instance using context menu on Elastic IP
  3. Configure your DNS to point on that public IP address with a new DNS record. You would possibly wish to create a new A record in the DNS configuration.
  4. If you provided the DNS name before deployment then skip this step, else connect to the server and make sure that in Apache configuration under
$(which apache2)/sites-available/Confluence.conf
$(which httpd)/sites-available/Confluence.conf

you have a valid name of a DNS record. Change it (for example using vim) and reload the Apache

service apache2 restart
service httpd restart
  1. Double check the Confluence's Tomcat config file under (proxyName parameter in Connector section)
/opt/atlassian/confluence/conf/server.xml
  1. Make sure your Confluence is started and access the URL in a browser
service confluence restart

Confluence is not working on 1GB droplet. What to do?

Commonly we do not recommend the 1GB installation, especially for a heavy software like Confluence. But if you have decided to make such installation then you would have to change the JVM memory limits:

  1. Connect to Confluence machine via SSH
  2. From the command line do the following:
sudo su
cd /opt/atlassian/confluence/bin
  1. Edit file setenv.sh to change value in JVM_MAXIMUM_MEMORY to 512m or smaller if required
  2. Execute
service confluence stop
./setenv.sh
service confluence restart