Amazon Instance configuration

From genomewiki
Jump to: navigation, search

Apache Modifications

On the amazon instance that you want to turn into a genome browser, add ucsc.conf to the directory /etc/httpd/conf.d and turn off the vhosts

# cd /etc/httpd/conf.d
# mv vhosts.conf vhosts.conf.0
# cat /etc/httpd/conf.d/ucsc.conf
#
# Executable files will be processed for SSI
#
XBitHack on

Alterations to /etc/httpd/conf/httpd.conf for the options to Directory /var/www/html

Options Indexes FollowSymLinks MultiViews Includes

Change ReadmeName and HeaderName to:

ReadmeName FOOTER
HeaderName README

Add handler:

AddHandler server-parsed .shtml

Add UCSC error documents:

ErrorDocument 403 /admin/403.html
ErrorDocument 404 /admin/404.html
ErrorDocument 500 /admin/500.html

And turn off VirtualHost business, comment this line out:

# NameVirtualHost *:80

Fetch UCSC html hierarchy:

# cd /var/www/html
# rsync -avzP rsync://hgdownload.cse.ucsc.edu/htdocs/ ./

Start Apache:

# cd /etc/init.d
# ./httpd start

Can now open URL to top-level page to see index.html The Browser will not yet work, other items need to be installed, MySQL and /gbdb/ business in particular.

User Accounts

Create group for browser development:

# groupadd -g 83492 ucscbrowser

Create user for browser development:

# useradd -c "UCSC Genome Browser Development" -d /home/browserdev \
     -g ucscbrowser -m -u 29438 -n -s /bin/bash browserdev

Set a password for that user, expires in 90 days:

# passwd browserdev
# passwd -x 90 browserdev