This is server-side software for managing new package issues submitted to the 'Contribution' Bioconductor github repository.
The software runs on an Amazon nano-instance at issues.bioconductor.org, accessible via ssh.
Setup is via a chef recipe. Update the recipe by commiting to the
github repository, then berks install. On the
issues.bioconductor.org
sudo su - www-data
chef-client
The software notifies the single package builder of the repository to retrieve the package from, and the package is build across Linux, Mac, and Windows.
- issues.bioconductor.org
- checkip.amazonaws.com
- EC2 Management Console
- NETWORK & SECURITY
- ssh; edit inbound rules
- ssh ubuntu@issues.bioconductor.org sudo su - www-data
The software is running under apache. Error and access logs are at
/var/log/apache2 as ubuntu.
Use pry for interactive ruby. Ensure that the approporiate pry is
available (via rbenv)
$ which pry
/var/www/.rbenv/shims/rpy
start pry (from the home directory of www-data) and require the core
application
$ pry
[1] pry(main)> require_relative './app/core'
=> true
[2] pry(main)>
Make changes in personal clone of git repository, push to github.com/Bioconductor/issue_tracker_github.
On ubuntu@issues.bioconductor.org
sudo chef-client
There are authentication credentials encoded by chef data bags. See git credentials for keys to authorize changes. To show contains of a data bag:
knife data bag show IssueTrackerConfig IssueTrackerConfig --secret-file /home/lori/Documents/databag.key
To edit the contents of the data bag:
knife data bag edit IssueTrackerConfig IssueTrackerConfig --secret-file /home/lori/Documents/databag.key
The data bag should remain encrypted but to test run the show command without including --secret-file and it should show encrypted contents.
You can list all managed data bags by doing
knife data bag list
The aws keys correspond to the sdk-email user security credentials.