Disaster recovery plan

[travishathaway]

What the idea?

This issue will document steps we can take if the plugin begins malfunctioning in production.

What should be done?

• In a comment, write an issue template that can be used with a check list of items to be completed in the case that the plugin begins malfunctioning in production systems
• Spike the possibility of programming a "kill switch" into the plugin itself; report findings as a comment on this issue

[travishathaway]

The procedure should be followed if the plugin is deemed to be in an unstable state and should be prevent from being installed further on users systems:

These should be done immediately:

• Provide instructions for users being affected by this plugin for how to remove it from their systems
  • Instructions should be posted to community forums:
    • https://github.com/ContinuumIO/anaconda-issues/
    • https://community.anaconda.cloud/
• Remove current Miniconda and Anaconda Distribution installations that include this plugin; revert latest to previous releases
• Revoke the current version of the plugin on the defaults channel

These should be done later as time permits:

• Patch the plugin to fix the error that was encountered and release to defaults channel
• Create new Miniconda and Anaconda Distribution releases that do not include the plugin or include a patched version of it

[tferguson47]

Team reviewed checklist today and approved