While OpenStack Release notes is a great source to get updated quickly on the changes and new features that are introduced in each release, the Release Notes do not provide much details on how to use and leverage the new features. Here’s a bit more detailed description of what’s new with HEAT in OpenStack Juno release and expected in the supported product “RHEL OpenStack Platform 6” to be released in late January 2015.
Recovery From Failures During Stack Updates
Heat allows to update existing stacks, while stack updates occur new resources are created created and added. If a failure happens during Stack Update Heat can attempt to roll back to the previous state or keep the Stack at failed state. If the roll-back fails or if roll-back is disabled for a Stack, Heat now tracks all changes that happen in Stack’s resources, so instead of just failing a Stack Update operation and keeping it in failed state,where in this state all we can do is to delete a stack, Heat now keeps track of all changes in resources and updates the current template accordingly.
API to cancel and roll back an in-progress stack update
Heat now allows to cancel a Stack Update operation while stack update is in progress.
usage: heat stack-cancel-update <NAME or ID>
Remember the previously-supplied parameters when updating a stack
Heat Update now can re-use the set of parameters of the current stack. Parameters specified in –parameters will patch over the existing values in the current stack. Parameters omitted will keep the existing values when –existing flag is specified.
Usage: heat stack-update --existing <NAME or ID>
Improved visibility into trees of nested stacks
Heat API and CLI added a new flag to list all stacks including nested stacks.
Usage: heat stack-list --nested
Implementation of new resource types:
New resource types that were added in the Juno release cycle to the already large list of existing resources you can use within Heat Orchestration Templates:
Provides the option to store Wait Condition (and Software Deployment) data in Swift
Groups code for multiple lifecycle events into a single deployment resource
Provides a way of avoiding circular dependencies when deploying an interdependent cluster of servers
All new Implemented HEAT blueprints in Juno:
What else are you interested in related to OpenStack HEAT? Let me know in the comments below.