New Features & API changes
- Add the ability to specify versions for clients in benchmark scenarios
You can call self.clients(“glance”, “2”) and get client initialized for specific API version.
- Add API for tempest uninstall
$ rally-manage tempest uninstall # removes fully tempest for active deployment
- Add a –uuids-only option to rally task list
$ rally task list –uuids-only # returns list with only task uuids
- Adds endpoint to –fromenv deployment creation
$ rally deployment create –fromenv
# recognizes standard OS_ENDPOINT environment variable
- Configure SSL per deployment
Now SSL information is deployment specific not Rally specific and rally.conf option is deprecated. Take a look at sample.
- Proposal for new task input file format
This spec describes new task input format that will allow us to generate multi scenario load which is crucial for HA and more real life testing. For more details take a look at spec.
Benchmark Scenario Runners:
- [improved] rps runner
Add a maximum concurrency option to rps runner.This allows to avoid running to heavy load you can set ‘concurrency’ to configuration and in case if cloud is not able to process all requests it won’t start more parallel requests then ‘concurrency’ value.
- [new] CeilometerAlarms.create_alarm_and_get_history
- [new] KeystoneBasic.get_entities
- [new] EC2Servers.boot_server
- [new] KeystoneBasic.create_and_delete_service
- [new] MuranoEnvironments.list_environments
- [new] MuranoEnvironments.create_and_delete_environment
- [new] NovaServers.suspend_and_resume_server
- [new] NovaServers.pause_and_unpause_server
- [new] NovaServers.boot_and_rebuild_server
- [new] KeystoneBasic.create_and_list_services
- [new] HeatStacks.list_stacks_and_events
- [improved] VMTask.boot_runcommand_delete
Restore ability to use fixed IP and floating IP to connect to VM via ssh
- [fix] NovaServers.boot_server_attach_created_volume_and_live_migrate
Kwargs in nova scenario were wrongly passed
- [new] aborted_on_sla
This is internal SLA criteria, that is added if task was aborted
- [new] something_went_wrong
This is internal SLA criteria, that is added if something went wrong,
context failed to create or runner raised some exceptions
14 bugs were fixed, the most critical are:
- Set default task uuid to running task. Before it was set only after
task was fully finished.
- The “rally task results” command showed a disorienting “task not found”
message for a task that is currently running.
- Rally didn’t know how to reconnect to OpenStack in case if token
- New tutorial about task templates
- Various fixes