Improve scalability of preview system when running on Kubernetes environment.
Description
Release Notes
Improved the scalability of the preview system when running in Kubernetes environment by separating out preview runs in their own individual pods. Preview manager pod now only responsible for handling preview REST api.
87% Done
is duplicated by
Activity
Show:
Sagar Kapare August 15, 2020 at 7:20 AM
Since tasks for 6.1 are done. Marking this umbrella bug as fixed. Will open new for any changes on the 6.3 and 6.2.2
Fixed
Pinned fields
Click on the next to a field label to start pinning.
Created April 29, 2020 at 2:53 AM
Updated December 9, 2020 at 9:30 PM
Resolved August 15, 2020 at 7:20 AM
This bug is to track the design and implementation of the scalable preview system.
On high level -
Preview runner pods will be responsible for running actual preview job.
These pods will be exposed as Kubernetes service object.
Preview Manager will talk to the service to start/stop the preview runs.
Tracing data along with logs will be shipped back to the preview manager by using TMS.