-
|
Any best practices from the community related to image management? Do you separate production images from development ones? How about naming conventions; do you add information about for example development branch name in the image tag? Do you use git hash to uniquely tag images? How do you keep track of images for multi-image workflows?
|
Beta Was this translation helpful? Give feedback.
Replies: 1 comment
-
|
So one thing I have seen multiple folks use is to create a new image per git sha, with the name as the name of the repo. Retention is managed based on what’s merged to master, where non-master/main branch images are purged more frequently. For multi-image, Flytesnacks is an example. We still call the image cookbook, and the base image is simply tagged with the sha or semver. |
Beta Was this translation helpful? Give feedback.
So one thing I have seen multiple folks use is to create a new image per git sha, with the name as the name of the repo.
Retention is managed based on what’s merged to master, where non-master/main branch images are purged more frequently.
For multi-image, Flytesnacks is an example. We still call the image cookbook, and the base image is simply tagged with the sha or semver.
Individual images are tagged with usecase-sha. Only the usecase is variable and Flytekit has a simple way for tasks to declare they want a specific image.
@kumare3