fix(ml): limit load retries #10494
Merged
fix(ml): limit load retries #10494
Conversation
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
jrasm91
approved these changes
Jun 20, 2024
claabs
pushed a commit
to claabs/immich-machine-learning-openvino-kernel-fix
that referenced
this pull request
Aug 11, 2024
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
Description
The ML service tries to load a model, then on failure clears the cache and retries once. However, a new request will mean another two attempts to load. Given that the service can receive many requests for a model, failing to load/download can cause an excessive number of attempts.
This PR limits a single model instance to one retry across its lifetime. After a model is unloaded or the service restarts, it will be able to re-attempt it.
How Has This Been Tested?
Tested by making a model throw an error on the first attempt, but not the second, and confirming that it still loaded. If I made it throw an error every time, it didn't try to load the model repeatedly.