monkeypatch kubernetes to avoid ThreadPool problems #169
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
avoids creating a new client for each reflector
…client kubernetes creates client objects in a number of places e.g. every time a Watch is instantiated. These objects create a max-size thread pool for async requests, which bogs down the process if enough client objects are instantiated. Since we don't use async requests, these pools go totally unused. api_client is monkeypatched to avoid creating these ThreadPools in the first place. A patch has been submitted upstream to swagger-codegen, which is responsible for creating the ThreadPool.
|
How about keeping an instance of |
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Outdated suggestions cannot be applied.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.
kubernetes creates client objects in a number of places e.g. every time a Watch is instantiated.
These objects create a max-size thread pool for async requests, which bogs down the process if enough client objects are instantiated (#165). Since we don't
use async requests, these pool objects go totally unused.
We setup shared clients to reduce the number of client objects created in #128, but there are some places we can't prevent clients from being instantiated (Watches), so the new EventReflector caused a problem in a big way by instantiating N-CPUs threads for every spawn.
To solve this at the root, api_client is monkeypatched to avoid creating these ThreadPools in the first place. A patch has been submitted upstream to swagger-codegen, which is responsible for creating the ThreadPool in the kubernetes client.
Related to #153, which creates a Watch per spawn, so we cannot avoid creating a very large number of ApiClient objects and thereby threadpools without monkeypatching Kubernetes.
The alternative is to go back to pinning kubernetes-3.0.