New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
Add cluster-aware quota limits for concurrent long-running requests #4662
Conversation
What is the difference between GlobalNamespaceReplicationInducingAPIsRPS and GlobalNamespaceNamespaceReplicationInducingAPIsRPS? |
This was a mistake. There should only be one. I left the old one and removed the new one |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
I don't think ns count limit is for protecting DB resource. I am under the impression that those limit exists to protect resources like network connections, file descriptors on a host. But please correct me if I am wrong.
This makes sense to me. I'm going to close this |
I'm keeping the refactoring changes though and putting them in a separate PR here: #4665 |
What changed?
I added some cluster-aware limits on the number of concurrent long-running requests and namespace-replication-inducing requests. I also did a lot of refactoring and added a bunch of documentation around here to make this clearer because the "namespace count" terminology was not very descriptive.
Why?
These limits are designed to protect database resources, so if we just use the per-instance limits, we could end up with a total limit which is very high when there are many instances, or, conversely, too low if there are few instances.
How did you test it?
I added unit tests.
Potential risks
Is hotfix candidate?