dnsdist: set backend server selection policy to chashed to optimize cache usage #158
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Outdated suggestions cannot be applied.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.
Set the backend server selection to chashed, which concentrates queries for the same domains from any dnsdist to the same recursors, improving the cache usage on the recursors.
The documentation recommends to set the weights between 100-1000, with 100 the distribution was a bit imbalanced, with 500 it is good (see first screenshot).
Here are a few monitoring screenshots showing the effects:
(The first day shown having the old wrandom active, then two days with chashed, and then back to wrandom)While on average the incoming requests to the recursors stayed roughly the same:
The outgoing queries decreased a bit:
The response times decreased a bit:
And the packet cache & recursor cache hit ratio increased:
While (recursor) cache size decreased:
https://dnsdist.org/guides/serverselection.html