add zstd and snappy compression for query api #6848
Conversation
Signed-off-by: Ahmed Hassan <[email protected]>
Signed-off-by: Ahmed Hassan <[email protected]>
Signed-off-by: Ahmed Hassan <[email protected]>
Signed-off-by: Ahmed Hassan <[email protected]>
Signed-off-by: Ahmed Hassan <[email protected]>
|
@yeya24 @SungJin1212 could someone take a look when you have some time? Thanks! |
|
@afhassan |
Signed-off-by: Ahmed Hassan <[email protected]>
Signed-off-by: Ahmed Hassan <[email protected]>
Signed-off-by: Ahmed Hassan <[email protected]>
|
@SungJin1212 Thanks for the review, I added unit tests and an integration test for compression. The change done to QFE response size limit is already covered by this test https://github.com/afhassan/cortex/blob/master/integration/query_frontend_test.go#L977 |
| @@ -102,8 +110,23 @@ func (c instantQueryCodec) DecodeResponse(ctx context.Context, r *http.Response, | |||
| return nil, err | |||
| } | |||
|
|
|||
| responseSize := 0 | |||
| responseSizeHeader := r.Header.Get("X-Uncompressed-Length") | |||
There was a problem hiding this comment.
This change means that we need a two phase deployment for querier to return this header for Query Frontend to properly apply the response size limit. Otherwise, the value is 0.
I think we should find a way (maybe a flag) to properly maintain the previous behavior during rollout.
There was a problem hiding this comment.
I didn't quite get it, do you mean during deployment if Query Frontend is updated first then the limit wouldn't apply because the header is not being returned from Querier?
The old behaviour only works for gzip which is why I changed it. I can reverse it to use the old behaviour when gzip is used and the new one for other compression types.
There was a problem hiding this comment.
Yes. If we use gzip then this limiter won't work until Querier is updated, which is a breaking behavior.
We just need a way to fallback to the old behavior if it is gzip. Other types are fine since they are new
There was a problem hiding this comment.
If header is missing we now fallback to applying limit using size of the response after decoding it. I think this is a cleaner way to still apply the limit and avoid a breaking change if the header is not there.
Applying the limit before decoding response was an optimization in the old behaviour, but I don't think it is considered breaking if we are still applying the limit.
Signed-off-by: Ahmed Hassan <[email protected]>
Signed-off-by: Ahmed Hassan <[email protected]>
Signed-off-by: Ahmed Hassan <[email protected]>
* add zstd and snappy compression for query api Signed-off-by: Ahmed Hassan <[email protected]> * parse X-Uncompressed-Length only if header exists Signed-off-by: Ahmed Hassan <[email protected]> * fix formatting Signed-off-by: Ahmed Hassan <[email protected]> * refactor query decompression Signed-off-by: Ahmed Hassan <[email protected]> * ensure zstd reader is closed after decompression Signed-off-by: Ahmed Hassan <[email protected]> * add tests for zstd and snappy compression Signed-off-by: Ahmed Hassan <[email protected]> * update changelog Signed-off-by: Ahmed Hassan <[email protected]> * update docs Signed-off-by: Ahmed Hassan <[email protected]> * apply query response size limit after decompression if header is missing Signed-off-by: Ahmed Hassan <[email protected]> * fix formatting Signed-off-by: Ahmed Hassan <[email protected]> --------- Signed-off-by: Ahmed Hassan <[email protected]> Signed-off-by: Ahmed Hassan <[email protected]>
What this PR does:
X-Uncompressed-Lengthto enforce QFE response size limit before decompressing responseWhich issue(s) this PR fixes:
Fixes #
Checklist
CHANGELOG.mdupdated - the order of entries should be[CHANGE],[FEATURE],[ENHANCEMENT],[BUGFIX]