fix(sidecar): strip ec_transfer_params from decode request in EPD mode#1704
Draft
revit13 wants to merge 2 commits into
Draft
fix(sidecar): strip ec_transfer_params from decode request in EPD mode#1704revit13 wants to merge 2 commits into
revit13 wants to merge 2 commits into
Conversation
Signed-off-by: Revital Sur <eres@il.ibm.com>
This file contains hidden or 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
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
What type of PR is this?
/kind bug
What this PR does / why we need it:
In EPD (Encoder-Prefill-Decode) mode, every decode request arrived at the
decode pod carrying
ec_transfer_params, causing vLLM to log a warning perrequest:
Requests succeeded (200 OK), but the warning is incorrect — the decode pod
has no EC connector and does not use these params.
Root cause
handleNIXLV2receives a request body thatrunPDPipelinealreadyserialized with
ec_transfer_paramspresent (injected by the encoderstage). The same
completionRequestmap is mutated in place for both theprefill and decode stages. Encoder-only fields were never removed before
marshaling the decode body.
Fix
Delete
ec_transfer_paramsfromcompletionRequestwhen preparing thedecode stage, before setting
kv_transfer_paramsand marshaling the decodebody (
connector_nixlv2.go).Test
Added a Ginkgo integration test in
connector_nixlv2_test.gothat drivesthe full EPD path (encoder backend + prefill mock + decode mock) and
asserts:
ec_transfer_params(correct — prefiller needs it)ec_transfer_params(the fix)The test fails before the fix and passes after.
Which issue(s) this PR fixes:
Fixes #1703
Release note (write
NONEif no user-facing change):