From 8dfd215060432aa700a10cc3545d04815a9d529d Mon Sep 17 00:00:00 2001 From: GLAMR Date: Wed, 16 Sep 2026 14:06:49 -0400 Subject: [PATCH] fix(sglang): stop the elastic-EP scale-up worker crash-looping at startup (#14568) Signed-off-by: svc-glamr@nvidia.com Co-authored-by: yunzhoul-nv <232973175+yunzhoul-nv@users.noreply.github.com> Signed-off-by: Tzu-Ling --- .../templates/sglang/run_sglang_elastic_ep_scale_up_test.sh | 1 + .../deploy/templates/sglang/sglang_elastic_ep.yaml | 6 ++++++ 2 files changed, 7 insertions(+) diff --git a/tests/fault_tolerance/deploy/templates/sglang/run_sglang_elastic_ep_scale_up_test.sh b/tests/fault_tolerance/deploy/templates/sglang/run_sglang_elastic_ep_scale_up_test.sh index daf35fe38863..38cc21e93745 100755 --- a/tests/fault_tolerance/deploy/templates/sglang/run_sglang_elastic_ep_scale_up_test.sh +++ b/tests/fault_tolerance/deploy/templates/sglang/run_sglang_elastic_ep_scale_up_test.sh @@ -167,6 +167,7 @@ launch_joiner() { --elastic-ep-backend mooncake --mooncake-ib-device $IB_DEVICE \ --enable-eplb --ep-num-redundant-experts 24 \ --elastic-ep-initial-size 4 --max-ep-size 8 \ + --cuda-graph-backend-decode disabled --cuda-graph-backend-prefill disabled \ --elastic-ep-join-mode scale --elastic-ep-join-rank-offset $rank_offset \ --dist-init-addr $DIST_INIT_ADDR \ --host 127.0.0.1 --port $((30000 + rank_offset)) \ diff --git a/tests/fault_tolerance/deploy/templates/sglang/sglang_elastic_ep.yaml b/tests/fault_tolerance/deploy/templates/sglang/sglang_elastic_ep.yaml index 4f7a5c281be4..50f753ea2076 100644 --- a/tests/fault_tolerance/deploy/templates/sglang/sglang_elastic_ep.yaml +++ b/tests/fault_tolerance/deploy/templates/sglang/sglang_elastic_ep.yaml @@ -31,6 +31,8 @@ # --mooncake-ib-device to the node's IB device list before applying. # * DYN_SYSTEM_ENABLED=true so the leader exposes the :9090 system server that # serves /engine/control/*. +# * BOTH the decode and the prefill CUDA graph disabled, passed explicitly +# below: SGLang asserts this for elastic-EP scale-up and exits on startup. # # The joining group is NOT launched by the operator yet (that orchestration is # tracked separately). The accompanying run_sglang_elastic_ep_scale_up_test.sh @@ -128,6 +130,10 @@ spec: - "4" - --max-ep-size - "8" + - --cuda-graph-backend-decode + - disabled + - --cuda-graph-backend-prefill + - disabled # Stable rendezvous address the joining group dials on scale-up. # Sourced from SGLANG_ELASTIC_SCALE_DIST_INIT (single source of truth). - --dist-init-addr