From f2caefe226e4c3283d1140aa16125748d1130027 Mon Sep 17 00:00:00 2001 From: Wei Zhao <51183510+wzhao18@users.noreply.github.com> Date: Thu, 28 May 2026 06:31:45 -0400 Subject: [PATCH] [UX] Increase DP Coordinator startup timeout from 30s to 120s (#42343) Signed-off-by: wzhao18 --- vllm/v1/engine/coordinator.py | 5 +++-- 1 file changed, 3 insertions(+), 2 deletions(-) diff --git a/vllm/v1/engine/coordinator.py b/vllm/v1/engine/coordinator.py index 87937cc6a8..151a07abe2 100644 --- a/vllm/v1/engine/coordinator.py +++ b/vllm/v1/engine/coordinator.py @@ -58,13 +58,14 @@ class DPCoordinator: def _wait_for_zmq_addrs(self, zmq_addr_pipe) -> tuple[str, str, str]: try: + timeout = 120 ready = multiprocessing.connection.wait( - [zmq_addr_pipe, self.proc.sentinel], timeout=30 + [zmq_addr_pipe, self.proc.sentinel], timeout=timeout ) if not ready: raise RuntimeError( "DP Coordinator process failed to report ZMQ addresses " - "during startup." + f"within timeout={timeout} seconds during startup." ) try: return zmq_addr_pipe.recv()