[PD Disaggregation] Write the cache of preempted req to storage and refine PD Disaggregation (#7107)

* Write the cache of preempted req to storage * up * fix
2026-04-23 00:17:25 +08:00 · 2026-04-01 13:15:52 +08:00
parent 3651113ee5
commit af51fc46d6
5 changed files with 35 additions and 19 deletions
@@ -910,7 +910,7 @@ class EngineService:
                                self.split_connector.send_splitwise_tasks([task], task.idx)
                                status, msg = self.split_connector.check_decode_allocated(task)
                                if not status:
-                                    self.llm_logger.error(
+                                    self.llm_logger.warning(
                                        f"D failed to allocate resource for request {task.request_id}, try again."
                                    )
                                    time.sleep(0.05)