{"record":{"id":"0e16666b29ae6251","repo":"sgl-project/sglang","slug":"sglang-disagg-staging-buffer-does-not-support-pref","errorCode":null,"errorMessage":"SGLANG_DISAGG_STAGING_BUFFER does not support prefill context parallelism.","messagePattern":"SGLANG_DISAGG_STAGING_BUFFER does not support prefill context parallelism\\.","errorType":"validation","errorClass":"RuntimeError","httpStatus":null,"severity":"error","filePath":"python/sglang/srt/disaggregation/prefill.py","lineNumber":185,"sourceCode":"            # and the grid cannot disagree after a post-publish override.\n            chunked_prefill_size = get_schedule().chunked_prefill_size\n            cps = chunked_prefill_size or 8192\n            # Staging slices each send into a fixed page-aligned grid, so an\n            # unbounded (-1) or non-page-aligned chunk size has no valid grid.\n            if cps <= 0 or cps % page_size != 0:\n                raise RuntimeError(\n                    f\"SGLANG_DISAGG_STAGING_BUFFER requires a positive \"\n                    f\"chunked_prefill_size that is a multiple of page_size \"\n                    f\"({page_size}); got {chunked_prefill_size}.\"\n                )\n            if self.pp_size > 1 and self.transfer_backend != TransferBackend.MOONCAKE:\n                raise RuntimeError(\n                    \"SGLANG_DISAGG_STAGING_BUFFER with pp_size > 1 is only \"\n                    \"supported by Mooncake.\"\n                )\n            if get_parallel().enable_prefill_context_parallel:\n                # CP rewrites index_slice per rank, breaking the chunk grid.\n                raise RuntimeError(\n                    \"SGLANG_DISAGG_STAGING_BUFFER does not support \"\n                    \"prefill context parallelism.\"\n                )\n        self.kv_manager = self._init_kv_manager()\n\n    def _init_kv_manager(self) -> CommonKVManager:\n        kv_args_class = get_kv_class(self.transfer_backend, KVClassType.KVARGS)\n        kv_args = kv_args_class()\n        kv_args.engine_rank = self.tp_rank\n        kv_args.pp_rank = self.pp_rank\n        kv_args.system_dp_rank = self.scheduler.ps.dp_rank\n        kv_args.kv_cache_dtype_str = (\n            self.scheduler.tp_worker.model_runner.kv_cache_dtype_str\n        )\n        layer_shard_enabled = getattr(\n            self.token_to_kv_pool, \"layer_shard_enabled\", False\n        )\n        layer_shard_rank = getattr(self.token_to_kv_pool, \"layer_shard_rank\", None)","sourceCodeStart":167,"sourceCodeEnd":203,"githubUrl":"https://github.com/sgl-project/sglang/blob/0132848349585cfe6aae51c4941cbae872505f8a/python/sglang/srt/disaggregation/prefill.py#L167-L203","documentation":"Prefill context parallelism (CP) rewrites index_slice per rank, which breaks the fixed page-aligned chunk grid that SGLANG_DISAGG_STAGING_BUFFER relies on, so enabling both is rejected at startup in the prefill controller.","triggerScenarios":"SGLANG_DISAGG_STAGING_BUFFER=1 with get_parallel().enable_prefill_context_parallel true (e.g. --enable-prefill-context-parallel or equivalent server arg).","commonSituations":"Long-context workloads enabling prefill CP for throughput while also enabling the staging buffer optimization for KV transfer in a disaggregated deployment.","solutions":["Disable prefill context parallelism (remove the CP flag) when using SGLANG_DISAGG_STAGING_BUFFER.","Or unset SGLANG_DISAGG_STAGING_BUFFER and keep context parallelism."],"exampleFix":"# before\nexport SGLANG_DISAGG_STAGING_BUFFER=1\npython -m sglang.launch_server --enable-prefill-context-parallel ...\n# after (choose one optimization)\nunset SGLANG_DISAGG_STAGING_BUFFER\npython -m sglang.launch_server --enable-prefill-context-parallel ...","handlingStrategy":"validation","validationCode":"if envs.SGLANG_DISAGG_STAGING_BUFFER.get():\n    assert not get_parallel().enable_prefill_context_parallel, \\\n        'staging buffer is incompatible with prefill CP'","typeGuard":null,"tryCatchPattern":null,"preventionTips":["Document mutually exclusive perf flags in the deploy runbook.","Assert flag compatibility in a preflight config check before server start."],"tags":["sglang","disaggregation","context-parallelism","staging-buffer","config-validation"],"backgroundTag":"unsupported-feature-configuration","analyzedSha":"0132848349585cfe6aae51c4941cbae872505f8a","analyzedAt":"2026-08-28T05:10:05.995Z","schemaVersion":2},"datasetVersion":"2026-08-28T06:17:29.519Z"}