[Bug][v0.5.0]: Benign error reported by Python multiprocessing resource_tracker

[mgoin]

Your current environment

python collect_env.py
Collecting environment information...
PyTorch version: 2.3.0+cu121
Is debug build: False
CUDA used to build PyTorch: 12.1
ROCM used to build PyTorch: N/A

OS: Ubuntu 22.04.4 LTS (x86_64)
GCC version: (Ubuntu 11.4.0-1ubuntu1~22.04) 11.4.0
Clang version: Could not collect
CMake version: version 3.29.5
Libc version: glibc-2.35

Python version: 3.10.12 (main, Nov 20 2023, 15:14:05) [GCC 11.4.0] (64-bit runtime)
Python platform: Linux-6.5.0-35-generic-x86_64-with-glibc2.35
Is CUDA available: True
CUDA runtime version: 12.5.40
CUDA_MODULE_LOADING set to: LAZY
GPU models and configuration: 
GPU 0: NVIDIA H100 80GB HBM3
GPU 1: NVIDIA H100 80GB HBM3
GPU 2: NVIDIA H100 80GB HBM3
GPU 3: NVIDIA H100 80GB HBM3
GPU 4: NVIDIA H100 80GB HBM3
GPU 5: NVIDIA H100 80GB HBM3
GPU 6: NVIDIA H100 80GB HBM3
GPU 7: NVIDIA H100 80GB HBM3

Nvidia driver version: 555.42.02
cuDNN version: Could not collect
HIP runtime version: N/A
MIOpen runtime version: N/A
Is XNNPACK available: True

CPU:
Architecture:                       x86_64
CPU op-mode(s):                     32-bit, 64-bit
Address sizes:                      46 bits physical, 57 bits virtual
Byte Order:                         Little Endian
CPU(s):                             128
On-line CPU(s) list:                0-127
Vendor ID:                          GenuineIntel
Model name:                         Intel(R) Xeon(R) Platinum 8462Y+
CPU family:                         6
Model:                              143
Thread(s) per core:                 2
Core(s) per socket:                 32
Socket(s):                          2
Stepping:                           8
CPU max MHz:                        4100.0000
CPU min MHz:                        800.0000
BogoMIPS:                           5600.00
Flags:                              fpu vme de pse tsc msr pae mce cx8 apic sep mtrr pge mca cmov pat pse36 clflush dts acpi mmx fxsr sse sse2 ss ht tm pbe syscall nx pdpe1gb rdtscp lm constant_tsc art arch_perfmon pebs bts rep_good nopl xtopology nonstop_tsc cpuid aperfmperf tsc_known_freq pni pclmulqdq dtes64 monitor ds_cpl vmx smx est tm2 ssse3 sdbg fma cx16 xtpr pdcm pcid dca sse4_1 sse4_2 x2apic movbe popcnt tsc_deadline_timer aes xsave avx f16c rdrand lahf_lm abm 3dnowprefetch cpuid_fault epb cat_l3 cat_l2 cdp_l3 invpcid_single intel_ppin cdp_l2 ssbd mba ibrs ibpb stibp ibrs_enhanced tpr_shadow flexpriority ept vpid ept_ad fsgsbase tsc_adjust bmi1 avx2 smep bmi2 erms invpcid cqm rdt_a avx512f avx512dq rdseed adx smap avx512ifma clflushopt clwb intel_pt avx512cd sha_ni avx512bw avx512vl xsaveopt xsavec xgetbv1 xsaves cqm_llc cqm_occup_llc cqm_mbm_total cqm_mbm_local split_lock_detect avx_vnni avx512_bf16 wbnoinvd dtherm ida arat pln pts hfi vnmi avx512vbmi umip pku ospke waitpkg avx512_vbmi2 gfni vaes vpclmulqdq avx512_vnni avx512_bitalg tme avx512_vpopcntdq la57 rdpid bus_lock_detect cldemote movdiri movdir64b enqcmd fsrm md_clear serialize tsxldtrk pconfig arch_lbr ibt amx_bf16 avx512_fp16 amx_tile amx_int8 flush_l1d arch_capabilities
Virtualization:                     VT-x
L1d cache:                          3 MiB (64 instances)
L1i cache:                          2 MiB (64 instances)
L2 cache:                           128 MiB (64 instances)
L3 cache:                           120 MiB (2 instances)
NUMA node(s):                       2
NUMA node0 CPU(s):                  0-31,64-95
NUMA node1 CPU(s):                  32-63,96-127
Vulnerability Gather data sampling: Not affected
Vulnerability Itlb multihit:        Not affected
Vulnerability L1tf:                 Not affected
Vulnerability Mds:                  Not affected
Vulnerability Meltdown:             Not affected
Vulnerability Mmio stale data:      Not affected
Vulnerability Retbleed:             Not affected
Vulnerability Spec rstack overflow: Not affected
Vulnerability Spec store bypass:    Mitigation; Speculative Store Bypass disabled via prctl
Vulnerability Spectre v1:           Mitigation; usercopy/swapgs barriers and __user pointer sanitization
Vulnerability Spectre v2:           Mitigation; Enhanced / Automatic IBRS; IBPB conditional; RSB filling; PBRSB-eIBRS SW sequence; BHI BHI_DIS_S
Vulnerability Srbds:                Not affected
Vulnerability Tsx async abort:      Not affected

Versions of relevant libraries:
[pip3] numpy==1.26.4
[pip3] nvidia-nccl-cu12==2.20.5
[pip3] torch==2.3.0
[pip3] transformers==4.41.2
[pip3] triton==2.3.0
[conda] Could not collect
ROCM Version: Could not collect
Neuron SDK Version: N/A
vLLM Version: 0.5.0
vLLM Build Flags:
CUDA Archs: Not Set; ROCm: Disabled; Neuron: Disabled
GPU Topology:
GPU0    GPU1    GPU2    GPU3    GPU4    GPU5    GPU6    GPU7    NIC0    NIC1    NIC2    NIC3    NIC4    NIC5    NIC6    NIC7    CPU Affinity    NUMA Affinity   GPU NUMA ID
GPU0     X      NV18    NV18    NV18    NV18    NV18    NV18    NV18    PIX     NODE    NODE    NODE    SYS     SYS     SYS     SYS     0-31,64-95      0               N/A
GPU1    NV18     X      NV18    NV18    NV18    NV18    NV18    NV18    NODE    PIX     NODE    NODE    SYS     SYS     SYS     SYS     0-31,64-95      0               N/A
GPU2    NV18    NV18     X      NV18    NV18    NV18    NV18    NV18    NODE    NODE    PIX     NODE    SYS     SYS     SYS     SYS     0-31,64-95      0               N/A
GPU3    NV18    NV18    NV18     X      NV18    NV18    NV18    NV18    NODE    NODE    NODE    PIX     SYS     SYS     SYS     SYS     0-31,64-95      0               N/A
GPU4    NV18    NV18    NV18    NV18     X      NV18    NV18    NV18    SYS     SYS     SYS     SYS     PIX     NODE    NODE    NODE    32-63,96-127    1               N/A
GPU5    NV18    NV18    NV18    NV18    NV18     X      NV18    NV18    SYS     SYS     SYS     SYS     NODE    PIX     NODE    NODE    32-63,96-127    1               N/A
GPU6    NV18    NV18    NV18    NV18    NV18    NV18     X      NV18    SYS     SYS     SYS     SYS     NODE    NODE    PIX     NODE    32-63,96-127    1               N/A
GPU7    NV18    NV18    NV18    NV18    NV18    NV18    NV18     X      SYS     SYS     SYS     SYS     NODE    NODE    NODE    PIX     32-63,96-127    1               N/A
NIC0    PIX     NODE    NODE    NODE    SYS     SYS     SYS     SYS      X      NODE    NODE    NODE    SYS     SYS     SYS     SYS
NIC1    NODE    PIX     NODE    NODE    SYS     SYS     SYS     SYS     NODE     X      NODE    NODE    SYS     SYS     SYS     SYS
NIC2    NODE    NODE    PIX     NODE    SYS     SYS     SYS     SYS     NODE    NODE     X      NODE    SYS     SYS     SYS     SYS
NIC3    NODE    NODE    NODE    PIX     SYS     SYS     SYS     SYS     NODE    NODE    NODE     X      SYS     SYS     SYS     SYS
NIC4    SYS     SYS     SYS     SYS     PIX     NODE    NODE    NODE    SYS     SYS     SYS     SYS      X      NODE    NODE    NODE
NIC5    SYS     SYS     SYS     SYS     NODE    PIX     NODE    NODE    SYS     SYS     SYS     SYS     NODE     X      NODE    NODE
NIC6    SYS     SYS     SYS     SYS     NODE    NODE    PIX     NODE    SYS     SYS     SYS     SYS     NODE    NODE     X      NODE
NIC7    SYS     SYS     SYS     SYS     NODE    NODE    NODE    PIX     SYS     SYS     SYS     SYS     NODE    NODE    NODE     X 

Legend:

  X    = Self
  SYS  = Connection traversing PCIe as well as the SMP interconnect between NUMA nodes (e.g., QPI/UPI)
  NODE = Connection traversing PCIe as well as the interconnect between PCIe Host Bridges within a NUMA node
  PHB  = Connection traversing PCIe as well as a PCIe Host Bridge (typically the CPU)
  PXB  = Connection traversing multiple PCIe bridges (without traversing the PCIe Host Bridge)
  PIX  = Connection traversing at most a single PCIe bridge
  NV#  = Connection traversing a bonded set of # NVLinks

NIC Legend:

  NIC0: mlx5_0
  NIC1: mlx5_1
  NIC2: mlx5_2
  NIC3: mlx5_3
  NIC4: mlx5_4
  NIC5: mlx5_5
  NIC6: mlx5_6
  NIC7: mlx5_7

🐛 Describe the bug

When running with tensor parallelism enabled on the multiprocessing backend (i.e. default with ray not installed) I see these errors:

Traceback (most recent call last):
  File "/usr/lib/python3.10/multiprocessing/resource_tracker.py", line 209, in main
    cache[rtype].remove(name)
KeyError: '/psm_10b3d52e'
Traceback (most recent call last):
  File "/usr/lib/python3.10/multiprocessing/resource_tracker.py", line 209, in main
    cache[rtype].remove(name)
KeyError: '/psm_10b3d52e'
Traceback (most recent call last):
  File "/usr/lib/python3.10/multiprocessing/resource_tracker.py", line 209, in main
    cache[rtype].remove(name)
KeyError: '/psm_10b3d52e'

They seemingly cause no issue during actual inference, but we should try to remove any reported errors even if benign.

Here is further context on when this is happening during creation:

/home/mgoin/venvs/vllm-rel/lib/python3.10/site-packages/huggingface_hub/file_download.py:1132: FutureWarning: `resume_download` is deprecated and will be removed in version 1.0.0. Downloads always resume when possible. If you want to force a new download, use `force_download=True`.
  warnings.warn(
WARNING 06-12 19:01:08 config.py:216] fp8 quantization is not fully optimized yet. The speed can be slower than non-quantized models.
INFO 06-12 19:01:11 config.py:623] Defaulting to use mp for distributed inference
INFO 06-12 19:01:11 llm_engine.py:161] Initializing an LLM engine (v0.5.0) with config: model='neuralmagic/Mixtral-8x7B-Instruct-v0.1-FP8', speculative_config=None, tokenizer='neuralmagic/Mixtral-8x7B-Instruct-v0.1-FP8', skip_tokenizer_init=False, tokenizer_mode=auto, revision=None, rope_scaling=None, rope_theta=None, tokenizer_revision=None, trust_remote_code=False, dtype=torch.bfloat16, max_seq_len=32768, download_dir=None, load_format=LoadFormat.AUTO, tensor_parallel_size=4, disable_custom_all_reduce=False, quantization=fp8, enforce_eager=False, kv_cache_dtype=auto, quantization_param_path=None, device_config=cuda, decoding_config=DecodingConfig(guided_decoding_backend='outlines'), seed=1234, served_model_name=neuralmagic/Mixtral-8x7B-Instruct-v0.1-FP8)
(VllmWorkerProcess pid=2584199) INFO 06-12 19:01:16 multiproc_worker_utils.py:215] Worker ready; awaiting tasks
(VllmWorkerProcess pid=2584200) INFO 06-12 19:01:17 multiproc_worker_utils.py:215] Worker ready; awaiting tasks
(VllmWorkerProcess pid=2584198) INFO 06-12 19:01:17 multiproc_worker_utils.py:215] Worker ready; awaiting tasks
(VllmWorkerProcess pid=2584198) INFO 06-12 19:01:21 utils.py:623] Found nccl from library libnccl.so.2
(VllmWorkerProcess pid=2584200) INFO 06-12 19:01:21 utils.py:623] Found nccl from library libnccl.so.2
INFO 06-12 19:01:21 utils.py:623] Found nccl from library libnccl.so.2
(VllmWorkerProcess pid=2584198) INFO 06-12 19:01:21 pynccl.py:65] vLLM is using nccl==2.20.5
(VllmWorkerProcess pid=2584200) INFO 06-12 19:01:21 pynccl.py:65] vLLM is using nccl==2.20.5
INFO 06-12 19:01:21 pynccl.py:65] vLLM is using nccl==2.20.5
(VllmWorkerProcess pid=2584199) INFO 06-12 19:01:21 utils.py:623] Found nccl from library libnccl.so.2
(VllmWorkerProcess pid=2584199) INFO 06-12 19:01:21 pynccl.py:65] vLLM is using nccl==2.20.5
Traceback (most recent call last):
  File "/usr/lib/python3.10/multiprocessing/resource_tracker.py", line 209, in main
    cache[rtype].remove(name)
KeyError: '/psm_10b3d52e'
Traceback (most recent call last):
  File "/usr/lib/python3.10/multiprocessing/resource_tracker.py", line 209, in main
    cache[rtype].remove(name)
KeyError: '/psm_10b3d52e'
Traceback (most recent call last):
  File "/usr/lib/python3.10/multiprocessing/resource_tracker.py", line 209, in main
    cache[rtype].remove(name)
KeyError: '/psm_10b3d52e'
(VllmWorkerProcess pid=2584198) INFO 06-12 19:01:24 custom_all_reduce_utils.py:179] reading GPU P2P access cache from /home/mgoin/.config/vllm/gpu_p2p_access_cache_for_0,1,2,3.json
(VllmWorkerProcess pid=2584199) INFO 06-12 19:01:24 custom_all_reduce_utils.py:179] reading GPU P2P access cache from /home/mgoin/.config/vllm/gpu_p2p_access_cache_for_0,1,2,3.json
INFO 06-12 19:01:24 custom_all_reduce_utils.py:179] reading GPU P2P access cache from /home/mgoin/.config/vllm/gpu_p2p_access_cache_for_0,1,2,3.json
(VllmWorkerProcess pid=2584200) INFO 06-12 19:01:24 custom_all_reduce_utils.py:179] reading GPU P2P access cache from /home/mgoin/.config/vllm/gpu_p2p_access_cache_for_0,1,2,3.json

[youkaichao]

Thanks for the report, I also noticed this in #5439 .

related code:

vllm/vllm/distributed/parallel_state.py

Lines 433 to 443 in 1a8bfd9

	# clean up the shared memory segment
	with contextlib.suppress(OSError):
	if rank == 0:
	if shm:
	shm.unlink()
	else:
	if shm:
	# fix to https://stackoverflow.com/q/62748654/9191338
	resource_tracker.unregister(
	shm._name, "shared_memory") # type: ignore[attr-defined]
	torch.distributed.all_reduce(is_in_the_same_node, group=pg)

Initially I added this code because Python tracks the shared memory even if it is not the creator, and it will print many warnings. After I add this, it works fine in my machine.

Not sure why it errors in some other machines.

[super-buster]

same here, H800

[youkaichao]

I will fix it tmr. Fortunately this is benign, and does not affect performance.

[majestichou]

same here, A100, vllm 0.5.0 docker start

[DefTruth]

@youkaichao same error in 0.5.0.post1, H20

Traceback (most recent call last):
  File "/usr/lib/python3.10/multiprocessing/resource_tracker.py", line 209, in main
    cache[rtype].remove(name)
KeyError: '/psm_6f36dc36'
Traceback (most recent call last):
  File "/usr/lib/python3.10/multiprocessing/resource_tracker.py", line 209, in main
    cache[rtype].remove(name)
KeyError: '/psm_6f36dc36'
Traceback (most recent call last):
  File "/usr/lib/python3.10/multiprocessing/resource_tracker.py", line 209, in main
    cache[rtype].remove(name)
KeyError: '/psm_6f36dc36'

[youkaichao]

#5512 is not included in 0.5.0.post1 . It will be included in the next release.

Note that this should not stop your program, so you can ignore the benign error. If you want to get rid of it, build from the latest code :)

[wentinghome]

same error here.

Traceback (most recent call last):
  File "/export/home/miniconda3/envs/vllm/lib/python3.9/multiprocessing/resource_tracker.py", line 201, in main
    cache[rtype].remove(name)
KeyError: '/psm_2ec8dce9'

[linpan]

esource_tracker.py", line 201, in main
cache[rtype].remove(name)
KeyError: '/psm_2ec8dce9'

[bestpredicts]

#5512 is not included in 0.5.0.post1 . It will be included in the next release.

Note that this should not stop your program, so you can ignore the benign error. If you want to get rid of it, build from the latest code :)

same here

[tszslovewanpu]

Thanks