Issue Station
Sign In
Tools hereToolbox

vllm - 💡(How to fix) Fix [Bug]: V1 engine hangs then crashes with "No available shared memory broadcast block found / RPC call to sample_tokens timed out" under chat completions burst load on Qwen3.5-122B-A10B [12 comments, 5 participants]

Official PRs (…)
ON THIS PAGE

Recommended Tools

×6

Utilities matched from this issue’s tags and category — try them while you read without losing context.

GitHub issue graph ai analysis

Paste a GitHub issue URL. We fetch that issue, discover linked issues from bodies/comments/timeline, collect linked pull requests, and produce a structured English report.

The report is written in English Markdown for sharing and archival.

Helpful · Quick feedback

Loading…
GitHub stats
vllm-project/vllm#36921•Fetched 2026-04-08 00:43:37
View on GitHub
Comments
12
Participants
5
Timeline
14
Reactions
1
Author
natech-personanatech-persona
Participants
lianjiezhlianjiezhmiguelmanlyxmiguelmanlyxnatech-personanatech-personawuzixiawuzixiaZJY0516ZJY0516
Timeline (top)
commented ×12closed ×1labeled ×1

Error Message

I'm seeing the error under versions:

Fix Action

Fix / Workaround

============================== CPU Info

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): 192 On-line CPU(s) list: 0-191 Vendor ID: GenuineIntel Model name: INTEL(R) XEON(R) PLATINUM 8568Y+ CPU family: 6 Model: 207 Thread(s) per core: 2 Core(s) per socket: 48 Socket(s): 2 Stepping: 2 BogoMIPS: 4600.00 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 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 cdp_l2 ssbd mba ibrs ibpb stibp ibrs_enhanced 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 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 amx_bf16 avx512_fp16 amx_tile amx_int8 flush_l1d arch_capabilities ibpb_exit_to_user L1d cache: 4.5 MiB (96 instances) L1i cache: 3 MiB (96 instances) L2 cache: 192 MiB (96 instances) L3 cache: 600 MiB (2 instances) NUMA node(s): 4 NUMA node0 CPU(s): 0-23,96-119 NUMA node1 CPU(s): 24-47,120-143 NUMA node2 CPU(s): 48-71,144-167 NUMA node3 CPU(s): 72-95,168-191 Vulnerability Gather data sampling: Not affected Vulnerability Indirect target selection: 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 Reg file data sampling: Not affected Vulnerability Retbleed: Not affected Vulnerability Spec rstack overflow: Not affected Vulnerability Spec store bypass: Mitigation; Speculative Store Bypass disabled via prctl and seccomp Vulnerability Spectre v1: Mitigation; usercopy/swapgs barriers and __user pointer sanitization Vulnerability Spectre v2: Mitigation; Enhanced / Automatic IBRS; IBPB conditional; PBRSB-eIBRS SW sequence; BHI BHI_DIS_S Vulnerability Srbds: Not affected Vulnerability Tsa: Not affected Vulnerability Tsx async abort: Not affected Vulnerability Vmscape: Mitigation; IBPB before exit to userspace

Code Example

uv is set
==============================
        System Info
==============================
OS                           : Ubuntu 22.04.5 LTS (x86_64)
GCC version                  : (Ubuntu 11.4.0-1ubuntu1~22.04.3) 11.4.0
Clang version                : Could not collect
CMake version                : Could not collect
Libc version                 : glibc-2.35

==============================
       PyTorch Info
==============================
PyTorch version              : 2.10.0+cu128
Is debug build               : False
CUDA used to build PyTorch   : 12.8
ROCM used to build PyTorch   : N/A

==============================
      Python Environment
==============================
Python version               : 3.10.12 (main, Mar  3 2026, 11:56:32) [GCC 11.4.0] (64-bit runtime)
Python platform              : Linux-5.15.0-164-generic-x86_64-with-glibc2.35

==============================
       CUDA / GPU Info
==============================
Is CUDA available            : True
CUDA runtime version         : Could not collect
CUDA_MODULE_LOADING set to   : 
GPU models and configuration : 
GPU 0: NVIDIA H200
GPU 1: NVIDIA H200
GPU 2: NVIDIA H200
GPU 3: NVIDIA H200
GPU 4: NVIDIA H200
GPU 5: NVIDIA H200
GPU 6: NVIDIA H200
GPU 7: NVIDIA H200

Nvidia driver version        : 570.211.01
cuDNN version                : Could not collect
HIP runtime version          : N/A
MIOpen runtime version       : N/A
Is XNNPACK available         : True

==============================
          CPU Info
==============================
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):                                  192
On-line CPU(s) list:                     0-191
Vendor ID:                               GenuineIntel
Model name:                              INTEL(R) XEON(R) PLATINUM 8568Y+
CPU family:                              6
Model:                                   207
Thread(s) per core:                      2
Core(s) per socket:                      48
Socket(s):                               2
Stepping:                                2
BogoMIPS:                                4600.00
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 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 cdp_l2 ssbd mba ibrs ibpb stibp ibrs_enhanced 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 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 amx_bf16 avx512_fp16 amx_tile amx_int8 flush_l1d arch_capabilities ibpb_exit_to_user
L1d cache:                               4.5 MiB (96 instances)
L1i cache:                               3 MiB (96 instances)
L2 cache:                                192 MiB (96 instances)
L3 cache:                                600 MiB (2 instances)
NUMA node(s):                            4
NUMA node0 CPU(s):                       0-23,96-119
NUMA node1 CPU(s):                       24-47,120-143
NUMA node2 CPU(s):                       48-71,144-167
NUMA node3 CPU(s):                       72-95,168-191
Vulnerability Gather data sampling:      Not affected
Vulnerability Indirect target selection: 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 Reg file data sampling:    Not affected
Vulnerability Retbleed:                  Not affected
Vulnerability Spec rstack overflow:      Not affected
Vulnerability Spec store bypass:         Mitigation; Speculative Store Bypass disabled via prctl and seccomp
Vulnerability Spectre v1:                Mitigation; usercopy/swapgs barriers and __user pointer sanitization
Vulnerability Spectre v2:                Mitigation; Enhanced / Automatic IBRS; IBPB conditional; PBRSB-eIBRS SW sequence; BHI BHI_DIS_S
Vulnerability Srbds:                     Not affected
Vulnerability Tsa:                       Not affected
Vulnerability Tsx async abort:           Not affected
Vulnerability Vmscape:                   Mitigation; IBPB before exit to userspace

==============================
Versions of relevant libraries
==============================
[pip3] flashinfer-python==0.6.4
[pip3] numpy==2.2.6
[pip3] nvidia-cublas-cu12==12.8.4.1
[pip3] nvidia-cuda-cupti-cu12==12.8.90
[pip3] nvidia-cuda-nvrtc-cu12==12.8.93
[pip3] nvidia-cuda-runtime-cu12==12.8.90
[pip3] nvidia-cudnn-cu12==9.10.2.21
[pip3] nvidia-cudnn-frontend==1.18.0
[pip3] nvidia-cufft-cu12==11.3.3.83
[pip3] nvidia-cufile-cu12==1.13.1.3
[pip3] nvidia-curand-cu12==10.3.9.90
[pip3] nvidia-cusolver-cu12==11.7.3.90
[pip3] nvidia-cusparse-cu12==12.5.8.93
[pip3] nvidia-cusparselt-cu12==0.7.1
[pip3] nvidia-cutlass-dsl==4.4.1
[pip3] nvidia-cutlass-dsl-libs-base==4.4.1
[pip3] nvidia-ml-py==13.590.48
[pip3] nvidia-nccl-cu12==2.27.5
[pip3] nvidia-nvjitlink-cu12==12.8.93
[pip3] nvidia-nvshmem-cu12==3.4.5
[pip3] nvidia-nvtx-cu12==12.8.90
[pip3] pyzmq==27.1.0
[pip3] torch==2.10.0
[pip3] torch-c-dlpack-ext==0.1.5
[pip3] torchaudio==2.10.0
[pip3] torchvision==0.25.0
[pip3] transformers==4.57.6
[pip3] triton==3.6.0
[conda] Could not collect

==============================
         vLLM Info
==============================
ROCM Version                 : Could not collect
vLLM Version                 : 0.17.1
vLLM Build Flags:
  CUDA Archs: Not Set; ROCm: Disabled
GPU Topology:
        GPU0    GPU1    GPU2    GPU3    GPU4    GPU5    GPU6    GPU7    NIC0    NIC1    NIC2    NIC3    NIC4    NIC5    NIC6    NIC7    NIC8    CPU Affinity    NUMA Affinity   GPU NUMA ID
GPU0     X      NV18    NV18    NV18    NV18    NV18    NV18    NV18    PIX     NODE    SYS     SYS     SYS     SYS     SYS     SYS     SYS     0,2-23,96,98-119        0               N/A
GPU1    NV18     X      NV18    NV18    NV18    NV18    NV18    NV18    NODE    PIX     SYS     SYS     SYS     SYS     SYS     SYS     SYS     0,2-23,96,98-119        0               N/A
GPU2    NV18    NV18     X      NV18    NV18    NV18    NV18    NV18    SYS     SYS     PIX     NODE    SYS     SYS     SYS     SYS     SYS     24-47,120-143   1               N/A
GPU3    NV18    NV18    NV18     X      NV18    NV18    NV18    NV18    SYS     SYS     NODE    PIX     SYS     SYS     SYS     SYS     SYS     24-47,120-143   1               N/A
GPU4    NV18    NV18    NV18    NV18     X      NV18    NV18    NV18    SYS     SYS     SYS     SYS     PIX     NODE    SYS     SYS     PIX     48-71,144-167   2               N/A
GPU5    NV18    NV18    NV18    NV18    NV18     X      NV18    NV18    SYS     SYS     SYS     SYS     NODE    PIX     SYS     SYS     NODE    48-71,144-167   2               N/A
GPU6    NV18    NV18    NV18    NV18    NV18    NV18     X      NV18    SYS     SYS     SYS     SYS     SYS     SYS     PIX     NODE    SYS     72-95,168-191   3               N/A
GPU7    NV18    NV18    NV18    NV18    NV18    NV18    NV18     X      SYS     SYS     SYS     SYS     SYS     SYS     NODE    PIX     SYS     72-95,168-191   3               N/A
NIC0    PIX     NODE    SYS     SYS     SYS     SYS     SYS     SYS      X      NODE    SYS     SYS     SYS     SYS     SYS     SYS     SYS
NIC1    NODE    PIX     SYS     SYS     SYS     SYS     SYS     SYS     NODE     X      SYS     SYS     SYS     SYS     SYS     SYS     SYS
NIC2    SYS     SYS     PIX     NODE    SYS     SYS     SYS     SYS     SYS     SYS      X      NODE    SYS     SYS     SYS     SYS     SYS
NIC3    SYS     SYS     NODE    PIX     SYS     SYS     SYS     SYS     SYS     SYS     NODE     X      SYS     SYS     SYS     SYS     SYS
NIC4    SYS     SYS     SYS     SYS     PIX     NODE    SYS     SYS     SYS     SYS     SYS     SYS      X      NODE    SYS     SYS     PIX
NIC5    SYS     SYS     SYS     SYS     NODE    PIX     SYS     SYS     SYS     SYS     SYS     SYS     NODE     X      SYS     SYS     NODE
NIC6    SYS     SYS     SYS     SYS     SYS     SYS     PIX     NODE    SYS     SYS     SYS     SYS     SYS     SYS      X      NODE    SYS
NIC7    SYS     SYS     SYS     SYS     SYS     SYS     NODE    PIX     SYS     SYS     SYS     SYS     SYS     SYS     NODE     X      SYS
NIC8    SYS     SYS     SYS     SYS     PIX     NODE    SYS     SYS     SYS     SYS     SYS     SYS     PIX     NODE    SYS     SYS      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_7
  NIC6: mlx5_8
  NIC7: mlx5_9
  NIC8: mlx5_bond_0

==============================
     Environment Variables
==============================
PYTORCH_NVML_BASED_CUDA_CHECK=1
TORCHINDUCTOR_COMPILE_THREADS=1
TORCHINDUCTOR_CACHE_DIR=/tmp/torchinductor_persona

---

vllm serve Qwen/Qwen3.5-122B-A10B \
 --tensor-parallel-size 8 \
 --reasoning-parser qwen3 \
 --served-model-name q3.5-122b \
 --trust-remote-code \
 --cudagraph-metrics \
 --enable-prefix-caching \
 --gpu-memory-utilization 0.95 \
 --max-model-len 16384 \
 --language-model-only \

---

No available shared memory broadcast block found in 60 seconds.
This typically happens when some processes are hanging or doing some time-consuming work
(e.g. compilation, weight/kv cache quantization).
RAW_BUFFERClick to expand / collapse

Your current environment

<details> <summary>The output of <code>python collect_env.py</code></summary> 
uv is set
==============================
        System Info
==============================
OS                           : Ubuntu 22.04.5 LTS (x86_64)
GCC version                  : (Ubuntu 11.4.0-1ubuntu1~22.04.3) 11.4.0
Clang version                : Could not collect
CMake version                : Could not collect
Libc version                 : glibc-2.35

==============================
       PyTorch Info
==============================
PyTorch version              : 2.10.0+cu128
Is debug build               : False
CUDA used to build PyTorch   : 12.8
ROCM used to build PyTorch   : N/A

==============================
      Python Environment
==============================
Python version               : 3.10.12 (main, Mar  3 2026, 11:56:32) [GCC 11.4.0] (64-bit runtime)
Python platform              : Linux-5.15.0-164-generic-x86_64-with-glibc2.35

==============================
       CUDA / GPU Info
==============================
Is CUDA available            : True
CUDA runtime version         : Could not collect
CUDA_MODULE_LOADING set to   : 
GPU models and configuration : 
GPU 0: NVIDIA H200
GPU 1: NVIDIA H200
GPU 2: NVIDIA H200
GPU 3: NVIDIA H200
GPU 4: NVIDIA H200
GPU 5: NVIDIA H200
GPU 6: NVIDIA H200
GPU 7: NVIDIA H200

Nvidia driver version        : 570.211.01
cuDNN version                : Could not collect
HIP runtime version          : N/A
MIOpen runtime version       : N/A
Is XNNPACK available         : True

==============================
          CPU Info
==============================
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):                                  192
On-line CPU(s) list:                     0-191
Vendor ID:                               GenuineIntel
Model name:                              INTEL(R) XEON(R) PLATINUM 8568Y+
CPU family:                              6
Model:                                   207
Thread(s) per core:                      2
Core(s) per socket:                      48
Socket(s):                               2
Stepping:                                2
BogoMIPS:                                4600.00
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 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 cdp_l2 ssbd mba ibrs ibpb stibp ibrs_enhanced 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 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 amx_bf16 avx512_fp16 amx_tile amx_int8 flush_l1d arch_capabilities ibpb_exit_to_user
L1d cache:                               4.5 MiB (96 instances)
L1i cache:                               3 MiB (96 instances)
L2 cache:                                192 MiB (96 instances)
L3 cache:                                600 MiB (2 instances)
NUMA node(s):                            4
NUMA node0 CPU(s):                       0-23,96-119
NUMA node1 CPU(s):                       24-47,120-143
NUMA node2 CPU(s):                       48-71,144-167
NUMA node3 CPU(s):                       72-95,168-191
Vulnerability Gather data sampling:      Not affected
Vulnerability Indirect target selection: 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 Reg file data sampling:    Not affected
Vulnerability Retbleed:                  Not affected
Vulnerability Spec rstack overflow:      Not affected
Vulnerability Spec store bypass:         Mitigation; Speculative Store Bypass disabled via prctl and seccomp
Vulnerability Spectre v1:                Mitigation; usercopy/swapgs barriers and __user pointer sanitization
Vulnerability Spectre v2:                Mitigation; Enhanced / Automatic IBRS; IBPB conditional; PBRSB-eIBRS SW sequence; BHI BHI_DIS_S
Vulnerability Srbds:                     Not affected
Vulnerability Tsa:                       Not affected
Vulnerability Tsx async abort:           Not affected
Vulnerability Vmscape:                   Mitigation; IBPB before exit to userspace

==============================
Versions of relevant libraries
==============================
[pip3] flashinfer-python==0.6.4
[pip3] numpy==2.2.6
[pip3] nvidia-cublas-cu12==12.8.4.1
[pip3] nvidia-cuda-cupti-cu12==12.8.90
[pip3] nvidia-cuda-nvrtc-cu12==12.8.93
[pip3] nvidia-cuda-runtime-cu12==12.8.90
[pip3] nvidia-cudnn-cu12==9.10.2.21
[pip3] nvidia-cudnn-frontend==1.18.0
[pip3] nvidia-cufft-cu12==11.3.3.83
[pip3] nvidia-cufile-cu12==1.13.1.3
[pip3] nvidia-curand-cu12==10.3.9.90
[pip3] nvidia-cusolver-cu12==11.7.3.90
[pip3] nvidia-cusparse-cu12==12.5.8.93
[pip3] nvidia-cusparselt-cu12==0.7.1
[pip3] nvidia-cutlass-dsl==4.4.1
[pip3] nvidia-cutlass-dsl-libs-base==4.4.1
[pip3] nvidia-ml-py==13.590.48
[pip3] nvidia-nccl-cu12==2.27.5
[pip3] nvidia-nvjitlink-cu12==12.8.93
[pip3] nvidia-nvshmem-cu12==3.4.5
[pip3] nvidia-nvtx-cu12==12.8.90
[pip3] pyzmq==27.1.0
[pip3] torch==2.10.0
[pip3] torch-c-dlpack-ext==0.1.5
[pip3] torchaudio==2.10.0
[pip3] torchvision==0.25.0
[pip3] transformers==4.57.6
[pip3] triton==3.6.0
[conda] Could not collect

==============================
         vLLM Info
==============================
ROCM Version                 : Could not collect
vLLM Version                 : 0.17.1
vLLM Build Flags:
  CUDA Archs: Not Set; ROCm: Disabled
GPU Topology:
        GPU0    GPU1    GPU2    GPU3    GPU4    GPU5    GPU6    GPU7    NIC0    NIC1    NIC2    NIC3    NIC4    NIC5    NIC6    NIC7    NIC8    CPU Affinity    NUMA Affinity   GPU NUMA ID
GPU0     X      NV18    NV18    NV18    NV18    NV18    NV18    NV18    PIX     NODE    SYS     SYS     SYS     SYS     SYS     SYS     SYS     0,2-23,96,98-119        0               N/A
GPU1    NV18     X      NV18    NV18    NV18    NV18    NV18    NV18    NODE    PIX     SYS     SYS     SYS     SYS     SYS     SYS     SYS     0,2-23,96,98-119        0               N/A
GPU2    NV18    NV18     X      NV18    NV18    NV18    NV18    NV18    SYS     SYS     PIX     NODE    SYS     SYS     SYS     SYS     SYS     24-47,120-143   1               N/A
GPU3    NV18    NV18    NV18     X      NV18    NV18    NV18    NV18    SYS     SYS     NODE    PIX     SYS     SYS     SYS     SYS     SYS     24-47,120-143   1               N/A
GPU4    NV18    NV18    NV18    NV18     X      NV18    NV18    NV18    SYS     SYS     SYS     SYS     PIX     NODE    SYS     SYS     PIX     48-71,144-167   2               N/A
GPU5    NV18    NV18    NV18    NV18    NV18     X      NV18    NV18    SYS     SYS     SYS     SYS     NODE    PIX     SYS     SYS     NODE    48-71,144-167   2               N/A
GPU6    NV18    NV18    NV18    NV18    NV18    NV18     X      NV18    SYS     SYS     SYS     SYS     SYS     SYS     PIX     NODE    SYS     72-95,168-191   3               N/A
GPU7    NV18    NV18    NV18    NV18    NV18    NV18    NV18     X      SYS     SYS     SYS     SYS     SYS     SYS     NODE    PIX     SYS     72-95,168-191   3               N/A
NIC0    PIX     NODE    SYS     SYS     SYS     SYS     SYS     SYS      X      NODE    SYS     SYS     SYS     SYS     SYS     SYS     SYS
NIC1    NODE    PIX     SYS     SYS     SYS     SYS     SYS     SYS     NODE     X      SYS     SYS     SYS     SYS     SYS     SYS     SYS
NIC2    SYS     SYS     PIX     NODE    SYS     SYS     SYS     SYS     SYS     SYS      X      NODE    SYS     SYS     SYS     SYS     SYS
NIC3    SYS     SYS     NODE    PIX     SYS     SYS     SYS     SYS     SYS     SYS     NODE     X      SYS     SYS     SYS     SYS     SYS
NIC4    SYS     SYS     SYS     SYS     PIX     NODE    SYS     SYS     SYS     SYS     SYS     SYS      X      NODE    SYS     SYS     PIX
NIC5    SYS     SYS     SYS     SYS     NODE    PIX     SYS     SYS     SYS     SYS     SYS     SYS     NODE     X      SYS     SYS     NODE
NIC6    SYS     SYS     SYS     SYS     SYS     SYS     PIX     NODE    SYS     SYS     SYS     SYS     SYS     SYS      X      NODE    SYS
NIC7    SYS     SYS     SYS     SYS     SYS     SYS     NODE    PIX     SYS     SYS     SYS     SYS     SYS     SYS     NODE     X      SYS
NIC8    SYS     SYS     SYS     SYS     PIX     NODE    SYS     SYS     SYS     SYS     SYS     SYS     PIX     NODE    SYS     SYS      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_7
  NIC6: mlx5_8
  NIC7: mlx5_9
  NIC8: mlx5_bond_0

==============================
     Environment Variables
==============================
PYTORCH_NVML_BASED_CUDA_CHECK=1
TORCHINDUCTOR_COMPILE_THREADS=1
TORCHINDUCTOR_CACHE_DIR=/tmp/torchinductor_persona
</details>

🐛 Describe the bug

We are repeatedly hitting what looks like a V1 engine / multiprocess shared-memory coordination bug under bursty but otherwise normal chat-completions load.

The engine will run normally for a while, then throughput suddenly drops to zero, we start seeing:

No available shared memory broadcast block found in 60 seconds.

repeated every minute, and eventually the engine dies with:

  • TimeoutError: RPC call to sample_tokens timed out.

This is happening even though:

  • KV cache usage is still low
  • /dev/shm is very large (1TB)
  • requests are ordinary chat completions
  • this reproduces across multiple versions/config tweaks

My launch args:

vllm serve Qwen/Qwen3.5-122B-A10B \
 --tensor-parallel-size 8 \
 --reasoning-parser qwen3 \
 --served-model-name q3.5-122b \
 --trust-remote-code \
 --cudagraph-metrics \
 --enable-prefix-caching \
 --gpu-memory-utilization 0.95 \
 --max-model-len 16384 \
 --language-model-only \

I'm seeing the error under versions:

  • 0.17.0
  • 0.17.1
  • 0.16.0rc2.dev471+g709eadbb0

Tested already with:

  • --max-num-batched-tokens 8192 and lower, other args e.g.
  • removing the --language-model-only,
  • lowering mem util,
  • using Ray backend

Using persistent caches for XDG, Torch, Triton & CUDA.

This is an offline inference workload, but sent through the OpenAI-compatible chat/completions API. So from the engines perspective it is a normal bursty chat workload.

  • we've observed crashes after 5 mins, 20 mins, 1hr, 2.5 hrs, or 4.5 hrs.
  • the failure often happens when a new stage in our pipeline starts and a fresh burst of requests is added

requests are not especially exotic; ordinary chat completions with prompts in the ~100–1000 token range and max_tokens at 1024. The sender has concurrency limiting and we even added a 10 second exponential rampup, but the issue still reproduces.

Symptoms

A typical sequence is:

  • Engine is healthy and serving requests. . A burst of new requests arrives.
  • Logs show normal throughput, then suddenly: Avg prompt throughput: 0.0 tokens/s, Avg generation throughput: 0.0 tokens/s while there are still many running requests.
  • Then every 60 seconds:
No available shared memory broadcast block found in 60 seconds.
This typically happens when some processes are hanging or doing some time-consuming work
(e.g. compilation, weight/kv cache quantization).

Eventually the engine aborts with:

TimeoutError: RPC call to sample_tokens timed out.

Then all worker processes are torn down.

Before submitting a new issue...

  • Make sure you already searched for relevant issues, and asked the chatbot living at the bottom right corner of the documentation page, which can answer lots of frequently asked questions.

extent analysis

Fix Plan

The issue seems to be related to shared memory broadcast block allocation. To fix this, we can try the following steps:

  • Increase the shared memory size by setting the SHM_SIZE environment variable to a larger value, e.g., export SHM_SIZE=16G.
  • Reduce the --gpu-memory-utilization flag to a lower value, e.g., --gpu-memory-utilization 0.8, to reduce memory usage and alleviate shared memory allocation issues.
  • Implement a retry mechanism for the sample_tokens RPC call to handle temporary timeouts.
  • Consider using a different shared memory allocation library or implementing a custom shared memory allocation solution.

Example code for implementing a retry mechanism:

import time
import torch

def sample_tokens_with_retry(model, prompt, max_retries=3, retry_delay=1):
    for attempt in range(max_retries):
        try:
            return model.sample_tokens(prompt)
        except TimeoutError:
            if attempt < max_retries - 1:
                time.sleep(retry_delay)
            else:
                raise

Verification

To verify that the fix worked, monitor the engine's logs and performance metrics after applying the changes. Check for the presence of the No available shared memory broadcast block found error and the TimeoutError: RPC call to sample_tokens timed out error. If the errors persist, further debugging and optimization may be necessary.

Extra Tips

  • Regularly clean up shared memory to prevent fragmentation and allocation issues.
  • Consider using a memory profiler to identify memory usage patterns and optimize memory allocation.
  • If using a Ray backend, ensure that the Ray configuration is optimized for the specific use case and hardware setup.

Vote matrix · Quick signals

Works
Did the solution work? Tap to confirm.
Easy Fix
Was it a quick fix?
Time Saver
Did it save you time?
Blocking
Was it severely blocking?
Common Issue
Are others likely hitting this too?
Flaky / Intermittent
Is it intermittent?
Verified / Reproducible
Can you reproduce it reliably?
Loading…
#api#ssr#installation#tensor shape#environment variable#container setup#orchestration issue#cache issue

Still need to ship something?

×6

Another batch ranked right after the header list — different links, same matching logic.

Back to top recommendations

TRENDING