deepseek-ai/DeepEP
1
0
Fork 0
mirror of https://github.com/deepseek-ai/DeepEP.git synced 2026-11-05 00:47:32 +00:00
Code Issues 308 Projects Packages Wiki Activity

[GH-ISSUE #121] test_low_latency.py failed on 2 nodes. #107

New issue
Closed
opened 2026-10-26 09:54:25 +00:00 by bsakaguchi · 3 comments
bsakaguchi commented 2026-10-26 09:54:25 +00:00
Owner
Copy link

Originally created by @liusy58 on GitHub (Apr 15, 2025).
Original GitHub issue: https://github.com/deepseek-ai/DeepEP/issues/121

node1

export MASTER_ADDR=
export WORLD_SIZE=2
export MASTER_PORT=9000
export RANK=0

node2

export MASTER_ADDR=
export WORLD_SIZE=2
export MASTER_PORT=9000
export RANK=0

I run test_low_latency but got the following error:

python tests/test_low_latency.py 
Allocating buffer size: 2115.111296 MB ...
/sglang/sglang/3rdparty/nvshmem_src/src/modules/bootstrap/uid/bootstrap_uid.cpp:bootstrap_net_recv:99: Message truncated : received 36 bytes instead of 1

/sglang/sglang/3rdparty/nvshmem_src/src/modules/bootstrap/uid/bootstrap_uid.cpp:499: non-zero status: -3 /sglang/sglang/3rdparty/nvshmem_src/src/modules/bootstrap/uid/bootstrap_uid.cpp:bootstrap_net_recv:99: Message truncated : received 40 bytes instead of 1

/sglang/sglang/3rdparty/nvshmem_src/src/modules/bootstrap/uid/bootstrap_uid.cpp:499: non-zero status: -3 /sglang/sglang/3rdparty/nvshmem_src/src/modules/bootstrap/uid/bootstrap_uid.cpp:bootstrap_net_recv:99: Message truncated : received 33554468 bytes instead of 40

/sglang/sglang/3rdparty/nvshmem_src/src/modules/bootstrap/uid/bootstrap_uid.cpp:499: non-zero status: -3 /sglang/sglang/3rdparty/nvshmem_src/src/host/topo/topo.cpp:526: non-zero status: -3 allgather of ipc handles failed 

/sglang/sglang/3rdparty/nvshmem_src/src/host/team/team.cu:nvshmem_team_split_strided:63: NVSHMEM API called before NVSHMEM initialization has completed

/sglang/sglang/3rdparty/nvshmem_src/src/modules/bootstrap/uid/bootstrap_uid.cpp:bootstrap_net_recv:99: Message truncated : received 40 bytes instead of 1

/sglang/sglang/3rdparty/nvshmem_src/src/modules/bootstrap/uid/bootstrap_uid.cpp:499: non-zero status: -3 /sglang/sglang/3rdparty/nvshmem_src/src/modules/bootstrap/uid/bootstrap_uid.cpp:bootstrap_net_recv:99: Message truncated : received 1743285613 bytes instead of 40

/sglang/sglang/3rdparty/nvshmem_src/src/modules/bootstrap/uid/bootstrap_uid.cpp:499: non-zero status: -3 /sglang/sglang/3rdparty/nvshmem_src/src/host/topo/topo.cpp:526: non-zero status: -3 allgather of ipc handles failed 

/sglang/sglang/3rdparty/nvshmem_src/src/host/team/team.cu:nvshmem_team_split_strided:63: NVSHMEM API called before NVSHMEM initialization has completed

/sglang/sglang/3rdparty/nvshmem_src/src/modules/bootstrap/uid/bootstrap_uid.cpp:bootstrap_net_recv:99: Message truncated : received 40 bytes instead of 1

/sglang/sglang/3rdparty/nvshmem_src/src/modules/bootstrap/uid/bootstrap_uid.cpp:499: non-zero status: -3 /sglang/sglang/3rdparty/nvshmem_src/src/modules/bootstrap/uid/bootstrap_uid.cpp:bootstrap_net_recv:99: Message truncated : received 126 bytes instead of 40

/sglang/sglang/3rdparty/nvshmem_src/src/modules/bootstrap/uid/bootstrap_uid.cpp:499: non-zero status: -3 /sglang/sglang/3rdparty/nvshmem_src/src/host/topo/topo.cpp:526: non-zero status: -3 allgather of ipc handles failed 

/sglang/sglang/3rdparty/nvshmem_src/src/host/team/team.cu:nvshmem_team_split_strided:63: NVSHMEM API called before NVSHMEM initialization has completed

/sglang/sglang/3rdparty/nvshmem_src/src/modules/bootstrap/uid/bootstrap_uid.cpp:bootstrap_net_recv:99: Message truncated : received 162 bytes instead of 40

/sglang/sglang/3rdparty/nvshmem_src/src/modules/bootstrap/uid/bootstrap_uid.cpp:499: non-zero status: -3 /sglang/sglang/3rdparty/nvshmem_src/src/host/topo/topo.cpp:526: non-zero status: -3 allgather of ipc handles failed 

/sglang/sglang/3rdparty/nvshmem_src/src/host/team/team.cu:nvshmem_team_split_strided:63: NVSHMEM API called before NVSHMEM initialization has completed

/sglang/sglang/3rdparty/nvshmem_src/src/modules/bootstrap/uid/bootstrap_uid.cpp:bootstrap_net_recv:99: Message truncated : received 40 bytes instead of 1

/sglang/sglang/3rdparty/nvshmem_src/src/modules/bootstrap/uid/bootstrap_uid.cpp:499: non-zero status: -3 /sglang/sglang/3rdparty/nvshmem_src/src/modules/bootstrap/uid/bootstrap_uid.cpp:bootstrap_net_recv:99: Message truncated : received 198 bytes instead of 40
/sglang/sglang/3rdparty/nvshmem_src/src/modules/bootstrap/uid/bootstrap_uid.cpp:bootstrap_net_recv:99: 
Message truncated : received 40 bytes instead of 1
/sglang/sglang/3rdparty/nvshmem_src/src/modules/bootstrap/uid/bootstrap_uid.cpp:499: non-zero status: -3 
/sglang/sglang/3rdparty/nvshmem_src/src/host/topo/topo.cpp:526: non-zero status: -3 /sglang/sglang/3rdparty/nvshmem_src/src/modules/bootstrap/uid/bootstrap_uid.cpp:499: non-zero status: -3 allgather of ipc handles failed 

/sglang/sglang/3rdparty/nvshmem_src/src/host/team/team.cu:nvshmem_team_split_strided:63: NVSHMEM API called before NVSHMEM initialization has completed

/sglang/sglang/3rdparty/nvshmem_src/src/modules/bootstrap/uid/bootstrap_uid.cpp:bootstrap_net_recv:99: Message truncated : received 40 bytes instead of 1

/sglang/sglang/3rdparty/nvshmem_src/src/modules/bootstrap/uid/bootstrap_uid.cpp:499: non-zero status: -3 /sglang/sglang/3rdparty/nvshmem_src/src/modules/bootstrap/uid/bootstrap_uid.cpp:bootstrap_net_recv:99: Message truncated : received 1054699037 bytes instead of 40

/sglang/sglang/3rdparty/nvshmem_src/src/modules/bootstrap/uid/bootstrap_uid.cpp:499: non-zero status: -3 /sglang/sglang/3rdparty/nvshmem_src/src/host/topo/topo.cpp:526: non-zero status: -3 allgather of ipc handles failed 

/sglang/sglang/3rdparty/nvshmem_src/src/host/team/team.cu:nvshmem_team_split_strided:63: NVSHMEM API called before NVSHMEM initialization has completed

W0415 10:42:30.133000 144043 torch/multiprocessing/spawn.py:160] Terminating process 144134 via signal SIGTERM
W0415 10:42:30.133000 144043 torch/multiprocessing/spawn.py:160] Terminating process 144135 via signal SIGTERM
W0415 10:42:30.134000 144043 torch/multiprocessing/spawn.py:160] Terminating process 144136 via signal SIGTERM
W0415 10:42:30.134000 144043 torch/multiprocessing/spawn.py:160] Terminating process 144137 via signal SIGTERM
W0415 10:42:30.134000 144043 torch/multiprocessing/spawn.py:160] Terminating process 144138 via signal SIGTERM
W0415 10:42:30.135000 144043 torch/multiprocessing/spawn.py:160] Terminating process 144139 via signal SIGTERM
W0415 10:42:30.135000 144043 torch/multiprocessing/spawn.py:160] Terminating process 144140 via signal SIGTERM
Traceback (most recent call last):
  File "/sglang/sglang/3rdparty/DeepEP/tests/test_low_latency.py", line 172, in <module>
    torch.multiprocessing.spawn(test_loop, args=(num_processes,), nprocs=num_processes)
  File "/sglang/sglang/py312/lib/python3.12/site-packages/torch/multiprocessing/spawn.py", line 328, in spawn
    return start_processes(fn, args, nprocs, join, daemon, start_method="spawn")
           ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "/sglang/sglang/py312/lib/python3.12/site-packages/torch/multiprocessing/spawn.py", line 284, in start_processes
    while not context.join():
              ^^^^^^^^^^^^^^
  File "/sglang/sglang/py312/lib/python3.12/site-packages/torch/multiprocessing/spawn.py", line 192, in join
    raise ProcessExitedException(
torch.multiprocessing.spawn.ProcessExitedException: process 0 terminated with exit code 255
Originally created by @liusy58 on GitHub (Apr 15, 2025). Original GitHub issue: https://github.com/deepseek-ai/DeepEP/issues/121 ## node1 ``` export MASTER_ADDR= export WORLD_SIZE=2 export MASTER_PORT=9000 export RANK=0 ``` ## node2 ``` export MASTER_ADDR= export WORLD_SIZE=2 export MASTER_PORT=9000 export RANK=0 ``` I run `test_low_latency` but got the following error: ``` python tests/test_low_latency.py Allocating buffer size: 2115.111296 MB ... /sglang/sglang/3rdparty/nvshmem_src/src/modules/bootstrap/uid/bootstrap_uid.cpp:bootstrap_net_recv:99: Message truncated : received 36 bytes instead of 1 /sglang/sglang/3rdparty/nvshmem_src/src/modules/bootstrap/uid/bootstrap_uid.cpp:499: non-zero status: -3 /sglang/sglang/3rdparty/nvshmem_src/src/modules/bootstrap/uid/bootstrap_uid.cpp:bootstrap_net_recv:99: Message truncated : received 40 bytes instead of 1 /sglang/sglang/3rdparty/nvshmem_src/src/modules/bootstrap/uid/bootstrap_uid.cpp:499: non-zero status: -3 /sglang/sglang/3rdparty/nvshmem_src/src/modules/bootstrap/uid/bootstrap_uid.cpp:bootstrap_net_recv:99: Message truncated : received 33554468 bytes instead of 40 /sglang/sglang/3rdparty/nvshmem_src/src/modules/bootstrap/uid/bootstrap_uid.cpp:499: non-zero status: -3 /sglang/sglang/3rdparty/nvshmem_src/src/host/topo/topo.cpp:526: non-zero status: -3 allgather of ipc handles failed /sglang/sglang/3rdparty/nvshmem_src/src/host/team/team.cu:nvshmem_team_split_strided:63: NVSHMEM API called before NVSHMEM initialization has completed /sglang/sglang/3rdparty/nvshmem_src/src/modules/bootstrap/uid/bootstrap_uid.cpp:bootstrap_net_recv:99: Message truncated : received 40 bytes instead of 1 /sglang/sglang/3rdparty/nvshmem_src/src/modules/bootstrap/uid/bootstrap_uid.cpp:499: non-zero status: -3 /sglang/sglang/3rdparty/nvshmem_src/src/modules/bootstrap/uid/bootstrap_uid.cpp:bootstrap_net_recv:99: Message truncated : received 1743285613 bytes instead of 40 /sglang/sglang/3rdparty/nvshmem_src/src/modules/bootstrap/uid/bootstrap_uid.cpp:499: non-zero status: -3 /sglang/sglang/3rdparty/nvshmem_src/src/host/topo/topo.cpp:526: non-zero status: -3 allgather of ipc handles failed /sglang/sglang/3rdparty/nvshmem_src/src/host/team/team.cu:nvshmem_team_split_strided:63: NVSHMEM API called before NVSHMEM initialization has completed /sglang/sglang/3rdparty/nvshmem_src/src/modules/bootstrap/uid/bootstrap_uid.cpp:bootstrap_net_recv:99: Message truncated : received 40 bytes instead of 1 /sglang/sglang/3rdparty/nvshmem_src/src/modules/bootstrap/uid/bootstrap_uid.cpp:499: non-zero status: -3 /sglang/sglang/3rdparty/nvshmem_src/src/modules/bootstrap/uid/bootstrap_uid.cpp:bootstrap_net_recv:99: Message truncated : received 126 bytes instead of 40 /sglang/sglang/3rdparty/nvshmem_src/src/modules/bootstrap/uid/bootstrap_uid.cpp:499: non-zero status: -3 /sglang/sglang/3rdparty/nvshmem_src/src/host/topo/topo.cpp:526: non-zero status: -3 allgather of ipc handles failed /sglang/sglang/3rdparty/nvshmem_src/src/host/team/team.cu:nvshmem_team_split_strided:63: NVSHMEM API called before NVSHMEM initialization has completed /sglang/sglang/3rdparty/nvshmem_src/src/modules/bootstrap/uid/bootstrap_uid.cpp:bootstrap_net_recv:99: Message truncated : received 162 bytes instead of 40 /sglang/sglang/3rdparty/nvshmem_src/src/modules/bootstrap/uid/bootstrap_uid.cpp:499: non-zero status: -3 /sglang/sglang/3rdparty/nvshmem_src/src/host/topo/topo.cpp:526: non-zero status: -3 allgather of ipc handles failed /sglang/sglang/3rdparty/nvshmem_src/src/host/team/team.cu:nvshmem_team_split_strided:63: NVSHMEM API called before NVSHMEM initialization has completed /sglang/sglang/3rdparty/nvshmem_src/src/modules/bootstrap/uid/bootstrap_uid.cpp:bootstrap_net_recv:99: Message truncated : received 40 bytes instead of 1 /sglang/sglang/3rdparty/nvshmem_src/src/modules/bootstrap/uid/bootstrap_uid.cpp:499: non-zero status: -3 /sglang/sglang/3rdparty/nvshmem_src/src/modules/bootstrap/uid/bootstrap_uid.cpp:bootstrap_net_recv:99: Message truncated : received 198 bytes instead of 40 /sglang/sglang/3rdparty/nvshmem_src/src/modules/bootstrap/uid/bootstrap_uid.cpp:bootstrap_net_recv:99: Message truncated : received 40 bytes instead of 1 /sglang/sglang/3rdparty/nvshmem_src/src/modules/bootstrap/uid/bootstrap_uid.cpp:499: non-zero status: -3 /sglang/sglang/3rdparty/nvshmem_src/src/host/topo/topo.cpp:526: non-zero status: -3 /sglang/sglang/3rdparty/nvshmem_src/src/modules/bootstrap/uid/bootstrap_uid.cpp:499: non-zero status: -3 allgather of ipc handles failed /sglang/sglang/3rdparty/nvshmem_src/src/host/team/team.cu:nvshmem_team_split_strided:63: NVSHMEM API called before NVSHMEM initialization has completed /sglang/sglang/3rdparty/nvshmem_src/src/modules/bootstrap/uid/bootstrap_uid.cpp:bootstrap_net_recv:99: Message truncated : received 40 bytes instead of 1 /sglang/sglang/3rdparty/nvshmem_src/src/modules/bootstrap/uid/bootstrap_uid.cpp:499: non-zero status: -3 /sglang/sglang/3rdparty/nvshmem_src/src/modules/bootstrap/uid/bootstrap_uid.cpp:bootstrap_net_recv:99: Message truncated : received 1054699037 bytes instead of 40 /sglang/sglang/3rdparty/nvshmem_src/src/modules/bootstrap/uid/bootstrap_uid.cpp:499: non-zero status: -3 /sglang/sglang/3rdparty/nvshmem_src/src/host/topo/topo.cpp:526: non-zero status: -3 allgather of ipc handles failed /sglang/sglang/3rdparty/nvshmem_src/src/host/team/team.cu:nvshmem_team_split_strided:63: NVSHMEM API called before NVSHMEM initialization has completed W0415 10:42:30.133000 144043 torch/multiprocessing/spawn.py:160] Terminating process 144134 via signal SIGTERM W0415 10:42:30.133000 144043 torch/multiprocessing/spawn.py:160] Terminating process 144135 via signal SIGTERM W0415 10:42:30.134000 144043 torch/multiprocessing/spawn.py:160] Terminating process 144136 via signal SIGTERM W0415 10:42:30.134000 144043 torch/multiprocessing/spawn.py:160] Terminating process 144137 via signal SIGTERM W0415 10:42:30.134000 144043 torch/multiprocessing/spawn.py:160] Terminating process 144138 via signal SIGTERM W0415 10:42:30.135000 144043 torch/multiprocessing/spawn.py:160] Terminating process 144139 via signal SIGTERM W0415 10:42:30.135000 144043 torch/multiprocessing/spawn.py:160] Terminating process 144140 via signal SIGTERM Traceback (most recent call last): File "/sglang/sglang/3rdparty/DeepEP/tests/test_low_latency.py", line 172, in <module> torch.multiprocessing.spawn(test_loop, args=(num_processes,), nprocs=num_processes) File "/sglang/sglang/py312/lib/python3.12/site-packages/torch/multiprocessing/spawn.py", line 328, in spawn return start_processes(fn, args, nprocs, join, daemon, start_method="spawn") ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ File "/sglang/sglang/py312/lib/python3.12/site-packages/torch/multiprocessing/spawn.py", line 284, in start_processes while not context.join(): ^^^^^^^^^^^^^^ File "/sglang/sglang/py312/lib/python3.12/site-packages/torch/multiprocessing/spawn.py", line 192, in join raise ProcessExitedException( torch.multiprocessing.spawn.ProcessExitedException: process 0 terminated with exit code 255 ```
bsakaguchi commented 2026-10-26 09:54:26 +00:00
Author
Owner
Copy link

@sphish commented on GitHub (Apr 16, 2025):

/sglang/sglang/3rdparty/nvshmem_src/src/host/team/team.cu:nvshmem_team_split_strided:63: NVSHMEM API called before NVSHMEM initialization has completed

This log line is very strange. Have you modified the test code?

<!-- gh-comment-id:2808022730 --> @sphish commented on GitHub (Apr 16, 2025): > /sglang/sglang/3rdparty/nvshmem_src/src/host/team/team.cu:nvshmem_team_split_strided:63: NVSHMEM API called before NVSHMEM initialization has completed This log line is very strange. Have you modified the test code?
bsakaguchi commented 2026-10-26 09:54:26 +00:00
Author
Owner
Copy link

@liusy58 commented on GitHub (Apr 17, 2025):

Not yet. I found the the main cause of my problem is the version nvidia gpu driver of two nodes are different.

<!-- gh-comment-id:2811531280 --> @liusy58 commented on GitHub (Apr 17, 2025): Not yet. I found the the main cause of my problem is the version nvidia gpu driver of two nodes are different.
bsakaguchi commented 2026-10-26 09:54:26 +00:00
Author
Owner
Copy link

@liusy58 commented on GitHub (Apr 17, 2025):

Thank you a lot~

<!-- gh-comment-id:2811531441 --> @liusy58 commented on GitHub (Apr 17, 2025): Thank you a lot~
Sign in to join this conversation.
No Branch/Tag specified
Branches Tags
main
hybrid-ep
chore/remove-clang-format-yapf
v2.5
nvDev
try-fix-cu128
mori-ep
antgroup-opt
compat-old-torch
add-linter
eager-rdma
v1.2.1
Labels
Clear labels   
pull-request
Mirrored from GitHub Pull Request

No labels
pull-request
Milestone
Clear milestone
No items
No milestone
Projects
Clear projects
No items
No project
Assignees
Clear assignees
No assignees
1 participant
Notifications
Due date
The due date is invalid or out of range. Please use the format "yyyy-mm-dd".

No due date set.

Dependencies

No dependencies set.

Reference
deepseek-ai/DeepEP#107
No description provided.