hello; how to fix this error; errno: 97 - Address family not supported by protocol
Nobody has claimed this yet.
- Dominant language
- Python
- Stars
- 682
- Forks
- 128
- Avg merge
- 9h 44m
- Merged PRs (30d)
- 1
Description
python /app/software1/vidur/vidur/profiling/collectives/main.py
--num_workers_per_node_combinations 2
--collective send_recv
2025-02-14 07:59:21,469 INFO worker.py:1821 -- Started a local Ray instance.
0%| | 0/994 [00:00<?, ?it/s](BenchmarkRunner pid=10922) INFO 02-14 07:59:30 benchmark_runner.py:83] Initializing gpu id: 1, Rank: 1, num_workers: 2, comm_id: 50704, devices_per_node: 2, max_devices_per_node: 8, ip_addr: 33.254.158.199, CUDA_VISIBLE_DEVICES: 1
(BenchmarkRunner pid=10922) [W socket.cpp:697] [c10d] The client socket cannot be initialized to connect to [x08j01357.cloud.sqa.na131.tbsite.net]:50704 (errno: 97 - Address family not supported by protocol).
(BenchmarkRunner pid=10896) [W socket.cpp:464] [c10d] The server socket cannot be initialized on [::]:50704 (errno: 97 - Address family not supported by protocol).
100%|███████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████| 994/994 [02:25<00:00, 6.84it/s]
(BenchmarkRunner pid=10896) INFO 02-14 07:59:30 benchmark_runner.py:83] Initializing gpu id: 0, Rank: 0, num_workers: 2, comm_id: 50704, devices_per_node: 2, max_devices_per_node: 8, ip_addr: 33.254.158.199, CUDA_VISIBLE_DEVICES: 0
(BenchmarkRunner pid=10896) [W socket.cpp:697] [c10d] The client socket cannot be initialized to connect to [x08j01357.cloud.sqa.na131.tbsite.net]:50704 (errno: 97 - Address family not supported by protocol).
Contributor guide
No contributing guide indexed for this repository
First steps
- Read the whole issue, then the project's contributing guide.
- Comment on the issue to say you are picking it up — it saves two people doing the same work.
- Fork the repository and make your change on a branch.
- Open a pull request that references the issue number.
Research direction
Start with vidur/profiling/collectives/main.py and reproduce the command shown for the send_recv collective with two workers per node. Trace how the worker address and communication port are selected, then identify a configuration that completes without the socket warnings and verify the profiling run succeeds.
Written by the indexing model from the issue text.
Assessment
- Tech stack
- python
- Domain
- distributed-systems, networking
- Issue type
- Bug
- Difficulty
- 4/5
- Estimated time
- 3-5 days
- Activity status
- Stale
- Clarity
- Needs clarification
- Newbie friendliness
- 35/100