Failed to register worker 01000000ffffffffffffffffffffffffffffffffffffffffffffffff to Raylet. IOError: [RayletClient] Unable to register worker with raylet. No such file or directory #100
-
Beta Was this translation helpful? Give feedback.
Replies: 4 comments 7 replies
-
|
Hi, Can you upgrade to the latest version ( |
Beta Was this translation helpful? Give feedback.
-
|
Thanks for the response, This is a very simple demo. Following your advice, I upgraded the version, but the problem still persists. from lean_dojo import *
repo = LeanGitRepo(
# mathlib的地址
"https://github.com/leanprover-community/mathlib4",
# 一个具体的版本
"3ce43c18f614b76e161f911b75a3e1ef641620ff",
)
repo.get_config("lean-toolchain")
# A few minutes if the traced repo is in the cache; many hours otherwise.
traced_repo = trace(repo)error /home/rise/anaconda3/envs/chx-py3.10/bin/python /home/rise/learn/leanDojo/extract.py
2023-11-30 11:18:52.309 | INFO | lean_dojo.data_extraction.trace:trace:182 - Loading the traced repo from /home/rise/.cache/lean_dojo/leanprover-community-mathlib4-3ce43c18f614b76e161f911b75a3e1ef641620ff/mathlib4
2023-11-30 11:18:53,092 INFO worker.py:1489 -- Connecting to existing Ray cluster at address: 198.18.0.1:6379...
[2023-11-30 11:18:53,096 W 108247 108247] global_state_accessor.cc:407: Some processes that the driver needs to connect to have not registered with GCS, so retrying. Have you run 'ray start' on this node?
[2023-11-30 11:18:54,098 W 108247 108247] global_state_accessor.cc:407: Some processes that the driver needs to connect to have not registered with GCS, so retrying. Have you run 'ray start' on this node?
[2023-11-30 11:18:55,099 W 108247 108247] global_state_accessor.cc:407: Some processes that the driver needs to connect to have not registered with GCS, so retrying. Have you run 'ray start' on this node?
[2023-11-30 11:18:56,100 W 108247 108247] global_state_accessor.cc:407: Some processes that the driver needs to connect to have not registered with GCS, so retrying. Have you run 'ray start' on this node?
[2023-11-30 11:18:57,102 W 108247 108247] global_state_accessor.cc:407: Some processes that the driver needs to connect to have not registered with GCS, so retrying. Have you run 'ray start' on this node?
[2023-11-30 11:18:58,103 W 108247 108247] global_state_accessor.cc:407: Some processes that the driver needs to connect to have not registered with GCS, so retrying. Have you run 'ray start' on this node?
[2023-11-30 11:18:59,105 W 108247 108247] global_state_accessor.cc:407: Some processes that the driver needs to connect to have not registered with GCS, so retrying. Have you run 'ray start' on this node?
[2023-11-30 11:19:00,106 W 108247 108247] global_state_accessor.cc:407: Some processes that the driver needs to connect to have not registered with GCS, so retrying. Have you run 'ray start' on this node?
[2023-11-30 11:19:01,107 W 108247 108247] global_state_accessor.cc:407: Some processes that the driver needs to connect to have not registered with GCS, so retrying. Have you run 'ray start' on this node?
[2023-11-30 11:19:02,109 W 108247 108247] global_state_accessor.cc:407: Some processes that the driver needs to connect to have not registered with GCS, so retrying. Have you run 'ray start' on this node?
2023-11-30 11:19:03,111 INFO worker.py:1638 -- Failed to connect to the default Ray cluster address at 198.18.0.1:6379. This is most likely due to a previous Ray instance that has since crashed. To reset the default address to connect to, run `ray stop` or restart Ray with `ray start`.
Traceback (most recent call last):
File "/home/rise/anaconda3/envs/chx-py3.10/lib/python3.10/site-packages/ray/_private/worker.py", line 1629, in init
_global_node = ray._private.node.Node(
File "/home/rise/anaconda3/envs/chx-py3.10/lib/python3.10/site-packages/ray/_private/node.py", line 254, in __init__
node_info = ray._private.services.get_node_to_connect_for_driver(
File "/home/rise/anaconda3/envs/chx-py3.10/lib/python3.10/site-packages/ray/_private/services.py", line 475, in get_node_to_connect_for_driver
return global_state.get_node_to_connect_for_driver(node_ip_address)
File "/home/rise/anaconda3/envs/chx-py3.10/lib/python3.10/site-packages/ray/_private/state.py", line 751, in get_node_to_connect_for_driver
return self.global_state_accessor.get_node_to_connect_for_driver(
File "python/ray/includes/global_state_accessor.pxi", line 218, in ray._raylet.GlobalStateAccessor.get_node_to_connect_for_driver
RuntimeError: b'GCS has started but no raylets have registered yet.'
During handling of the above exception, another exception occurred:
Traceback (most recent call last):
File "/home/rise/learn/leanDojo/extract.py", line 10, in <module>
traced_repo = trace(repo)
File "/home/rise/anaconda3/envs/chx-py3.10/lib/python3.10/site-packages/lean_dojo/data_extraction/trace.py", line 183, in trace
traced_repo = TracedRepo.load_from_disk(cached_path)
File "/home/rise/anaconda3/envs/chx-py3.10/lib/python3.10/site-packages/lean_dojo/data_extraction/traced_data.py", line 1481, in load_from_disk
with ray_actor_pool(_TracedRepoHelper, root_dir, repo) as pool:
File "/home/rise/anaconda3/envs/chx-py3.10/lib/python3.10/contextlib.py", line 135, in __enter__
return next(self.gen)
File "/home/rise/anaconda3/envs/chx-py3.10/lib/python3.10/site-packages/lean_dojo/utils.py", line 72, in ray_actor_pool
ray.init()
File "/home/rise/anaconda3/envs/chx-py3.10/lib/python3.10/site-packages/ray/_private/client_mode_hook.py", line 103, in wrapper
return func(*args, **kwargs)
File "/home/rise/anaconda3/envs/chx-py3.10/lib/python3.10/site-packages/ray/_private/worker.py", line 1645, in init
raise ConnectionError
ConnectionError
Process finished with exit code 1
|
Beta Was this translation helpful? Give feedback.
-
|
What will happen if you first run import ray
ray.init()Is there any error? BTW, if the problem is with Ray, a (slower) workaround is to disable Ray by setting the |
Beta Was this translation helpful? Give feedback.
-
|
If you cannot do import ray
ray.init(), the problem is specific to your particular system and Ray installation. There is nothing we can do on LeanDojo's side (other than you can set |
Beta Was this translation helpful? Give feedback.


If you cannot do
, the problem is specific to your particular system and Ray installation. There is nothing we can do on LeanDojo's side (other than you can set
NUM_PROCS=1to disable Ray). I'd suggest you ask in Ray's GitHub repo.