All-Hands-AI
[Bug]: Released resolver doesn't run due to "Invalid token" error
Is there an existing issue for the same bug? (If one exists, thumbs up or comment on the issue instead).
- [x] I have checked the existing issues.
Describe the bug and reproduction steps
https://github.com/malhotra5/test-repo/actions/runs/14543443533/job/40805579570
^ I've attached logs for the error. Can't run the latest released resolver at all because of this. I think this would be affecting all users?
OpenHands Installation
GitHub resolver
OpenHands Version
No response
Operating System
None
Logs, Errors, Screenshots, and Additional Context
ERROR:root: File "<frozen runpy>", line 198, in _run_module_as_main
File "<frozen runpy>", line 88, in _run_code
File "/opt/hostedtoolcache/Python/3.12.10/x64/lib/python3.12/site-packages/openhands/resolver/resolve_issue.py", line 773, in <module>
main()
File "/opt/hostedtoolcache/Python/3.12.10/x64/lib/python3.12/site-packages/openhands/resolver/resolve_issue.py", line 769, in main
asyncio.run(issue_resolver.resolve_issue())
File "/opt/hostedtoolcache/Python/3.12.10/x64/lib/python3.12/asyncio/runners.py", line 195, in run
return runner.run(main)
^^^^^^^^^^^^^^^^
File "/opt/hostedtoolcache/Python/3.12.10/x64/lib/python3.12/asyncio/runners.py", line 118, in run
return self._loop.run_until_complete(task)
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "/opt/hostedtoolcache/Python/3.12.10/x64/lib/python3.12/asyncio/base_events.py", line 691, in run_until_complete
return future.result()
^^^^^^^^^^^^^^^
File "/opt/hostedtoolcache/Python/3.12.10/x64/lib/python3.12/site-packages/openhands/resolver/resolve_issue.py", line 568, in resolve_issue
output = await self.process_issue(
^^^^^^^^^^^^^^^^^^^^^^^^^
File "/opt/hostedtoolcache/Python/3.12.10/x64/lib/python3.12/site-packages/openhands/resolver/resolve_issue.py", line 279, in process_issue
await runtime.connect()
File "/opt/hostedtoolcache/Python/3.12.10/x64/lib/python3.12/site-packages/openhands/runtime/impl/docker/docker_runtime.py", line 189, in connect
await call_sync_from_async(self.setup_initial_env)
File "/opt/hostedtoolcache/Python/3.12.10/x64/lib/python3.12/site-packages/openhands/utils/async_utils.py", line 18, in call_sync_from_async
result = await coro
^^^^^^^^^^
File "/opt/hostedtoolcache/Python/3.12.10/x64/lib/python3.12/concurrent/futures/thread.py", line 59, in run
result = self.fn(*self.args, **self.kwargs)
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "/opt/hostedtoolcache/Python/3.12.10/x64/lib/python3.12/site-packages/openhands/utils/async_utils.py", line 17, in <lambda>
coro = loop.run_in_executor(None, lambda: fn(*args, **kwargs))
^^^^^^^^^^^^^^^^^^^
File "/opt/hostedtoolcache/Python/3.12.10/x64/lib/python3.12/site-packages/openhands/runtime/base.py", line 167, in setup_initial_env
self.add_env_vars(self.initial_env_vars)
File "/opt/hostedtoolcache/Python/3.12.10/x64/lib/python3.12/site-packages/openhands/runtime/base.py", line 227, in add_env_vars
obs = self.run(CmdRunAction(cmd))
^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "/opt/hostedtoolcache/Python/3.12.10/x64/lib/python3.12/site-packages/openhands/runtime/impl/action_execution/action_execution_client.py", line 310, in run
return self.send_action_for_execution(action)
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "/opt/hostedtoolcache/Python/3.12.10/x64/lib/python3.12/site-packages/openhands/runtime/impl/action_execution/action_execution_client.py", line 292, in send_action_for_execution
response = self._send_action_server_request(
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "/opt/hostedtoolcache/Python/3.12.10/x64/lib/python3.12/site-packages/tenacity/__init__.py", line 338, in wrapped_f
return copy(f, *args, **kw)
^^^^^^^^^^^^^^^^^^^^
File "/opt/hostedtoolcache/Python/3.12.10/x64/lib/python3.12/site-packages/tenacity/__init__.py", line 477, in __call__
do = self.iter(retry_state=retry_state)
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "/opt/hostedtoolcache/Python/3.12.10/x64/lib/python3.12/site-packages/tenacity/__init__.py", line 378, in iter
result = action(retry_state)
^^^^^^^^^^^^^^^^^^^
File "/opt/hostedtoolcache/Python/3.12.10/x64/lib/python3.12/site-packages/tenacity/__init__.py", line 400, in <lambda>
self._add_action_func(lambda rs: rs.outcome.result())
^^^^^^^^^^^^^^^^^^^
File "/opt/hostedtoolcache/Python/3.12.10/x64/lib/python3.12/concurrent/futures/_base.py", line 449, in result
return self.__get_result()
^^^^^^^^^^^^^^^^^^^
File "/opt/hostedtoolcache/Python/3.12.10/x64/lib/python3.12/concurrent/futures/_base.py", line 401, in __get_result
raise self._exception
File "/opt/hostedtoolcache/Python/3.12.10/x64/lib/python3.12/site-packages/tenacity/__init__.py", line 480, in __call__
result = fn(*args, **kwargs)
^^^^^^^^^^^^^^^^^^^
File "/opt/hostedtoolcache/Python/3.12.10/x64/lib/python3.12/site-packages/openhands/runtime/impl/action_execution/action_execution_client.py", line 128, in _send_action_server_request
return send_request(self.session, method, url, **kwargs)
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "/opt/hostedtoolcache/Python/3.12.10/x64/lib/python3.12/site-packages/tenacity/__init__.py", line 338, in wrapped_f
return copy(f, *args, **kw)
^^^^^^^^^^^^^^^^^^^^
File "/opt/hostedtoolcache/Python/3.12.10/x64/lib/python3.12/site-packages/tenacity/__init__.py", line 477, in __call__
do = self.iter(retry_state=retry_state)
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "/opt/hostedtoolcache/Python/3.12.10/x64/lib/python3.12/site-packages/tenacity/__init__.py", line 378, in iter
result = action(retry_state)
^^^^^^^^^^^^^^^^^^^
File "/opt/hostedtoolcache/Python/3.12.10/x64/lib/python3.12/site-packages/tenacity/__init__.py", line 400, in <lambda>
self._add_action_func(lambda rs: rs.outcome.result())
^^^^^^^^^^^^^^^^^^^
File "/opt/hostedtoolcache/Python/3.12.10/x64/lib/python3.12/concurrent/futures/_base.py", line 449, in result
return self.__get_result()
^^^^^^^^^^^^^^^^^^^
File "/opt/hostedtoolcache/Python/3.12.10/x64/lib/python3.12/concurrent/futures/_base.py", line 401, in __get_result
raise self._exception
File "/opt/hostedtoolcache/Python/3.12.10/x64/lib/python3.12/site-packages/tenacity/__init__.py", line 480, in __call__
result = fn(*args, **kwargs)
^^^^^^^^^^^^^^^^^^^
File "/opt/hostedtoolcache/Python/3.12.10/x64/lib/python3.12/site-packages/openhands/runtime/utils/request.py", line 54, in send_request
raise RequestHTTPError(
ERROR:root:<class 'openhands.runtime.utils.request.RequestHTTPError'>: Server error '500 Internal Server Error' for url 'http://localhost:30526/execute_action'
For more information check: https://developer.mozilla.org/en-US/docs/Web/HTTP/Status/500
Details: Traceback (most recent call last):
File "/openhands/code/openhands/events/serialization/action.py", line 122, in action_from_dict
decoded_action = action_class(**args)
^^^^^^^^^^^^^^^^^^^^
TypeError: CmdRunAction.__init__() got an unexpected keyword argument 'is_static'
During handling of the above exception, another exception occurred:
Traceback (most recent call last):
File "/openhands/code/openhands/runtime/action_execution_server.py", line 632, in execute_action
action = event_from_dict(action_request.action)
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "/openhands/code/openhands/events/serialization/event.py", line 56, in event_from_dict
evt = action_from_dict(data)
^^^^^^^^^^^^^^^^^^^^^^
File "/openhands/code/openhands/events/serialization/action.py", line 132, in action_from_dict
raise LLMMalformedActionError(
openhands.core.exceptions.LLMMalformedActionError
Looking into this now
- I think experimental is down due to action serialization issue
- I think released resolver is down but the fix was merged today in #7934
Looks like experimental works. I think the reason for the is_static issue is the runtime is on a different version (older) than the code running.
Hmm this is weird, I ran openhands experimental on a test repo and it still failed: https://github.com/malhotra5/test-repo/actions/runs/14544104353/job/40807085671
EDIT: Specifically this workflow run installed openhands using pip install git+https://github.com/all-hands-ai/openhands.git
^ Oh sorry for the noise experimental does work, my workflow definition in my personal repo was incorrect
Although looks like we should patch release #7934 maybe
Hmm this is weird, I ran openhands experimental on a test repo and it still failed: https://github.com/malhotra5/test-repo/actions/runs/14544104353/job/40807085671
EDIT: Specifically this workflow run installed openhands using
pip install git+https://github.com/all-hands-ai/openhands.git
I see, but for some reason this one ran nikolaik https://github.com/malhotra5/test-repo/actions/runs/14544104353/job/40807085671#step:10:40
Yeah the workflow definition wasn't passing the experimental env to resolve_issue.py; so instead of building the runtime image it was pulling it
That's was an improper workflow definition on my part
This run worked as expected when I fixed my workflow definition: https://github.com/malhotra5/test-repo/actions/runs/14544289306/job/40807522801
Note for released resolver, the error logs are
ERROR:root: File "<frozen runpy>", line 198, in _run_module_as_main
File "<frozen runpy>", line 88, in _run_code
File "/opt/hostedtoolcache/Python/3.12.10/x64/lib/python3.12/site-packages/openhands/resolver/resolve_issue.py", line 734, in <module>
main()
File "/opt/hostedtoolcache/Python/3.12.10/x64/lib/python3.12/site-packages/openhands/resolver/resolve_issue.py", line 673, in main
Error: raise ValueError('Token is invalid.')
ERROR:root:<class 'ValueError'>: Token is invalid.
Error: Process completed with exit code 1.
This should be fixed with a release
Answering my own question: when i run with 'fix-me', it fails, but when i run with 'fix-me-experimental' with the same setup it works.
Would there be any interest in adding a openhands-version flag to the resolver? I could make a PR. I notice that when I run the resolver every couple of days or so, if i dont update my workflow, it sometimes will fail, and I don't really want to do that much work in my own repo to keep up with resolver bugs and workflow updates, I would like something more stable. a simple openhands-version would allow people like me to always pull the same stable version that i set my workflow to. otherwise, my workflow is the way it is but the newest """stable""" version expects something different or otherwise errors like this
Yes I think that's a good idea @ryx2. It should be like anything else. You choose the version to run and whenever you choose to, you can upgrade it. This will definitely help with things all of a sudden not working.
