Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Fail to run the example #96

Open
YaoJiayi opened this issue Mar 27, 2025 · 3 comments
Open

Fail to run the example #96

YaoJiayi opened this issue Mar 27, 2025 · 3 comments

Comments

@YaoJiayi
Copy link

YaoJiayi commented Mar 27, 2025

Running the python example under examples/ will result in "nixl._bindings.nixlBackendError: NIXL_ERR_BACKEND"

Using NIXL Plugins from:
/home/ubuntu/anaconda3/envs/nixl-integration/lib/python3.11/site-packages/.nixl.mesonpy.libs/plugins/
Loaded plugin UCX_MO
Loaded plugin UCX
Traceback (most recent call last):
  File "/home/ubuntu/jiayi/nixl/test/python/nixl_api_test.py", line 32, in <module>
    nixl_agent1 = nixl_agent("target")
                  ^^^^^^^^^^^^^^^^^^^^
  File "/home/ubuntu/anaconda3/envs/nixl-integration/lib/python3.11/site-packages/nixl/_api.py", line 89, in __init__
    self.backends[bknd] = self.agent.createBackend(bknd, init)
                          ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
nixl._bindings.nixlBackendError: NIXL_ERR_BACKEND
@tstamler
Copy link
Contributor

Hi @YaoJiayi this issue seems to be rising from just initializing the backend. This example uses the UCX backend by default. How are you building/installing NIXL for this example, and how did you install UCX? Thanks!

@thesues
Copy link

thesues commented Mar 31, 2025

I met the similar issue. it could be 'ucx does not support multithread'. You have to follow Dockfile's instructions not the README.md.
build ucx like this:

    ./autogen.sh && ./configure     \
        --enable-shared             \
        --disable-static            \
        --disable-doxygen-doc       \
        --enable-optimizations      \
        --enable-cma                \
        --enable-devel-headers      \
        --with-cuda=/usr/local/cuda \
        --with-verbs                \
        --with-dm                   \
        --with-gdrcopy=/usr/local   \
        --enable-mt &&              \
    make -j &&                      \
    make -j install-strip &&        \
    ldconfig

@mkhazraee
Copy link
Contributor

Hi Jiayi, any updates? Did the ucx rebuild fix the issue?

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
None yet
Projects
None yet
Development

No branches or pull requests

4 participants