I am trying to set up a new worker on our network that will connect to our Cryosparc master node. We can initiate jobs, but they hang at startup. The event log gives this:
================= CRYOSPARCW ======= 2022-10-12 20:39:00.928549 =========
Project P121 Job J115
Master node08.server.internal Port 39002
===========================================================================
========= monitor process now starting main process
MAINPROCESS PID 3462
Traceback (most recent call last):
File "<string>", line 1, in <module>
File "cryosparc_worker/cryosparc_compute/run.py", line 170, in cryosparc_compute.run.run
File "/opt/cryosparc-home/cryosparc_worker/cryosparc_compute/jobs/runcommon.py", line 98, in connect
cli.test_authentication(project_uid, job_uid)
AttributeError: 'CommandClient' object has no attribute 'test_authentication'
Process Process-1:
Traceback (most recent call last):
File "/opt/cryosparc-home/cryosparc_worker/deps/anaconda/envs/cryosparc_worker_env/lib/python3.7/multiprocessing/process.py", line 297, in _bootstrap
self.run()
File "/opt/cryosparc-home/cryosparc_worker/deps/anaconda/envs/cryosparc_worker_env/lib/python3.7/multiprocessing/process.py", line 99, in run
self._target(*self._args, **self._kwargs)
File "cryosparc_worker/cryosparc_compute/run.py", line 32, in cryosparc_compute.run.main
File "/opt/cryosparc-home/cryosparc_worker/cryosparc_compute/jobs/runcommon.py", line 98, in connect
cli.test_authentication(project_uid, job_uid)
AttributeError: 'CommandClient' object has no attribute 'test_authentication'
Since we can successfully intitiate jobs, I don’t believe it’s an ssh error. Also, all firewalls are disabled, so the ports aren’t being blocked.
The server in question runs Ubuntu 22.04. Any help is appreciated.
Are you referring to the new worker?
What version of CryoSPARC are you running?
What is the output of the command (run on the new worker) curl node08.server.internal:39002
?
It seems like the worker version you’re trying to connect is v4.0.1, since it’s looking for a function that only exists in this version.
You might need to re-download the correct worker package by specifying the version in the download link. curl -L https://get.cryosparc.com/download/worker-v3.3.1/$LICENSE_ID -o cryosparc_worker.tar.gz
After this is done you can install normally.
@stephan Thanks. I installed the 3.3.1 like you mentioned, but it seems like 3.3.1 cannot detect the GPUs. I get a pycuda error:
Traceback (most recent call last):
File "bin/connect.py", line 221, in <module>
gpu_devidxs = check_gpus()
File "bin/connect.py", line 91, in check_gpus
num_devs = print_gpu_list()
File "bin/connect.py", line 23, in print_gpu_list
import pycuda.driver as cudrv
File "/cryosparc-worker/cryosparc_worker/deps/anaconda/envs/cryosparc_worker_env/lib/python3.7/site-packages/pycuda/driver.py", line 62, in <module>
from pycuda._driver import * # noqa
ImportError: /cryosparc-worker/cryosparc_worker/deps/anaconda/envs/cryosparc_worker_env/lib/python3.7/site-packages/pycuda/_driver.cpython-37m-x86_64-linux-gnu.so: undefined symbol: _ZSt28__throw_bad_array_new_lengthv
4.0 worked, so is there a limit to the cuda version I can install?