Worker test failed. Kill signal sent by unknown user

Install, Configure and Update
1

I just connected a new worker to a master node (two different workstations). When I test a job, the job launches but not running on the working. So I ran a worker test. But I got the following error:

image
image968×180 30.7 KB

Could you let me know what the problem is? Please let me know if any additional information is needed. Thanks!

2

@wxh180 What are the version and patch level of this CryoSPARC instance? If they are not 4.1.1 and 230110, please update and let us know if the problem persists.

3

@wtempel I just updated to the latest and patch. However, the problem persists and the previous worker does not work either now. It shows the job is launched but the log file is stuck at the “Running job on remote worker node” step.

4

Please can you:

  • inspect /path/to/cryosparc_master/run/update.log for errors
  • confirm that, for all worker installations (or is a single worker installation shared between all workers?),
    cat /path/to/cryosparc_worker/patch shows “230110”
  • the job directory is shared with and writeable on the workers at the same path as on the master
  • confirm password-less ssh to the cryosparc account on the workers from the master, also ensuring there are no non-password prompts for key changes, etc.
  • if a job.log file is already present in the stuck job’s directory, please inspect the file for error messages
  • inspect the output of
    cryosparcm log command_core for messages related to the stuck job
5

@wtempel I dig into it a bit today. I did not find any useful information in the log file. In fact, the one stuck has no log file generated. But I just realized that it is because the project was set on the SSD of the master node but not on the shared NAS between the master and worker. However, when I move the project to the share NAS that both master and worker have access to. I got a different error now:

image
image2556×614 91.2 KB

PS: when the master has restarted, do we need to disconnect the worker first and then reestablish the connection? Thanks!

6

Please post the log contents seen in the screen shot as text so that forum visitors with a similar problem may easily find this topic.
Please can you also post a screenshot of the Inputs and Parameters tab.

image
image1204×909 49.2 KB

If you are referring to cryosparcw connect ..., the answer is no:
cryosparcw connect ... needs to be run only when

  • a new worker needs to be connected to the CryoSPARC instance or
  • the configuration of an existing worker has changed

Worker configuration information is stored in the CryoSPARC database and persists between restarts of the master services or the master computer.

1 Like