Cryosparc2_worker installation problem on Centos7

This is the problem I met when installing cryosparc2 on a computer with GPU and cuda installed correctly(but it’s not connected to the internet). So I tried to install cryosparc2_master on another CPU computer( connected to internet ) and then install cryosparc2_worker on the GPU computer. But I also met problems when connecting them.

Hope for your help

Hi @chenkj,

Is this related to this?

Running into the same issue here too. I recently changed the hostname on the computer (Single installation). So, I accordingly requested the name change using the bin/cryosparcw but I kept getting errors. So, I uninstalled and reinstalled cryosparc (stopped cryosparcm, deleted master and worker, untarred, reinstalled) with the same port number and a different port number to no avail.
The database and the command core start but then the error is thrown. Specifically,
HTTPConnectionPool(host="jake, port=39102) where the actual hostname is Jake.

Using Centos 7. Tried changing the firewall settings. Same error.

Any help is appreciated! Thank you

Cryosparcm start results in

cryosparcm log command_core output

    [JSONRPC ERROR  2020-04-02 12:23:31.516327  at  get_num_active_licenses ]
    Traceback (most recent call last):
      File "cryosparc2_command/command_core/", line 114, in wrapper
        res = func(*args, **kwargs)
      File "cryosparc2_command/command_core/", line 1421, in get_num_active_licenses
        for j in jobs_running:
      File "/home/cryosparc_user/software/cryosparc/cryosparc2_master/deps/anaconda/lib/python2.7/site-packages/pymongo/", line 1114, in next
        if len(self.__data) or self._refresh():
      File "/home/cryosparc_user/software/cryosparc/cryosparc2_master/deps/anaconda/lib/python2.7/site-packages/pymongo/", line 1036, in _refresh
      File "/home/cryosparc_user/software/cryosparc/cryosparc2_master/deps/anaconda/lib/python2.7/site-packages/pymongo/", line 873, in __send_message
      File "/home/cryosparc_user/software/cryosparc/cryosparc2_master/deps/anaconda/lib/python2.7/site-packages/pymongo/", line 888, in _send_message_with_response
        server = topology.select_server(selector)
      File "/home/cryosparc_user/software/cryosparc/cryosparc2_master/deps/anaconda/lib/python2.7/site-packages/pymongo/", line 214, in select_server
      File "/home/cryosparc_user/software/cryosparc/cryosparc2_master/deps/anaconda/lib/python2.7/site-packages/pymongo/", line 189, in select_servers
    ServerSelectionTimeoutError: jake:39101: [Errno -2] Name or service not known
    Traceback (most recent call last):
      File "cryosparc2_command/command_core/", line 198, in background_worker
      File "cryosparc2_command/command_core/", line 1428, in concurrent_job_monitor
      File "cryosparc2_command/command_core/", line 123, in wrapper
        raise e
    ServerSelectionTimeoutError: jake:39101: [Errno -2] Name or service not known

and cryopsparcm status output

    [cryosparc_user@Jake cryosparc2_master]$ cryosparcm status
    CryoSPARC System master node installed at
    Current cryoSPARC version: v2.14.2

    cryosparcm process status:

    app                              STOPPED   Not started
    app_dev                          STOPPED   Not started
    command_core                     RUNNING   pid 185275, uptime 0:09:44
    command_proxy                    RUNNING   pid 185440, uptime 0:09:11
    command_rtp                      STOPPED   Not started
    command_vis                      STARTING  
    database                         RUNNING   pid 185189, uptime 0:09:47
    watchdog_dev                     STOPPED   Not started
    webapp                           RUNNING   pid 185460, uptime 0:09:09
    webapp_dev                       STOPPED   Not started



I seem to have fixed it.
This is my understanding of the issue. Please correct if you think otherwise. The reference to the hostname was being called from /etc/hosts and the hostname was listed as localhost. This should normally work but it was parsing the hostname with a lower case “jake” vs an uppercase “Jake” which is the actual hostname. When I changed the hostname in the /etc/hosts file to “Jake”, it ended up starting fine but then I could not connect using “localhost:39100” in the browser. I had to use “Jake:39100”.
To make sure, everything actually works, I uninstalled and reinstalled everything with the usual port number (39000) and it installs seamlessly.

The original post seems like a firewall issue though.