Debugging DDS#

Overview#

This topic describes procedures for debugging the status of the DDS interface that powers the CSCs that run on the summit. There are 3 use-cases for DDS Daemons defined by their location and application.

  1. Node specific daemons which are mostly used for production environments

  2. Local daemons for local CSCs or services which are mostly used for non-production environments (i.e. development)

  3. Nublado daemons which are mainly used as a way to manually ensure system stability for night time operations.

Prerequisites#

  • Inside of a node where the DDS Daemon is running

Responsibilities#

Software Engineer

Procedure#

Checking the status of the Daemon#

In the case that a CSC has trouble starting due to DDS failures. Check the status of the ospl daemon using the following procedure.

  1. Setup the appropriate environment variables.

  2. Check that the daemon is running:

    ospl status
Vortex OpenSplice System with domain name "ospl_sp_ddsi" has died

  3. Start the daemon

    ospl start

Verification#

  • ospl daemon status is nominal when running ospl status

Troubleshooting#

In the case of a production node, the tmp directory which houses the daemon files can be wiped out. In the case of a non-production node, bad conditions can cause the daemon to crash In the case of a nublado node, these daemons can be forcibly shut down due to updates on the service end.