Aida Process Management

Guidelines For The Use Of This Document

This document briefly describes how to check the health of the Aida processes and to stop or restart them.

Checking the Health of Aida Processes

Checking the health of the production Aida processes is best done through the aidaDpServerStatus.bash script, which runs on the Unix watchdog machine. One may see the results through the UNIX watchdog interface. One can also run this script directly to verify that all of the Aida data servers (aka, "data providers") are responding correctly. The only process that might fail to respond with "success" is the VMS Aida BPM server, which responds with a "failure" indication when the beam is not on. One can run the cmlogviewer at the same time one runs the aidaDpServerStatus.bash script to verify that a message is produced from the VMS Aida BPM server indicating that it responded in any case. If all of the Aida data servers are responding with "success" (or just the VMS Aida BPM server is not due to the beam not being on), one can be sure that the Aida infrastructure processes (DaServer and DaNameServer) are also functioning correctly.

One may also check whether the UNIX Aida production servers are running by issuing the following command on a machine such as slcs2:

This will also show information about test servers. All of the production UNIX Aida servers currently run on mccas0, with the exception of the Aida NLCTA Channel Archiver server, which runs on opi00gtw04. The UNIX Aida production servers are currently:

  1. DaServer
  2. DaNameServer
  3. DpCaServer (Aida PEPII Channel Access Server)
  4. DpCaLclsServer (Aida LCLS Channel Access Server)
  5. DpChadsServer (Aida PEPII Channel Archiver Server)
  6. DpChadsServer (Aida NLCTA Channel Archiver Server, which runs on opi00gtw04)
  7. DpChadsLclsServer (Aida LCLS Channel Archiver Server)
  8. DpRdbServer (Aida RDB [aka Oracle] Server)
  9. DpModelServer (Aida UNIX Model Server)

The aidamanager utility may be run with no arguments to produce information about the parameters it expects.

If you get this error: (when restarting DpCaServer)

  • Fri Sep 02 15:28:47 PDT 2011: org.omg.CORBA.COMM_FAILURE: bind() failed:
    • Kill the CaRepeater then restart S99st.DpCaServer
      • (caRepeater is started automatically)
    • Check for /etc/rc3.d/S9xst.caRepeater
      • If file is missing contact Jingchen or Ken


UNIX Aida Process Management

To stop all of the UNIX Aida production processes (including test processes):

To restart all of the UNIX Aida production processes:

VMS Aida Process Management

The VMS Aida production server processes are:

  1. aida_dpslc (Aida SLC Database server)
  2. aida_dpslchist (Aida SLC History server)
  3. aida_dpslcklys (Aida SLC Klystron server)
  4. aida_dpslcmgnt (Aida SLC Magnet server)
  5. aida_dpslcmosc (Aida SLC Master Osscillator server)
  6. aida_dpslcutil (Aida SLC Utility server)
  7. aida_dpslcbpm (Aida SLC BPM server)
  8. aida_dpslcmodel (Aida SLC Model server)

They may be restarted individually on the VMS MCC machine under the SLCSHR account with commands of the following form:

where "process_name" is one of the VMS Aida server names listed above.

If it is desired to simply stop any of these processes, a command of the following form may be entered:

Author:  Bob Hall 17-Nov-2009

Modified: Jingchen Zhou 05-16-2012