sddssnapshot

• description: sddssnapshot reads values of process variables and writes them to a file. An input file lists the process variables to be read.

  sddssnapshot differs from burtrb in that sddssnapshot may operate in a server mode in which a new file is written to the named output file whenever the signal SIGUSR1 is received by sddssnapshot. Another improvement over burtrb is that all data in the input file (even those not needed by the program) are transfered to the output file. sddssnapshot is more for data collection as opposed to backup and restore.

• example: The state of the APS storage ring is saved by writing values of process variables listed in SR.req to the snapshot file SR.snp:

  sddssnapshot SR.req SR.snp -nameOfData="Value"
  

  where the contents of the file SR.req are

  SDDS1
  &column
   name = ControlName,  type = string, &end
  &data
   mode = ascii, &end
  &data
   mode = "ascii", no_row_counts=1 &end
  S1A:Q1:CurrentAO
  S1A:Q2:CurrentAO
  ...
  

• synopsis:

  usage: sddssnapshot [-pipe[=input][,output]] [<input>] [<output>]
  [-ezcaTiming=<timeout>,<retries>] [-unitsOfData=<string>] [-nameOfData=<string>]
  [-serverMode=<pidFile>] [-average=<number>,<intervalInSec>]
  Takes a snapshot of EPICS scalar process variables.
  Requires the column "ControlName" with the process variable names.
  For server mode, writes a new file to the given filename whenever
  SIGUSR1 is received.  Exits when SIGUSR2 is received.
  

• files:
  • input file:

    The input file is an SDDS file with at least one column:

    • ControlName -- Required string column for the process variable or device name.

  • output file:

    The output file contains all columns of the input file including those not needed by the program plus a column named on the command line option -nameOfData. This column is defined as a double type and contains the readback values. Optionally the units of that readback column may be specified on the command line. Of course this option is useful only if all the process variables have the same units, as in the case of recording orbit values from all bpms. If the -average option is requested, then an additional double column is created with the name of the readback column with the ``StDev'' appended to it.

  • pid file:

    A process id file is created with option serverMode=<pidFile>. This file contains a single number which is the pid number of the running sddssnapshot process.

• switches:
  • -pipe[=input][,output] -- The standard SDDS Toolkit pipe option.
  • -ezcaTiming=<timeout>,<retries> -- Specifies tuning of ezca, the channel access interface used by sddssnapshot.
  • -nameOfData=<string> -- Column name to be given to the data collected. Default name is ``Value''.
  • -unitsOfData=<string> -- Optional. Name given to the units field of the column definition of the data to be collected. Default value is the null string.
  • -serverMode=<pidFile> -- Optional. Enables the server more. The file specified will be created and contain the process number of the present sddssnapshot process. This file is the mechanism through which the user will know to which process should the SIGUSR1 be sent. To activate one snapshot write, the user can type the command ``kill -SIGUSR1 `cat <pidFile>`''.
  • -average=<number>,<intervalInSec> -- Optional. On can specify the number of readings to average and the number of seconds interval between readings.

• see also:
  • burtrb
  • sddsmonitor
• author: M. Borland, ANL