Statistics Formatter Syntax

Syntax

To format a statistics file generated by the Statistics Parser too, use the fsstatsformatter tool: fsstatsformatter[-c | -C | -f | -h | -H | -i | -l | -m | -n | -o | -r | -s | -t | -T | -v | -V]

Options

-c = COMPONENT

Specifies the name of the component for which statistics are collected. If a subcomponent or statistics type is not specified, the statistics for all subcomponents and statistics types for this component are collected.

-C = COLUMNS

Specifies the column names to include in the collected statistics. COLUMNS is a comma separated list of the column names to retrieve. If column names contain a space, the COLUMNS list must be double-quoted:

fsstatsformatter -C = "first name,second name,third name"

The statsTime and duration columns are always included, so you do not need to specify them.

When listing Controller components statistics for named columns, the nodename is an optional column that is not included by default. Oracle recommends that you include the nodename on the list of named columns so that the data can be sorted and easily identified by node.

-f = FORMAT

Specifies the output format. CSV (comma-separated value) is the default format and currently the only option that can be specified, but a .csv extension is not automatically added to the output file name. To specify the .csv extension, specify an output file name using the -o option. If you are using a spreadsheet application, Oracle recommends that you use .csv as the file name extension.

-h

Displays this help page.

-H = HEADERTYPE

Header rows specify the component, subcomponent, type, and column names in the output file by default. Specify NONE to hide the headers and display only data rows.

-i = FILENAME

Specifies the name of the file from which to read the statistics data that was processed by fsstatsparser. If no file name is specified, the input defaults to the fsstats.sps file in the current working directory.

-l

Lists the available statistics in the statistics file. Statistics are listed by component, subcomponent, and statistics type.

-m

Specify this option to have a separate data file written for each component, subcomponent, and statistics type. This option applies only if the -o option is also used. The data files names include the component, subcomponent, and statistics type concatenated to the specified output file name. An output file name extension (such as .csv) can be specified with the file name. Oracle recommends that you specify a file name extension with the output file name. If you are using a spreadsheet application, Oracle recommends that you use .csv as the file name extension.

-n = NODENAME

Specifies that only data records from this Oracle FS System Controller or Drive Enclosure are collected. Specify the World-Wide Name (WWN) of the Controller or Drive Enclosure that generated the statistical data you want to obtain.

-o = FILENAME

Specifies the name of the file to which to write the formatted output. If not specified, the output defaults to stdout. File name extensions are not automatically added to the output file name. See also the -m option, which can be used in conjunction with the -o option to separate the output by component, subcomponent, and statistics type. Note that errors are printed to stderr.

-r

Specifies that all data is displayed from the data file without any processing applied. Without the -r option, counter data is normalized to turn it into rates, and ID values are translated into Fully Qualified Names (FQNs) where possible. With the -r option, counter data is expressed as accumulated values rather than rates, and ID values will not be translated into FQNs.

-s = NAME

Specifies that statistics collected are limited to those that match the specified subcomponent.

-t = STATISTICSTYPE

Must be used with the -c option. Specifies that statistics collected are limited to those that match the specified STATISTICSTYPE for the specified component. STATISTICSTYPE must be expressed as one of the following values:

• ERROR
• GENERAL
• UI

-T = TIME

Specifies a time range within which to retrieve the statistics. TIME is specified in the following format: START,END. START and END are formatted as follows:

MMDDYY:HH:MM:SS

The time is the local time of the client machine on which fsstatsformatter is running.

-v

Prints the fsstatsformatter version number.

-V

Specifies verbose output. While the statistics formatting process is running, status progress is printed to standard output (stdout). If you use verbose output, you should also use the -o option to send program output to a file to avoid mixing program output with progress output in stdout.

Examples

Run the following command to send CSV-formatted data from the default fsstatsparser output file fsstats.sps to your standard output (stdout) device:

fssstatsformatter

Run the following command to list the names and definitions of all the statistical values in the default fsstats.sps file:

fsstatsformatter -l

Run the following command to list the names and definitions of the statistical values for the Block Services component:

fsstatsformatter -l -c=BS

Run the following command to send CSV-formatted data from an fsstatsparser output file named my_stats.sps to your standard output (stdout) device:

fsstatsformatter -i=my_stats.sps

Run the following command to send CSV-formatted port errors to a file named porterrors.csv:

fsstatsformatter -c=VS -s=Port -o=porterrors.csv

Run the following command to send CSV-formatted data from the default fsstats.sps file to a series of files, one for each component, subcomponent, and statistic type, while sending verbose progress output to your standard output (stdout) device:

fsstatsformatter -m -o=mystats.csv -V

Run the following command to send CSV-formatted port error information for the time range 10:00 to 10:05 local client time on December 1, 2014 to your standard output (stdout) device:

fsstatsformatter -c=VS -s=Port -T=120114:10:00:00,120114:10:05:00