Data ONTAP enables you to control the format and destination of the output of the stats command. This could be useful if you are processing the information with another tool or script, or if you want to store the output in a file so you can process it at a later time.
|If you want to...||Then...|
|Send stats output to a file||Add -o filename to your stats show or stats stop command line.
filename is the pathname to the file you want to receive the stats output. The file does not need to exist, although any directory in the path must already exist.
|Determine whether the output is formatted in rows or columns||Add the -r or -c option to your stats show or stats stop command line.
The -r option formats the output in rows and is the default if the -I option is not specified.
|Specify a delimiter so that your output can be imported into a database or spreadsheet||Add the -d delimiter option to your stats show or stats stop command line.
The -d option only has effect if your output is in column format.
|Filter the output of the stats show command||Add -O name=value to the stats show command.
name is the name of the option you want to filter and value is on or off.
See the na_stats_preset(5) man page for a list of options.
The following example displays output in rows:
toaster> stats show qtree:*:nfs_ops qtree:vol1/proj1:nfs_ops:186/s qtree:vol3/proj2:nfs_ops:208/s
The -c option formats the output in columns and is the default only if the -I option is specified.
The following example displays output in columns:
toaster> stats show -c qtree:*:nfs_ops Instance nfs_ops /s vol1/proj1 143 vol3/proj2 408
If you are displaying multiple objects that have different counters, the column format may be difficult to read. In this case, use the row format.
In the following example, the same counter is listed as for the column output example, except that it is comma-delimited.
cli> stats show -d , -c qtree:*:nfs_ops Instance nfs_ops /s vol1/proj1,265 vol3/proj2,12
The command in the following example filters output of the stats show command with zero counter values: