>>- svctask -- -- startstats -- --------------------------------> >-- -interval -- time_in_minutes ------------------------------><
Statistics are collected at the end of each sampling period (as specified by the -interval parameter). These statistics are written to a file. A new file is created at the end of each sampling period. Separate files are created for managed disks and virtual disks statistics.
The files generated are written to the /dumps/iostats directory.
A maximum of 12 files are stored in the directory at any one time for each disk type, for example Nm_stats_<nodeid>_<date>_<time>, m_stats_<nodeid>_<date>_<time> and v_stats_<nodeid>_<date>_<time> files. Before the 13th file (for each type) is created, the oldest file of that type will be deleted.
These files can be listed by using the svcinfo lsiostatsdumps command.
The naming convention for these files is: <disk_type>_stats_<frontpanelid>_<date>_<time>. Where <disk_type> is m or Nm for managed disks and v for virtual disks, <frontpanelid> is the current configuration node ID, <date> is in the form of yymmdd, and <time> is in the form of hhmmss.
Examples of managed disk file names are: m_stats_lynn02_031123_07246 and Nm_stats_lynn02_031123_07246.
An example of a virtual disk file name is: v_stats_lynn02_031123_072426 or.
The statistics collected for each managed disk and each virtual disk, that appear in the file name with the format: m_stats_<nodeid>_<date>_<time> and v_stats_<nodeid>_<date> respectively, contain the following statistical information:
| Note: These statistics are collected for each node in the cluster at that point in time. |
The statistics collected for each managed disk that appear in the file name with the format: Nm_stats_<nodeid>_<date>_<time>, contain the following statistical information:
| Note: These statistics are collected only for the configuration node, at that point in time. |
An invocation example
svctask startstats -interval 25
The resulting output
No feedback