OS
Watcher (OSW) is a collection of UNIX shell scripts intended to collect
and archive operating system and network metrics to aid support in
diagnosing performance issues. OSW operates as a set of background
processes on the server and gathers OS data on a regular basis, invoking
such Unix utilities as vmstat, netstat and iostat. OSW can be
downloaded from this metalink note “Doc ID: 301137.1 OS Watcher User Guide ”.
Control is passed to individually spawned operating system data collector processes, which in turn collect specific data, timestamp the data output, and append the data to pre-generated and named files.
Each data collector will have its own file, created and named by the File Manager process.
OSW invokes the distinct operating system utilities listed below as data collectors.
OSW will not put any significant performance affecting load on the system. It will have the same impact as running the regular OS command like netstat, ps etc.
These utilities will be supported, or their equivalents, as available for each supported target platform:
* ps
* top
* mpstat
* iostat
* netstat
* traceroute
* vmstat
The size of the archived files saved during the running of the OSW will be based on the user parameters set at the starting of OSW and the OS information. For example, if each file will contain an hour of data and the collection interval is 10 sec the amount of data will be bigger compared to collecting with an interval of 60 sec.
It is highly recommended that OSW be installed and run continuously on ALL cluster nodes, at all times.
Be sure to use separate directories per node for storing OSW output. When using OS Watcher in a RAC environment, each node must write its output files to a separate archive directory. Combining the output files under one archive (on shared storage) is not supported and causes the OSWg tool to crash. Shared storage is fine, but each node needs a separate archive directory.
You can refer to metalink document “Doc ID: 580513.1 How To Start OSWatcher Every System Boot ” to automatically start the data collection on system boot.
If you open SRs with Oracle, they may ask for this data, so it always handy having it in.
~~~ Cheers
Control is passed to individually spawned operating system data collector processes, which in turn collect specific data, timestamp the data output, and append the data to pre-generated and named files.
Each data collector will have its own file, created and named by the File Manager process.
OSW invokes the distinct operating system utilities listed below as data collectors.
OSW will not put any significant performance affecting load on the system. It will have the same impact as running the regular OS command like netstat, ps etc.
These utilities will be supported, or their equivalents, as available for each supported target platform:
* ps
* top
* mpstat
* iostat
* netstat
* traceroute
* vmstat
The size of the archived files saved during the running of the OSW will be based on the user parameters set at the starting of OSW and the OS information. For example, if each file will contain an hour of data and the collection interval is 10 sec the amount of data will be bigger compared to collecting with an interval of 60 sec.
It is highly recommended that OSW be installed and run continuously on ALL cluster nodes, at all times.
Be sure to use separate directories per node for storing OSW output. When using OS Watcher in a RAC environment, each node must write its output files to a separate archive directory. Combining the output files under one archive (on shared storage) is not supported and causes the OSWg tool to crash. Shared storage is fine, but each node needs a separate archive directory.
You can refer to metalink document “Doc ID: 580513.1 How To Start OSWatcher Every System Boot ” to automatically start the data collection on system boot.
If you open SRs with Oracle, they may ask for this data, so it always handy having it in.
~~~ Cheers
No comments :
Post a Comment