Chapter 9: Performance Agent Commands › ADVISE COLLECT MODIFY › Fragmented File with an Extension Header › Command Qualifiers

Command Qualifiers

/BEGINNING=date

Specifies the date on which the daily data collection is to begin. Date represents both the date and time in standard DCL format.

If the date is in the future, any node running the data collection process begins collecting data when the beginning date is reached. If the beginning date is in the past, any node running the data collection process collects data provided that the ending date has not been reached and that the weekly data collection schedule indicates that data should be collected. The default value for the /BEGINNING qualifier is the date that the Performance Agent software was installed or the date on which the schedule file was created.

/ENDING=date

Specifies the date on which the daily data collection is to end. The ending date represents both the date and time in standard DCL format. By default, /ENDING is set to 1-JAN-2010. When the date specified with the /ENDING qualifier is reached, the data collection process is suspended.

If the ending date is in the future, all nodes listed in the schedule file collect data according to the data collection schedule and the /BEGINNING qualifier. If the ending date is in the past, the data collection is suspended and is not resumed until you change the ending date.

/CLASSES= list of classes

Specifies the classes of daily data that are kept. The default is ALL.

• ALL_METRICS
• PROCESS_METRICS
• IO_METRICS
• SYSTEM_METRICS

/EXCLUDE= ([NO]PROCESSES=(process_name,. . .),
[NO]PIDS=(pid,. . . ),
[NO]DEVICES=(device_name,. . .),
[NO]VOLUMES=(volume_name,. . .))

This qualifier allows you to change or add items to the exclude list associated with the collection definition. The main CPD collection definition does not allow exclusion items to be specified, however any alternate collection definition does.

If you choose to exclude specific processes for data collection, those processes are omitted from the collection. All other processes have data collected. Processes may be identified by either the PID, or the Process Name, but not both. Processes may not be 'excluded' if they are being included, so the /EXCLUDE and /INCLUDE qualifier are mutually exclusive for process data.

If you choose to exclude specific devices or volumes for data collection, those devices or volumes are omitted from the collection. All other devices or volumes have data collected. Disk devices may be identified by either the device name, or the volume name, but not both. Devices may not be 'excluded' if they are being included, so the /EXCLUDE and /INCLUDE qualifier are mutually exclusive for device data.

Negating the exclude keywords PROCESSES, PIDS, DEVICES, or VOLUMES allows you to remove all or any entries from the exclude list.

/DELETE_AFTER=number_of_days

Specifies the number of days that daily data is kept before the data collection process automatically deletes it. Number_of_days must be an integer from 1 to 99999. The default value is 7 for the CPD collection definition. The default value is 99,999 for alternate collection definitions.

/HOTFILE_QUEUE=value

OR/NOHOTFILE_QUEUE

Specifies the minimum average number of jobs in a queue for any disk before the hottest files for that disk are collected. This applies to the CPD collection definition only as alternate Performance Agents do not collect hot file data.

The queue size is averaged over a 2-minute time sample for each disk. If at the end of the 2 minutes, the number of jobs in the queue is greater than or equal to the set value, then the primary Performance Agent records the top five hottest files in the following six categories:

• Operations count
• Throughput
• Paging
• Swapping operations
• Split I/O rate
• Service time
• Read operations

The default is 0.33. The valid values are any positive floating point decimal numbers less than 100.00 and greater than or equal to 0.00.

The /NOHOTFILE_QUEUE qualifier disables the collection of hot files and as a side effect, reduces the Performance Agent overhead and disk space used.

/INCLUDE= ([NO]PROCESSES=(process_name,. . . ),
[NO]PIDS=(pid,. . .),
[NO]DEVICES=(device_name,. . .),
[NO]VOLUMES=(volume_name,. . .)) 

This qualifier allows you to change or add items to the list associated with the collection definition. You can specify items for inclusion for all but the CPD collection definition.

If you choose to include specific processes for data collection, only those processes are targeted for collection. All other processes do not have any data collected. Processes may be identified by either the PID, or the Process Name, but not both. Processes may not be 'excluded' if they are being included, so the /EXCLUDE and /INCLUDE qualifier are mutually exclusive for process data.

If you choose to include specific devices or volumes for data collection, only those devices or volumes are targeted for collection. All other devices or volumes do not have any data collected. Disk devices may be identified by either the device name, or the volume name, but not both. Devices may not be 'excluded' if they are being included, so the /EXCLUDE and /INCLUDE qualifier are mutually exclusive for device data.

Negating the include keywords PROCESSES, PIDS, DEVICES, or VOLUMES allows you to remove all or any entries from the include list.

/INTERVAL=seconds

Specifies the interval at which daily data is collected. The default value is 120 seconds. This applies only to alternate collection definitions. The interval is fixed for the primary Performance Agent.

/REMOVE_NODES=(nodename,[,nodename . . .])

This qualifier allows you to remove nodes associated with a particular Collection Definition. Wildcards may be used to remove all nodes or a single node for all Collection Definitions.

/REORDER_NODES=(nodename[,. . .])

Use this qualifier to change the order of the nodes listed in a particular collection definition or all collection definitions by using the wildcard (*) for the collection definition in the command.

/SCHEDULE=({day=(hour-range)[,. . .]|NOday}[,. . .])

Revises the weekly schedule for the specified collection definition. You can use this qualifier to suspend data collection to control the size of the daily data files.

day

Set as SUNDAY, MONDAY, TUESDAY, WEDNESDAY, THURSDAY, FRIDAY, SATURDAY, EVERYDAY, WEEKDAYS, or WEEKENDS.

hour-range

Is specified as m-n, where m and n are numbers from 0 to 24, and m is less than n. More than one hour range can be specified for a given day. Hour-range is mutually exclusive with the NO option.

If a day keyword is omitted, the schedule for that day is not changed. Data collection can be suspended for individual days of the week by negating the keyword (for example, NOSUNDAY) or for all of the days of the week by specifying the NOEVERYDAY keyword. The values [NO]WEEKDAYS and [NO]WEEKENDS enable or disable all data collection on weekdays and weekends.

You must specify an hour range for any day keyword. Do not include an hour range if you are specifying a negated day keyword, such as NOMONDAY.

Less inclusive keyword values override more inclusive values. For example, MONDAY=10-12 overrides EVERYDAY=8-17 for Monday however, the Performance Agent collects data from 8:00 a.m. to 5:00 p.m. for all of the other days of the week.