setessthresh
You can use the
setessthresh command
to set or change the ESS performance thresholds. To modify an existing threshold
setting for a specific device, use the
setessthresh command
again. Only new values specified change; unspecified settings are preserved.
- All changes to threshold settings apply immediately, even if a current
performance data collection task is already active and running.
- Threshold checking does not occur on a device unless a performance data
collection task has been created for that device and the task has been started
as part of a performance collection job. Performance data collection jobs,
when run, check devices to determine which thresholds are enabled in order
to see if any of the performance statistics, for which thresholds are set,
exceed the threshold value.
- Enabling a threshold status results in any detected exception data to
be recorded in the Performance Manager database. However, if the alert level
is set to none, a log message is entered into the MDM
event log and the exception information is saved in the Performance Manager
database. If the alert level is set to anything other than none, the Performance Manager sends an internal event to the Director.
Users can configure how these alerts should manifest (for example, SNMP, E-mail,
and so on) using Director's Event Action Planner.
- All ESS performance thresholds, except for I/O rate, have default critical
and warning values, and are enabled for threshold checking by default. None
of the SAN Volume Controller thresholds are enabled
by default, nor do they have default warning and critical values.
Syntax

>>-perfcli-- --setessthresh-- ---threshtype--+-diskutil--+-- --->
+-nvsfull---+
+-cachehold-+
+-iorate----+
'-all-------'
>---dev--+-ID [...]-+-- --+------------------+-- --------------->
'-all------' '--status--+-on--+-'
'-off-'
>--+------------------+-- --+-----------------+-- -------------->
'--warn--threshold-' '--err--threshold-'
>--+------------------+-- --+----------+-- --+--------+--------><
'--alert--+-warn-+-' '--restore-' '--quiet-'
+-err--+
'-none-'
Parameters
- -threshtype diskutil | nvsfull | cachehold
| iorate | all
- Specifies which ESS threshold is to be set. Optionally, the diskutil threshold (ESS) includes the percentage of the sequential I/O
filter when you use the setfilter command. One of the
following ESS thresholds can be specified:
- diskutil
- The disk utilization value (percent of time busy), for each ESS array.
- nvsfull
- The nonvolatile storage full value (percent of time full), for each
ESS cluster.
- cachehold
- The average cache hold time (in seconds), for each ESS cluster.
- iorate
- The input-output rate (total number of I/O requests), for each ESS cluster.
- all
- Sets all the thresholds.
- -dev id [...] | all
- Specifies the ESS device that will apply the threshold to be set. The
device ID is the nickname or model-serial-manufacturer of the device. Multiple
devices must be separated by commas. All devices must be of the same device
type. When you specify all the threshold applies to
all ESS devices.
- -status on | off
- An optional parameter that specifies whether the threshold type should
be enabled or disabled for checking. For all threshold types, except the I/O
rate, the default is on and enabled for checking.
- -warn threshold
- An optional parameter that specifies the warning threshold value. Valid
ranges are 0 - 100 for all devices. The default percentage values for each
warning threshold type are as follows:
- diskutil = 50
- nvsfull = 3
- cachehold = 60
- io = no default
- -err threshold
- An optional parameter that specifies the error threshold value. Valid
ranges are 0 - 100. The default percentage values for each error threshold
type are as follow:
- diskutil = 80
- nvsfull = 10
- cachehold = 30
- io = no default
- -alert warn | err | none
- An optional parameter that specifies which thresholds should send an
alert if the threshold is exceeded. One of the following alert values can
be specified:
- Warn. An alert is sent when the warning threshold is exceeded.
- Err. An alert is sent when the warning and error threshold is exceeded.
- None. No alerts are sent. This is the default.
 | Note: If the threshold
status is enabled for checking, but no alerts are set to be sent, a log message
is entered into the Multiple-Device Resource Manager event log, and exception
information is recorded in the database. |
- -restore
- An optional parameter that restores all default values for the specified
threshold type. This argument works only on those thresholds that have default
values.
- -quiet
- An optional parameter that turns off the confirmation prompt for this
command.
Description
You can use the setessthresh command to set or change
the ESS performance thresholds.
Return values
Threshold successfully set.
Possible failures
- Device device does not exist.
- Historical data must exist in the database for threshold checking to be
activated. You must use the mkcollector command or schedule
a data collection task in the GUI.
Examples
An invocation example:
perfcli setessthresh -threshtype iorate -dev 2105.111222 -warn 60 -alert err
The resulting output:
Threshold successfully set.