Documentation

Performance alerting

STOR2RRD has build-in alerting feature. Alerting is based on reaching of performance thresholds.
You can define alarms for any storage and its volumes, pools or host.

Sources

  • Storage total (summary of all pools)
  • Pool
  • Volume
  • Host

Metrics

  • io: IO/sec total (read + write)
  • io read: IO/sec read
  • io write: IO/sec write
  • data: data throughput in MB/sec (read + write)
  • read: read data throughput
  • write: write data throughput
  • resp time: response time, latency
  • resp time read: response time read
  • resp time write: response time write
Note that some storage types do not support all metrics.

Types

  • Emailing. You can place direct email address on each directive, use email groups or default email address.
    Alert email has attached graph which shows utilization of the metric for past 25 hours.
  • Nagios support. You can configure Nagios to pick up alarms from STOR2RRD via standard NRPE module.
    Nagios plug-in installation
  • External alerting via external shell script. Each alert can invoke defined script with given parameters.
    You can use it for your integration needs.
  • SNMP trap: follow this to configure it
  • Alerting plug-ins to other monitoring tools can be developed on demand especially for customers under support contract

Configurables

  • Limit: threshold level in IO per second, MB per seconds or mili seconds
  • Peak: The period of time in which avg traffic utilization has to be over the specified limit to generate an alert.
  • Repeat: Default time in minutes which says how often you should be alerted.
    You can specify per storage/pool/volume different value in alert repeat time column of each alert
  • Exclude hours: time range in hours when alerts will be ignored
    It is useful for exclusion of nightly batch jobs which usualy causing a big load.
  • Email group: you can create different email groups and direct alarms to them
All configuration is done through the UI item "Alerting Configuration"

Email setup

Use your company's default email setup on the operating system you run STOR2RRD.
There is no any specific setup in the tool, it uses setting on the operating system.
In case of problems
  • Try send out a testing email from cmd line 
    $ echo "test message"| mailx -s test your_email@company.com
  • Check succcess in mail log /var/log/maillog (Linux), /var/log/mail (AIX)
Email setup on the Virtual Appliance

3rd party integrations

  • AlarmsOne
  • Service Now
  • Riverbed NetIM
  • HPE OpenView
  • Jira
  • Nagios
  • others can be done as needed

Email alert example 1

    Alert when average write response time for storwize-home and pool01 is greater then 2.5 mili seconds during last 15 mins

    Subject 
    STOR2RRD: POOL alert for storwize-home:pool01: response time write, actual value: 2.67 ms (limit 2.5)
    Text 
    STOR2RRD alert
 Time: 18:24:00 18/06/2016
 Storage: storwize-home
 POOL: pool01
 Metric: response time write
 Average response time during last 15mins: 2.67 ms  
 Limit: 2.5
    Attachment
    STOR2RRD alert example 1


Email alert example 2

    Alert when average overall IO for storage storwize-home is greater then 50 IO/sec during last 5 mins

    Subject 
    STOR2RRD: POOL-ALL: alert for storwize-home:all pools: io, actual value: 65 IO/sec (Limit 50)
    Text 
    
STOR2RRD alert
 Time: 19:48:00 18/06/2016
 Storage: storwize-home
 POOL-ALL: all pools
 Metric: io
 Average throughput during last 5mins: 65 IO/sec
 Limit: 50
    Attachment
    STOR2RRD alert example 2