Skip to:

  1. Skip to Jira Navigation
  2. Skip to Side Navigation
  3. Skip to Main Content

Problems with "ResponseTimeSummary_Availability_Offenders" subreport

Description

There are several problems regarding this report.

I'll comment one problem at a time and ID them with numbers, in order to be more easily trackable.

Acceptance / Success Criteria

None

is duplicated by

Lucidchart Diagrams

Activity

Show:

Seth Leger June 16, 2016 at 1:57 AM

These changes have been merged into foundation-2016. Marking as fixed.

commit da1ea5dee2be518f29d610a4fc692a4e49ff3dc8

Seth Leger April 9, 2015 at 2:07 PM

We'll try and code-review these changes and then integrate them into 16.

Vitor Moreira February 16, 2015 at 2:12 PM

Problem 4) Unresolved outages aren't accounted because ifregainedservice is null and isn't considered by the OVERLAPS function.

Solution) In query "outage_scope", instead of using the ifregainedservice, we always use a non null value: ifregainedservice is there's a value, report's "end_time" if ifregainedservice is null.

Instead of

(SELECT outages.nodeid, greatest('$P!{START_TIME_STRING}'::TIMESTAMP,outages.iflostservice) as iflostservice, least('$P!{END_TIME_STRING}'::TIMESTAMP,COALESCE(outages.ifregainedservice,'$P!{END_TIME_STRING}'::TIMESTAMP)) as ifregainedservice FROM outages, service, events WHERE (iflostservice, ifregainedservice) OVERLAPS ('$P!{START_TIME_STRING}'::TIMESTAMP, '$P!{START_TIME_STRING}'::TIMESTAMP + '$P!{TIME_RANGE}'::INTERVAL) AND outages.serviceid = service.serviceid AND service.servicename = 'ICMP' AND outages.svclosteventid = events.eventid AND events.eventuei = 'uei.opennms.org/nodes/nodeDown') AS outages_scope

Should be

(SELECT outages.nodeid, greatest('$P!{START_TIME_STRING}'::TIMESTAMP,outages.iflostservice) as iflostservice, least('$P!{END_TIME_STRING}'::TIMESTAMP,COALESCE(outages.ifregainedservice,'$P!{END_TIME_STRING}'::TIMESTAMP)) as ifregainedservice FROM outages, service, events WHERE (iflostservice, COALESCE(ifregainedservice,'$P!{END_TIME_STRING}'::TIMESTAMP)) OVERLAPS ('$P!{START_TIME_STRING}'::TIMESTAMP, '$P!{START_TIME_STRING}'::TIMESTAMP + '$P!{TIME_RANGE}'::INTERVAL) AND outages.serviceid = service.serviceid AND service.servicename = 'ICMP' AND outages.svclosteventid = events.eventid AND events.eventuei = 'uei.opennms.org/nodes/nodeDown') AS outages_scope

Vitor Moreira February 16, 2015 at 2:05 PM

Problem 3a) if an outage is still active and the "lost" timestamp is older than the start_date of the report, the report will account the outage's time since it's beginning instead of the beginning of the report.
Problem 3b) if an outage is resolved and the "regained" timestamp is newer then the end_date of the report, the report will account the outage's time until it was resolved, instead of the ending of the report.

Solution 3a)
Instead of using the field iflostservice in query "outages_scope", the GREATEST function should be used, in order to return the greatest of start_time,iflostservice.

Solution 3b)
Instead of using the field ifregainedservice in query "outages_scope", the LEAST function should be used, in order to return the least of end_time,ifregainedservice.
NOTE: end_time is a new internal parameter, used to specify the end_time of the report

<parameter name="START_TIME" class="java.util.Date" isForPrompting="false"> <defaultValueExpression><![CDATA[new java.util.Date(new java.util.Date().getTime() - (86400 * 7 * 1000))]]></defaultValueExpression> </parameter> <parameter name="END_TIME" class="java.util.Date" isForPrompting="false"> <defaultValueExpression><![CDATA[new Date($P{START_TIME}.getTime() + (86400000L * $P{TIME_RANGE_NUMBER}))]]></defaultValueExpression> </parameter> <parameter name="END_TIME_STRING" class="java.lang.String" isForPrompting="false"> <defaultValueExpression><![CDATA[new SimpleDateFormat("yyyy-MM-dd HH:mm:ss").format($P{END_TIME})]]></defaultValueExpression> </parameter>

Instead of

(SELECT outages.nodeid, outages.iflostservice, outages.ifregainedservice FROM outages, service, events WHERE (iflostservice, ifregainedservice) OVERLAPS ('$P!{START_TIME_STRING}'::TIMESTAMP, '$P!{START_TIME_STRING}'::TIMESTAMP + '$P!{TIME_RANGE}'::INTERVAL) AND outages.serviceid = service.serviceid AND service.servicename = 'ICMP' AND outages.svclosteventid = events.eventid AND events.eventuei = 'uei.opennms.org/nodes/nodeDown') AS outages_scope

Should be

(SELECT outages.nodeid, greatest('$P!{START_TIME_STRING}'::TIMESTAMP,outages.iflostservice) as iflostservice, least('$P!{END_TIME_STRING}'::TIMESTAMP,COALESCE(outages.ifregainedservice,'$P!{END_TIME_STRING}'::TIMESTAMP)) as ifregainedservice FROM outages, service, events WHERE (iflostservice, ifregainedservice) OVERLAPS ('$P!{START_TIME_STRING}'::TIMESTAMP, '$P!{START_TIME_STRING}'::TIMESTAMP + '$P!{TIME_RANGE}'::INTERVAL) AND outages.serviceid = service.serviceid AND service.servicename = 'ICMP' AND outages.svclosteventid = events.eventid AND events.eventuei = 'uei.opennms.org/nodes/nodeDown') AS outages_scope

Vitor Moreira February 16, 2015 at 1:43 PM

Problem 2) Outage percentage is calculated using the average of outage's time, when should be used the total time of the outage.

Solution)
Instead of:

CASE WHEN AVG(EXTRACT (epoch from node_outages.duration)) > node_outages.avail_total THEN 100 ELSE AVG(EXTRACT (epoch from node_outages.duration)) * 100 / node_outages.avail_total

This should be:

CASE WHEN SUM(EXTRACT (epoch from node_outages.duration)) > node_outages.avail_total THEN 100 ELSE SUM(EXTRACT (epoch from node_outages.duration)) * 100 / node_outages.avail_total
Fixed

Details

Assignee

Reporter

Labels

Components

Fix versions

Affects versions

Priority

PagerDuty

Created February 16, 2015 at 1:17 PM
Updated June 16, 2016 at 5:41 AM
Resolved June 16, 2016 at 1:57 AM

Flag notifications