Troubleshooting
Problem
Health Monitor long-running queries.
Symptom
The Health Monitor agent fires the following
count(*) queries:
SELECT /*YANTRA*/ count(:"SYS_B_0") from YFS_ORDER_HEADER
WHERE ORDER_CLOSED = :"SYS_B_1" AND NEXT_ALERT_TS <= :"SYS_B_2" AND
SELECT /*YANTRA*/ count(:"SYS_B_0") from YFS_ITEM WHERE
(SYNCTS IS NULL OR MAX_MODIFYTS > SYNCTS) AND ITEM_GROUP_CODE = :"SYS_B_1" AND
KIT_CODE IN (:"SYS_B_2", :"SYS_B_3", :"SYS_B_4");
Depending on the
customer environment, the execution times can vary. In this particular
instance, the first query took 30 minutes per execution and the second one took
2 minutes per execution.
The Health Monitor agent to get the
count of pending jobs for a particular agent fires these queries. The System
Management Console displays this information (pending job count) for various
monitoring purposes.
In this case, an Order Monitor agent
fires the first query and a SEND_ITEM_CHANGES agent fires the second
query.
Resolving The Problem
If one does not to see the pending job count on
System Management Console, then set the CollectPendingJobs to N for Order
Monitors and the SEND_ITEM_CHANGES agent . By doing so, one can completely
avoid these queries. See related article HTG2804.
[{"Product":{"code":"SS6PEW","label":"IBM Sterling Order Management"},"Business Unit":{"code":"BU048","label":"IBM Software"},"Component":"Not Applicable","Platform":[{"code":"PF025","label":"Platform Independent"}],"Version":"All","Edition":"","Line of Business":{"code":"LOB59","label":"Sustainability Software"}}]
Historical Number
NFX6902
Product Synonym
[<p><b>]Severity[</b><p>];Normal;[<p><b>]Type[</b><p>];NormalFix
Was this topic helpful?
Document Information
Modified date:
16 June 2018
UID
swg21559659