The Agent logs show metrics related errors in the following format:
<metric_name> has negative value: <negative value>. value has been changed to zero
The monitored Instance was reset, or was taken down while the agent is working and collecting database metrics
In many metrics that are being collected we submit the differences between the current value and the previous value, for those metrics we perform a logical check to see if the current value is indeed bigger than the previous one, If it not - we print an error and submit a zero value.
This errors can happened if the DB instance is being reset during the agent work. Since these metrics are being accumulated over time - they are being initialized on DB restart and that can lead to a situation where the current value is smaller than the previous one.
Here is an example:
ECHO <DB_SQL_Server/5.7.5.35/DB_SQL_Server/DBSS-myinstance.mydomain.com> ERROR [Agent DBSS-myinstance.mydomain.com OSDBMODEL thread-1] com.quest.qsi.dbwc.sqlserver.agent.submitters.DataModelParser - BytesTotalPerSec has negative value: -1.436172314E9. value has been changed to zero
ECHO <DB_SQL_Server/5.7.5.35/DB_SQL_Server/DBSS-myinstance.mydomain.com> ERROR [Agent DBSS-myinstance.mydomain.com JDBCMONITORED thread-2] com.quest.qsi.dbwc.sqlserver.agent.submitters.DataModelParser - number_of_super_latches has negative value: -2.0. Current value 140.0, old value 142.0. value has been changed to zero
ECHO <DB_SQL_Server/5.7.5.35/DB_SQL_Server/DBSS-myinstance.mydomain.com> ERROR [Agent DBSS-myinstance.mydomain.com JDBCMONITORED thread-2] com.quest.qsi.dbwc.sqlserver.agent.submitters.DataModelParser - number_of_super_latches has negative value: -2.0. value has been changed to zero
These kind of errors do not indicate any corruption or malfunction on the Agent side.
© ALL RIGHTS RESERVED. Feedback Terms of Use Privacy Cookie Preference Center