Wednesday, 18 September 2019

[JAGENT] [ERROR] [OGGMON-20415] java.lang.IllegalArgumentException: source parameter must not be null

Unable to Discover jagent in OEM with error [JAGENT] [ERROR] [OGGMON-20415] java.lang.IllegalArgumentException: source parameter must not be null

Solution:
1. Stop all goldengate process including manager and jagent
ggsci> stop jagent
ggsci> stop *
ggsci> stop manager

2.Delete datastore
ggsci> delete datastore

3.Create datastore
ggsci>create datastore

4.Start all process now
ggsci> start manager
ggsci> start *
ggsci> start jagent

Note: Datastore holds monitoring information supplied by the Oracle GoldenGate Extract, Replicat, and Manager processes

Delete and recreate datastore will help in multiple jagent monitoring issues

ERROR(OGG-05500) Detected database metadata mismatch between current trail file

Issue:

Goldengate replicat failed with error

ERROR   OGG-05500  Oracle GoldenGate Delivery for Oracle, *****.prm:  Detected database metadata mismatch between current trail file ./dirdat/ti005021 and the previous sequence

 

Solution:

This procedure stops Extract, and then creates a new trail file. The new database metadata is included in this new file with the transactions that started after any possible changes.


1.Stop transaction activity on the source database.

2.In GGSCI on the source system

issue the SEND EXTRACT command with the LOGEND option until it shows there is no more redo data to capture.

SEND EXTRACT EXTRACTNAME LOGEND

3.Stop Extract.

STOP EXTRACT EXTRACTNAME

4.On each target system

issue the SEND REPLICAT command with the STATUS

option until it shows a status of ??At EOF?? to indicate that it finished

processing all of the data in the trail. This must be done on all target

systems until all Replicat processes return ??At EOF.??

 

SEND REPLICAT REPLICATNAME STATUS

 

5.Stop the data pumps and Replicat

STOP EXTRACT PUMPNAME

STOP REPLICAT REPLICATNAME

 

6. Extract to roll over the local trail to the start of a new file

ALTER EXTRACT command with the ETROLLOVER option for the primary.

 

ALTER EXTRACT EXTRACTNAME, ETROLLOVER


7. Roll over the remote trail to the start of a new file

ALTER EXTRACT command with the ETROLLOVER option for the pumps

 

ALTER EXTRACT PUMPNAME, ETROLLOVER

 

8.Start Extract.

START EXTRACT EXTRACTNAME

 

9. Alter Extract and replicat to new sequence

New sequence number refers to the new trail sequence created by the ETROLLOVER..you could list the trail file directory to find out using ls -lrt. Use RBA 0 which tells the extract/pump or replicat at the beginning

 

ls -ltr ./dirdat/*

ALTER EXTRACT PUMPNAME, EXTSEQNO 11, EXTRBA 0

ALTER REPLICAT REPLICATNAME, EXTSEQNO seqno xx, EXTRBA 0

 

10.Start the data pumps.

START EXTRACT PUMPNAME

 

11.Start the Replicat processes.

START REPLICAT REPLICATNAME

 

Note: OGG-05500 Detected Database Metadata Mismatch between Current Trail File (Doc ID 1605434.1)

 

 


Monday, 8 April 2019

Cluster log location in 18c

The location of cluster log’s has been changed in 18c to $ORACLE_BASE/diag/crs/<hostname>/crs/trace/

[grid@node01~]$ ls -ltr $ORACLE_HOME/log/node01/crsd
 Total 0
[grid@node01 ~]$ crsctl query crs activeversion
Oracle Clusterware active version on the cluster is [18.0.0.0.0]

18c standard cluster:
Name of alert log: alert.log
Location of alert log: $ORACLE_BASE/diag/crs/node1/crs/trace

18c flex cluster:
Name of alert log: alert<hostname>.log
Location of alert log: $ORACLE_HOME/log/node1

13C OEM java.sql.SQLException: Connection Cache with this Cache Name does not exist


Perform the following steps on the OMS server as a work around
1. <OMS_HOME>/bin/emctl set property -name use_pooled_target_connections -value false

2. Restart oms
 <OMS_HOME>/bin/emctl stop oms -all
 <OMS_HOME>/bin/emctl start oms

Note: EM 13c: Target DB Home Page Shows 'Failed To Connect To The Target Database, Regions That Pull Real-time Data' (Doc ID 2474441.1)