DEV Community: Prasenjit Mahato

Datapump - Exclude table AUD$

Prasenjit Mahato — Thu, 22 Oct 2026 10:10:59 +0000

There was a requirement to exclude the AUD$ table of the SYS schema from the export dump.
The AUD$ table is already being collected daily in a separate database and retained for three years for reporting purposes. Excluding it from the export avoids redundant data transfer and ensures consistency with the existing retention and reporting processes.
Initially, I assumed excluding the table would be straightforward. My parameter file consisted of the following lines:

directory=DATAPUMP_BACKUP
full=Y
dumpfile=full_datapump_<dbname>_%L.dmp
logfile=full_datapump_<dbname>_18-08-2025.log
metrics=Y
logtime=ALL
parallel=8
encryption=ALL
encryption_algorithm=AES256
encryption_password=xxxxxxxxxxxxxx
compression=ALL
compression_algorithm=medium
flashback_time=systimestamp
EXCLUDE=STATISTICS
EXCLUDE=SCHEMA:"IN ('MAPLE')"
EXCLUDE=TABLE:"IN ('AUD$', 'EXT_BIWKNZIEKTE', 'EXT_BIWKNZIEKTEP')"

The datapump backup was initiated and the resulting logfile was reviewed. Unexpectedly, the AUD$ table was included in the backup despite being excluded in the parameter file. Excerpts from the logfile:

;;;
Export: Release 19.0.0.0.0 - Production on Mon Aug 18 08:42:40 2025
Version 19.27.0.0.0

Copyright (c) 1982, 2019, Oracle and/or its affiliates.  All rights reserved.
;;;
Connected to: Oracle Database 19c EE Extreme Perf Release 19.0.0.0.0 - Production
18-AUG-25 08:42:53.841: ;;; **************************************************************************
18-AUG-25 08:42:53.866: ;;; Parfile values:
18-AUG-25 08:42:53.885: ;;;  parfile:  exclude=STATISTICS,SCHEMA:"IN ('MAPLE')",
18-AUG-25 08:42:53.905: ;;;  _parfile: TABLE:"IN ('AUD$', 'EXT_BIWKNZIEKTE', 'EXT_BIWKNZIEKTEP')"
18-AUG-25 08:42:53.937: ;;;  parfile:  flashback_time=systimestamp
18-AUG-25 08:42:53.954: ;;;  parfile:  compression_algorithm=medium
18-AUG-25 08:42:53.978: ;;;  parfile:  compression=ALL
18-AUG-25 08:42:53.992: ;;;  parfile:  encryption_password=********
18-AUG-25 08:42:54.023: ;;;  parfile:  encryption_algorithm=AES256
18-AUG-25 08:42:54.041: ;;;  parfile:  encryption=ALL
18-AUG-25 08:42:54.069: ;;;  parfile:  parallel=8
18-AUG-25 08:42:54.097: ;;;  parfile:  logtime=ALL
18-AUG-25 08:42:54.117: ;;;  parfile:  metrics=Y
18-AUG-25 08:42:54.147: ;;;  parfile:  logfile=full_datapump_<dbname>_18-08-2025.log
18-AUG-25 08:42:54.181: ;;;  parfile:  dumpfile=full_datapump_<dbname>_%L.dmp
18-AUG-25 08:42:54.204: ;;;  parfile:  full=Y
18-AUG-25 08:42:54.222: ;;;  parfile:  directory=DATAPUMP_BACKUP
18-AUG-25 08:42:54.244: ;;; **************************************************************************
18-AUG-25 08:42:55.630: FLASHBACK automatically enabled to preserve database integrity.
18-AUG-25 08:43:05.376: Starting "SYSTEM"."SYS_EXPORT_FULL_01":  system/******** parfile=exp_<dbname>.par
18-AUG-25 08:43:07.026: W-1 Startup took 1 seconds
18-AUG-25 08:43:10.201: W-3 Startup took 1 seconds
.
.
.
.
18-AUG-25 08:45:40.207: W-5      Completed 1654 TABLE objects in 110 seconds
18-AUG-25 08:45:53.451: W-4 . . exported "SYS"."AUD$"                                110.2 MB  197763 rows in 46 seconds using direct_path
.
.
.

During the investigation, I began to doubt whether I had correctly updated the parameter file. Further review of the Oracle documentation led me to Doc ID 2383356.1, which recommends that the preferred method to exclude SYS.AUD$ objects during export or import is to use EXCLUDE=AUDIT_TRAILS. I then modified my parameter file to include this parameter.

directory=DATAPUMP_BACKUP
full=Y
dumpfile=full_datapump_<dbname>_%L.dmp
logfile=full_datapump_<dbname>_18-08-2025.log
metrics=Y
logtime=ALL
parallel=8
encryption=ALL
encryption_algorithm=AES256
encryption_password=xxxxxxxxxxxxx
compression=ALL
compression_algorithm=medium
flashback_time=systimestamp
EXCLUDE=AUDIT_TRAILS
EXCLUDE=STATISTICS
EXCLUDE=SCHEMA:"IN ('MAPLE')"
EXCLUDE=TABLE:"IN ('EXT_BIWKNZIEKTE', 'EXT_BIWKNZIEKTEP')"

It worked like a charm !

In conclusion, excluding the AUD$ table from the export dump using the EXCLUDE=AUDIT_TRAILS parameter in the Oracle datapump utility proved to be an effective solution. This experience highlights the importance of consulting official documentation.

Further Links and Information
Oracle Doc ID 2383356.1 - DataPump - EXCLUDE AUD$ does not work in 12c

Automatic maintenance tasks controls the automatic optimizer statistics

Prasenjit Mahato — Thu, 22 Oct 2026 10:07:25 +0000

Query to check the current state of jobs in CDB_AUTOTASK_CLIENT of the database, auto optimizer stats collection job is disabled in PDBs.

SELECT 
    COALESCE(p.name, (SELECT name FROM v$database)) AS DATABASE,
    CASE 
        WHEN c.con_id = 1 THEN 'CDB'
        WHEN c.con_id = 2 THEN 'SEED'
        ELSE 'PDB'
    END AS type_db,
    c.client_name,
    c.status,
    c.attributes
FROM cdb_autotask_client c
LEFT JOIN v$pdbs p
       ON c.con_id = p.con_id
ORDER BY type_db, client_name;

       DATABASE    TYPE_DB                        CLIENT_NAME     STATUS                                                ATTRIBUTES
_______________ __________ __________________________________ __________ _________________________________________________________
CON1            CDB        auto optimizer stats collection    ENABLED    ON BY DEFAULT, VOLATILE, SAFE TO KILL
CON1            CDB        auto space advisor                 ENABLED    ON BY DEFAULT, VOLATILE, SAFE TO KILL
CON1            CDB        sql tuning advisor                 ENABLED    ONCE PER WINDOW, ON BY DEFAULT, VOLATILE, SAFE TO KILL
CON1PDB1        PDB        auto optimizer stats collection    DISABLED   ON BY DEFAULT, VOLATILE, SAFE TO KILL
CON1PDB2        PDB        auto optimizer stats collection    DISABLED   ON BY DEFAULT, VOLATILE, SAFE TO KILL
CON1PDB1        PDB        auto space advisor                 DISABLED   ON BY DEFAULT, VOLATILE, SAFE TO KILL
CON1PDB2        PDB        auto space advisor                 DISABLED   ON BY DEFAULT, VOLATILE, SAFE TO KILL
CON1PDB2        PDB        sql tuning advisor                 DISABLED   ONCE PER WINDOW, ON BY DEFAULT, VOLATILE, SAFE TO KILL
CON1PDB1        PDB        sql tuning advisor                 DISABLED   ONCE PER WINDOW, ON BY DEFAULT, VOLATILE, SAFE TO KILL

9 rows selected.

Query to check the configured window, here the autotask was disabled in PDBs.

SELECT 
    COALESCE(p.name, (SELECT name FROM v$database)) AS DATABASE,
    CASE 
        WHEN c.con_id = 1 THEN 'CDB'
        WHEN c.con_id = 2 THEN 'SEED'
        ELSE 'PDB'
    END AS type_db,
    c.window_name,
    c.autotask_status,
    c.optimizer_stats
FROM cdb_autotask_window_clients c
LEFT JOIN v$pdbs p
       ON c.con_id = p.con_id
ORDER BY type_db, window_name;

      DATABASE    TYPE_DB         WINDOW_NAME    AUTOTASK_STATUS    OPTIMIZER_STATS
_______________ __________ ___________________ __________________ __________________
CON1            CDB        FRIDAY_WINDOW       ENABLED            ENABLED
CON1            CDB        MONDAY_WINDOW       ENABLED            ENABLED
CON1            CDB        SATURDAY_WINDOW     ENABLED            ENABLED
CON1            CDB        SUNDAY_WINDOW       ENABLED            ENABLED
CON1            CDB        THURSDAY_WINDOW     ENABLED            ENABLED
CON1            CDB        TUESDAY_WINDOW      ENABLED            ENABLED
CON1            CDB        WEDNESDAY_WINDOW    ENABLED            ENABLED
CON1PDB1        PDB        FRIDAY_WINDOW       DISABLED           DISABLED           
CON1PDB2        PDB        FRIDAY_WINDOW       DISABLED           DISABLED           
CON1PDB1        PDB        MONDAY_WINDOW       DISABLED           DISABLED           
CON1PDB2        PDB        MONDAY_WINDOW       DISABLED           DISABLED           
CON1PDB1        PDB        SATURDAY_WINDOW     DISABLED           DISABLED           
CON1PDB2        PDB        SATURDAY_WINDOW     DISABLED           DISABLED           
CON1PDB1        PDB        SUNDAY_WINDOW       DISABLED           DISABLED           
CON1PDB2        PDB        SUNDAY_WINDOW       DISABLED           DISABLED           
CON1PDB1        PDB        THURSDAY_WINDOW     DISABLED           DISABLED           
CON1PDB2        PDB        THURSDAY_WINDOW     DISABLED           DISABLED           
CON1PDB1        PDB        TUESDAY_WINDOW      DISABLED           DISABLED           
CON1PDB2        PDB        TUESDAY_WINDOW      DISABLED           DISABLED           
CON1PDB1        PDB        WEDNESDAY_WINDOW    DISABLED           DISABLED           
CON1PDB2        PDB        WEDNESDAY_WINDOW    DISABLED           DISABLED           

21 rows selected.

PL/SQL block to enable the auto optimizer stats job for each of the maintenance window

Also, enable the AUTO_TASKS window

set serveroutput on;
declare
        type all_windows_name is table of varchar2(30);
        all_windows all_windows_name := all_windows_name('SUNDAY_WINDOW','MONDAY_WINDOW','TUESDAY_WINDOW','WEDNESDAY_WINDOW','THURSDAY_WINDOW','FRIDAY_WINDOW','SATURDAY_WINDOW');
begin
        for wname in all_windows.first .. all_windows.last
        loop
                dbms_auto_task_admin.enable(
                        client_name     => 'auto optimizer stats collection',
                        operation       => 'auto optimizer stats job',
                        window_name     =>  all_windows(wname)
                );
                dbms_output.put_line('Enable stats job for: ' || all_windows(wname));
        end loop;
        dbms_auto_task_admin.enable(client_name => 'auto optimizer stats collection',operation   => NULL,window_name => NULL);
        dbms_auto_task_admin.enable;

end;
/

Used the above PL/SQL block to execute it in all the PDBs,

 perl /u02/app/oracle/product/19.0.0.0/dbhome_2/rdbms/admin/catcon.pl -b autostatsupdate -l /tmp /tmp/autostats.sql

Check the configured window and the stats collection job

      DATABASE    TYPE_DB         WINDOW_NAME    AUTOTASK_STATUS    OPTIMIZER_STATS
_______________ __________ ___________________ __________________ __________________
CON1            CDB        FRIDAY_WINDOW       ENABLED            ENABLED
CON1            CDB        MONDAY_WINDOW       ENABLED            ENABLED
CON1            CDB        SATURDAY_WINDOW     ENABLED            ENABLED
CON1            CDB        SUNDAY_WINDOW       ENABLED            ENABLED
CON1            CDB        THURSDAY_WINDOW     ENABLED            ENABLED
CON1            CDB        TUESDAY_WINDOW      ENABLED            ENABLED
CON1            CDB        WEDNESDAY_WINDOW    ENABLED            ENABLED
CON1PDB1        PDB        FRIDAY_WINDOW       ENABLED            ENABLED
CON1PDB2        PDB        FRIDAY_WINDOW       ENABLED            ENABLED
CON1PDB1        PDB        MONDAY_WINDOW       ENABLED            ENABLED
CON1PDB2        PDB        MONDAY_WINDOW       ENABLED            ENABLED
CON1PDB1        PDB        SATURDAY_WINDOW     ENABLED            ENABLED
CON1PDB2        PDB        SATURDAY_WINDOW     ENABLED            ENABLED
CON1PDB1        PDB        SUNDAY_WINDOW       ENABLED            ENABLED
CON1PDB2        PDB        SUNDAY_WINDOW       ENABLED            ENABLED
CON1PDB1        PDB        THURSDAY_WINDOW     ENABLED            ENABLED
CON1PDB2        PDB        THURSDAY_WINDOW     ENABLED            ENABLED
CON1PDB1        PDB        TUESDAY_WINDOW      ENABLED            ENABLED
CON1PDB2        PDB        TUESDAY_WINDOW      ENABLED            ENABLED
CON1PDB1        PDB        WEDNESDAY_WINDOW    ENABLED            ENABLED
CON1PDB2        PDB        WEDNESDAY_WINDOW    ENABLED            ENABLED

21 rows selected.

Check the current state of jobs in DBA_AUTOTASK_CLIENT

       DATABASE    TYPE_DB                        CLIENT_NAME     STATUS                                                ATTRIBUTES
_______________ __________ __________________________________ __________ _________________________________________________________
CON1            CDB        auto optimizer stats collection    ENABLED    ON BY DEFAULT, VOLATILE, SAFE TO KILL
CON1            CDB        auto space advisor                 ENABLED    ON BY DEFAULT, VOLATILE, SAFE TO KILL
CON1            CDB        sql tuning advisor                 ENABLED    ONCE PER WINDOW, ON BY DEFAULT, VOLATILE, SAFE TO KILL
CON1PDB1        PDB        auto optimizer stats collection    ENABLED    ON BY DEFAULT, VOLATILE, SAFE TO KILL
CON1PDB2        PDB        auto optimizer stats collection    ENABLED    ON BY DEFAULT, VOLATILE, SAFE TO KILL
CON1PDB1        PDB        auto space advisor                 DISABLED   ON BY DEFAULT, VOLATILE, SAFE TO KILL
CON1PDB2        PDB        auto space advisor                 DISABLED   ON BY DEFAULT, VOLATILE, SAFE TO KILL
CON1PDB2        PDB        sql tuning advisor                 DISABLED   ONCE PER WINDOW, ON BY DEFAULT, VOLATILE, SAFE TO KILL
CON1PDB1        PDB        sql tuning advisor                 DISABLED   ONCE PER WINDOW, ON BY DEFAULT, VOLATILE, SAFE TO KILL

9 rows selected.

Other relevant queries

select * from dba_autotask_client;
select * from dba_autotask_window_clients;
select * from dba_scheduler_windows;
select * from dba_autotask_schedule order by start_time;
-- Disable Auto Task
exec dbms_auto_task_admin.disable;
-- Disable auto stats collection job
exec dbms_auto_task_admin.disable(client_name=>'auto optimizer stats collection',operation=>NULL,window_name=>NULL);
-- To change timing of the auto stats collection job
exec dbms_scheduler.set_attribute(name=>'SUNDAY_WINDOW',attribute=>'repeat_interval',value=>'freq=daily;byday=MON;byhour=5;byminute=0;bysecond=0');
-- To change duration of the maintenance window
exec dbms_scheduler.set_attribute(name=>'SUNDAY_WINDOW',attribute=>'duration',value=> numtodsinterval(12,'hour'));

-- Create new or additional maintenance window
select * from dba_scheduler_groups;
select * from dba_scheduler_group_members;
exec dbms_scheduler.create_window(window_name=>'EXTRA_WINDOW',duration=>numtodsinterval(4,'hour'),resource_plan=>'DEFAULT_MAINTENANCE_PLAN',repeat_interval=>'FREQ=DAILY;BYHOUR=23;BYMINUTE=0;BYSECOND=0');
exec dbms_scheduler.add_group_member(group_name=>'MAINTENANCE_WINDOW_GROUP',member=>'EXTRA_WINDOW');
-- Drop maintenance window
exec dbms_scheduler.drop_window(window_name=>'EXTRA_WINDOW');
exec dbms_scheduler.remove_group_member(group_name=>'MAINTENANCE_WINDOW_GROUP',member=>'EXTRA_WINDOW');

OEM : migrate monitoring templates

Prasenjit Mahato — Thu, 22 Oct 2026 10:03:52 +0000

Query to find the monitoring templates currently used on old server

SELECT TEMPLATE_NAME,TARGET_TYPE FROM SYSMAN.MGMT_TEMPLATES;

Old Server : OEM 13.5.0.24

[oracle@oldserver1 ~]$ $OMS_HOME/bin/emcli login -username=sysman
Enter password

Login successful
[oracle@oldserver1 ~]$
[oracle@oldserver1 ~]$ $OMS_HOME/bin/emcli export_template -name="att_oracle_instance" -target_type=oracle_database -output_file=/home/oracle/att_oracle_instance.template.xml
File(s) exported successfully.
[oracle@oldserver1 ~]$ $OMS_HOME/bin/emcli export_template -name="att_rac_database" -target_type=rac_database -output_file=/home/oracle/att_rac_database.template.xml
File(s) exported successfully.
[oracle@oldserver1 ~]$

Copy the xml file /home/oracle/att_oracle_instance.template.xml to the new server.

On new server, remove the entries for the metric extension.

New Server : 24.1.0.5

[oracle@newserver1 ~]$ $OMS_HOME/bin/emcli login -username=sysman
Enter password

Login successful
[oracle@newserver1 ~]$ $OMS_HOME/bin/emcli import_template -files=att_oracle_instance.template.xml
File(s) imported successfully.
[oracle@newserver1 ~]$ $OMS_HOME/bin/emcli import_template -files=att_rac_database.template.xml
File(s) imported successfully.

Removing the entries of the metric extension allows to create the monitoring template without importing ME

[oracle@newserver1  ~]$ $OMS_HOME/bin/emcli import_template -files=att_oracle_instance.template.xml
This Monitoring Template Export file att_oracle_instance.template.xml contains information related to the following Metric Extension(s) : ME$ForceLoggingNoRac(Version1), ME$check_database_backup(Version4) , ME$fra_space_norac(Version2) , ME$manager_max_job_norac(Version1) , ME$memory_resize_errors(Version1) , ME$streams_apply_errors(Version1) , ME$stuck_autotask_windows(Version2) , which do not exist in the repository or are not in Published status. Proceed with this operation only after these Metric Extension(s) are already imported into EM and Published.

Enable RSA key-pair based authentication - GoldenGate to Snowflake (On Premise)

Prasenjit Mahato — Thu, 22 Oct 2026 09:51:36 +0000

With Snowflake enforcing the deprecation of legacy password-based authentication, all data integration pipelines must transition to Key-Pair authentication.

Prerequisites & Architecture

OGG User: Standard OS account (typically oracle) with access to the GoldenGate installation directory.

Snowflake User: Dedicated service user (e.g., GGDEV) with appropriate roles, warehouse access and permissions (SECURITYADMIN or ACCOUNTADMIN access required to assign keys).

Dependencies: Snowflake JDBC driver and Bouncy Castle FIPS cryptographic libraries (required for decrypting PKCS#8 encrypted private keys).

1. Generate an encrypted RSA key pair
Run the following commands on the GoldenGate server as the oracle user to generate a 2048-bit encrypted private key and its public key:

# Private key - prompted for passphrase twice

[oracle@sdevcdc01 ogg_keys]$ openssl genrsa 2048 | openssl pkcs8 -topk8 -v2 des3 -inform PEM -out rsa_GGDEV.p8
Generating RSA private key, 2048 bit long modulus (2 primes)
......................................................................................+++++
.......................................................................................................+++++
e is 65537 (0x010001)
Enter Encryption Password:
Verifying - Enter Encryption Password:
You have mail in /var/spool/mail/oracle
[oracle@sdevcdc01 ogg_keys]$



# Generate public key from encrypted private key - promted for the passphrase once

[oracle@sdevcdc01 ogg_keys]$ openssl rsa -in rsa_GGDEV.p8 -pubout -out rsa_GGDEV.pub
Enter pass phrase for rsa_GGDEV.p8:
writing RSA key
[oracle@sdevcdc01 ogg_keys]$

chmod 600 /u01/app/oracle/ogg_keys/rsa_GGDEV.p8

Note : Ensure you note your encryption passphrase. Supported special characters tested with OGG (These are tested) ! @ - _ . ? ; * ,

2. Extract the public key in a single line
Snowflake requires the raw base64 string without headers, footers, or newlines:

grep -v "PUBLIC KEY" /u01/app/oracle/ogg_keys/rsa_GGDEV.pub | tr -d '\n'

3. Assign the public key for the user of Snowflake
Log in to the Snowflake Web UI or SnowSQL as an administrator and assign the key to GoldenGate user:

ALTER USER GGDEV SET RSA_PUBLIC_KEY='MIIBkjrekqhdtwe46874-3A.......';

-- Verify the key fingerprint matches
DESCRIBE USER GGUSER;

(Do not include the line containing -----BEGIN PUBLIC KEY----- or -----END PUBLIC KEY-----)

Validate public key was correctly set, run the command describe user GGDEV and note the value of RSA_PUBLIC_KEY. It would match the value which was set earlier for the user.

4. Download FIPS & Bouncy Castle (BC) 2.x dependencies

The script snowflake-fips.sh can be used for the purpose. It can be found under the directory (OGGDAA home) . It might download older version of BC jar which will not work. Raised SR with Oracle support to get this sorted. The replicat requires 2.x version. Following jar files are needed to make it operational. Without the bc*.jar files, it would not be possible to connect to Snowflake. The decrypt mechanism just fails to decrypt the key without the correct version of bouncycastle.

bcutil-fips-2.0.7.jar
bc-fips-2.0.1.jar
bcpkix-fips-2.0.8.jar
snowflake-jdbc-fips-3.28.0.jar

GoldenGate 23ai running the Snowflake FIPS driver requires the Bouncy Castle 2.x FIPS spec (bc-fips-2.0.x), whereas earlier automated scripts pulled 1.x dependencies, causing silent decryption errors (NoSuchAlgorithmException or PKCS8KeySpec parsing errors).
This part is not clearly mentioned in the documentation present on the website of Oracle.

[oracle@sdevcdc01 DependencyDownloader]$ ./snowflake-fips-custom.sh 3.28.0
openjdk version "21.0.11" 2026-04-21 LTS
Java is installed.
Apache Maven 3.9.11 (3e54c93a704957b63ee3494413a2b544fd3d825b)
Maven is accessible.
Root Configuration Script
INFO: This is the Maven binary [../../ggjava/maven-3.9.11/bin/mvn].
INFO: This is the location of the settings.xml file [./docs/settings_np.xml].
INFO: This is the location of the toolchains.xml file [./docs/toolchains.xml].
INFO: The dependencies will be written to the following directory[/u01/app/ogg/ogg23ai_daa_ma/opt/DependencyDownloader/dependencies/snowflake-jdbc-3.28.0].
INFO: The Maven coordinates are the following:
INFO: Group ID [net.snowflake].s
INFO: Artifact ID [snowflake-jdbc-fips].
INFO: Version [3.28.0]
INFO: bc-fips version [2.0.1]
INFO: bcpkix-fips version [2.0.8]
INFO: bcutil-fips version [2.0.7]
[INFO] Scanning for projects...
[INFO]
[INFO] ---------------< oracle.goldengate:dependencyDownloader >---------------
[INFO] Building dependencyDownloader 1.0
[INFO]   from snowflake-fips-custom.ZsJCgD.xml
[INFO] --------------------------------[ pom ]---------------------------------
Downloading from central: https://repo.maven.apache.org/maven2/org/bouncycastle/bc-fips/2.0.1/bc-fips-2.0.1.pom
Downloaded from central: https://repo.maven.apache.org/maven2/org/bouncycastle/bc-fips/2.0.1/bc-fips-2.0.1.pom (1.3 kB at 3.2 kB/s)
Downloading from central: https://repo.maven.apache.org/maven2/org/bouncycastle/bcpkix-fips/2.0.8/bcpkix-fips-2.0.8.pom
Downloaded from central: https://repo.maven.apache.org/maven2/org/bouncycastle/bcpkix-fips/2.0.8/bcpkix-fips-2.0.8.pom (1.8 kB at 37 kB/s)
Downloading from central: https://repo.maven.apache.org/maven2/org/bouncycastle/bc-fips/maven-metadata.xml
Downloaded from central: https://repo.maven.apache.org/maven2/org/bouncycastle/bc-fips/maven-metadata.xml (807 B at 22 kB/s)
Downloading from central: https://repo.maven.apache.org/maven2/org/bouncycastle/bc-fips/2.0.0/bc-fips-2.0.0.pom
Downloaded from central: https://repo.maven.apache.org/maven2/org/bouncycastle/bc-fips/2.0.0/bc-fips-2.0.0.pom (1.3 kB at 24 kB/s)
Downloading from central: https://repo.maven.apache.org/maven2/org/bouncycastle/bc-fips/2.0.2/bc-fips-2.0.2.pom
Downloaded from central: https://repo.maven.apache.org/maven2/org/bouncycastle/bc-fips/2.0.2/bc-fips-2.0.2.pom (1.3 kB at 23 kB/s)
Downloading from central: https://repo.maven.apache.org/maven2/org/bouncycastle/bcutil-fips/maven-metadata.xml
Downloaded from central: https://repo.maven.apache.org/maven2/org/bouncycastle/bcutil-fips/maven-metadata.xml (582 B at 15 kB/s)
Downloading from central: https://repo.maven.apache.org/maven2/org/bouncycastle/bcutil-fips/2.0.2/bcutil-fips-2.0.2.pom
Downloaded from central: https://repo.maven.apache.org/maven2/org/bouncycastle/bcutil-fips/2.0.2/bcutil-fips-2.0.2.pom (1.6 kB at 27 kB/s)
Downloading from central: https://repo.maven.apache.org/maven2/org/bouncycastle/bcutil-fips/2.0.3/bcutil-fips-2.0.3.pom
Downloaded from central: https://repo.maven.apache.org/maven2/org/bouncycastle/bcutil-fips/2.0.3/bcutil-fips-2.0.3.pom (1.6 kB at 34 kB/s)
Downloading from central: https://repo.maven.apache.org/maven2/org/bouncycastle/bcutil-fips/2.0.5/bcutil-fips-2.0.5.pom
Downloaded from central: https://repo.maven.apache.org/maven2/org/bouncycastle/bcutil-fips/2.0.5/bcutil-fips-2.0.5.pom (1.6 kB at 28 kB/s)
Downloading from central: https://repo.maven.apache.org/maven2/org/bouncycastle/bcutil-fips/2.0.6/bcutil-fips-2.0.6.pom
Downloaded from central: https://repo.maven.apache.org/maven2/org/bouncycastle/bcutil-fips/2.0.6/bcutil-fips-2.0.6.pom (1.6 kB at 27 kB/s)
Downloading from central: https://repo.maven.apache.org/maven2/org/bouncycastle/bcutil-fips/2.0.7/bcutil-fips-2.0.7.pom
Downloaded from central: https://repo.maven.apache.org/maven2/org/bouncycastle/bcutil-fips/2.0.7/bcutil-fips-2.0.7.pom (1.6 kB at 30 kB/s)
Downloading from central: https://repo.maven.apache.org/maven2/org/bouncycastle/bc-fips/2.0.1/bc-fips-2.0.1.jar
Downloaded from central: https://repo.maven.apache.org/maven2/org/bouncycastle/bc-fips/2.0.1/bc-fips-2.0.1.jar (4.1 MB at 21 MB/s)
Downloading from central: https://repo.maven.apache.org/maven2/org/bouncycastle/bcpkix-fips/2.0.8/bcpkix-fips-2.0.8.jar
Downloading from central: https://repo.maven.apache.org/maven2/org/bouncycastle/bcutil-fips/2.0.7/bcutil-fips-2.0.7.jar
Downloaded from central: https://repo.maven.apache.org/maven2/org/bouncycastle/bcutil-fips/2.0.7/bcutil-fips-2.0.7.jar (718 kB at 9.4 MB/s)
Downloaded from central: https://repo.maven.apache.org/maven2/org/bouncycastle/bcpkix-fips/2.0.8/bcpkix-fips-2.0.8.jar (1.0 MB at 10 MB/s)
[INFO]
[INFO] --- clean:3.2.0:clean (default-clean) @ dependencyDownloader ---
[INFO] Deleting /u01/app/ogg/ogg23ai_daa_ma/opt/DependencyDownloader/dependencies/snowflake-jdbc-3.28.0
[INFO]
[INFO] --- dependency:2.9:copy-dependencies (copy-dependencies) @ dependencyDownloader ---
[INFO] Copying bcutil-fips-2.0.7.jar to /u01/app/ogg/ogg23ai_daa_ma/opt/DependencyDownloader/dependencies/snowflake-jdbc-3.28.0/bcutil-fips-2.0.7.jar
[INFO] Copying bc-fips-2.0.1.jar to /u01/app/ogg/ogg23ai_daa_ma/opt/DependencyDownloader/dependencies/snowflake-jdbc-3.28.0/bc-fips-2.0.1.jar
[INFO] Copying bcpkix-fips-2.0.8.jar to /u01/app/ogg/ogg23ai_daa_ma/opt/DependencyDownloader/dependencies/snowflake-jdbc-3.28.0/bcpkix-fips-2.0.8.jar
[INFO] Copying snowflake-jdbc-fips-3.28.0.jar to /u01/app/ogg/ogg23ai_daa_ma/opt/DependencyDownloader/dependencies/snowflake-jdbc-3.28.0/snowflake-jdbc-fips-3.28.0.jar
[INFO] ------------------------------------------------------------------------
[INFO] BUILD SUCCESS
[INFO] ------------------------------------------------------------------------
[INFO] Total time:  2.607 s
[INFO] Finished at: 2026-09-04T11:33:27+02:00
[INFO] ------------------------------------------------------------------------
[oracle@sdevcdc01 DependencyDownloader]$

5. Store the private key file information and the password of the snowflake user to the credential store(Wallet) of Golden Gate. Default domain is OracleGoldenGate

# Add private key path alias
alter credentialstore add user GGDEV ALIAS ggDEVWRK domain OracleGoldenGate password /u01/app/oracle/ogg_keys/rsa_GGDEV.p8

# Add private key passphrase alias
alter credentialstore add user GGDEV ALIAS ggDEVPASS domain OracleGoldenGate password "<YOUR_KEY_PASSPHRASE>"

6. Configure the properties file of the Replicat file

Remove legacy password authentication properties (gg.eventhandler.snowflake.UserName / gg.eventhandler.snowflake.Password)

Accomodate username and private key file and it's passphrase in the gg.eventhandler.snowflake.connectionURL

Also adapt the classpath to include the BC jar files.

Sample .properties file of the replicat

gg.target=snowflake

# Snowflake Connection URL using OGG Wallet aliases
gg.eventhandler.snowflake.connectionURL=jdbc:snowflake://<account_identifier>.snowflakecomputing.com/?warehouse=<WH>&db=<db>&user=GGDEV&role=DEV_ROLE&private_key_file=ORACLEWALLETPASSWORD[ggDEVWRK OracleGoldenGate]&private_key_pwd=ORACLEWALLETPASSWORD[ggDEVPASS OracleGoldenGate]&CLIENT_SESSION_KEEP_ALIVE=true&SESSION_IDLE_TIMEOUT_MINS=1440&CLIENT_SESSION_KEEP_ALIVE_HEARTBEAT_FREQUENCY=900&disablePlatformDetection=true&disableOCSPChecks=true

# Classpath pointing to Snowflake JDBC & BC-FIPS 2.x dependencies
gg.classpath=/u01/app/ogg/ogg23ai_daa_ma/opt/DependencyDownloader/dependencies/snowflake-jdbc-3.28.0/snowflake-jdbc-fips-3.28.0.jar:/u01/app/ogg/ogg23ai_daa_ma/opt/DependencyDownloader/dependencies/snowflake-jdbc-3.28.0/bcutil-fips-2.0.7.jar:/u01/app/ogg/ogg23ai_daa_ma/opt/DependencyDownloader/dependencies/snowflake-jdbc-3.28.0/bc-fips-2.0.1.jar:/u01/app/ogg/ogg23ai_daa_ma/opt/DependencyDownloader/dependencies/snowflake-jdbc-3.28.0/bcpkix-fips-2.0.8.jar

jvm.bootoptions=-Xmx8g -Xms8g
gg.eventhandler.snowflake.maxConnections=200
gg.maxInlineLobSize=128000

7. Start the replicat
Start the process from the Admin Client or GUI:

start replicat RDEVWRK
view report RDEVWRK

Verify that the process status transitions to RUNNING without cryptographic or connection errors.

Source :
https://docs.oracle.com/en/database/goldengate/big-data/26/gadbd/snowflake1.html#GUID-AFEF7C23-B2FE-4CCB-A7BB-67E1E9038F5F