Zero Downtime Migration Release Notes
These release notes provide downloading instructions for the latest product software and documentation, and describe new features, fixed bugs, known issues, and troubleshooting information for Zero Downtime Migration Release 21c (21.2).
- What's New in This Release
- Bugs Fixed
- Downloading the Zero Downtime Migration Installation Software
- Downloading the Zero Downtime Migration Documentation
- General Information
- Known Issues
- Troubleshooting
- Additional Information for Migrating to Exadata Cloud Service
- Additional Information for Migrating to Exadata Cloud at Customer
- Documentation Accessibility
What's New in This Release
Zero Downtime Migration Release 21c (21.2) includes the following new features.
-
Physical Migration Supports Direct Data Transfer
You can migrate databases using direct data transfer during a physical migration to avoid backing up the source database to an intermediate store such as Object Storage or NFS. Recovery Manager (RMAN) allows Zero Downtime Migration to support active database duplication and restore from service.
-
Reuse an Existing RMAN Backup as a Migration Source
You can use an existing level 0 backup to skip the full backup phase of a migration job. Zero Downtime Migration takes level 0 and level 1 backups on the fly for both online and offline physical migration jobs. Now Zero Downtime Migration lets you re-use existing source database backup in place of performing a full back up.
-
CPAT Lists Remedies and Can Create Remedy Scripts
The Cloud Premigration Advisor Tool (CPAT) is integrated with Zero Downtime Migration, and can be used with logical migration jobs. CPAT can generate migration remedial scripts for failing checks that you can run if you choose.
-
Run CPAT Updates in Zero Downtime Migration Home
Keep the Cloud Premigration Advisor Tool (CPAT) up to date to get the latest tool version.
-
Migrate From Amazon Web Services RDS to Oracle Autonomous Database
You can now migrate an Oracle Database from Amazon Web Services (AWS) RDS to Oracle Autonomous Database (ADB) using the Zero Downtime Migration offline logical migration method. To transfer the data from AWS, you can use an Amazon Simple Storage Service (Amazon S3) Bucket or a database link (DBLINK).
Migrating from Amazon Web Services RDS to Oracle Autonomous Database
-
Migrate to Oracle Autonomous Database on Exadata Cloud@Customer
Zero Downtime Migration supports migrations to Oracle Autonomous Database on Exadata Cloud@Customer from any on-premises Oracle Database, including existing Exadata Cloud@Customer systems, using the offline logical migration method and NFS as a data transfer medium.
Migrating to Oracle Autonomous Database on Exadata Cloud@Customer
-
Platform Support Includes IBM AIX and Oracle Solaris
In addition to Linux-x86-64, Zero Downtime Migration now supports IBM AIX and Oracle Solaris source environments for logical offline migration mode.
-
Estimate Data Pump Dump Size with STATISTICS Method
In logical migrations, Zero Downtime Migration estimates the Data Pump dump size using the BLOCKS method by default as part of the pre-check and as part of the actual migration part of phase
ZDM_DATAPUMP_ESTIMATE_SRC
. A new response file parameter,DATAPUMPSETTINGS_DATAPUMPPARAMETERS_ESTIMATEBYSTATISTICS
, lets you choose the STATISTICS method to provide you with an estimated dump size. -
Automatically Create or Remap Tablespaces
For logical migrations, Zero Downtime Migration can automatically discover the source database tablespaces associated with user schemas that are being migrated, and automatically create them in the target database before the Data Pump import phase.
An important benefit is that you can create BIGFILE tablespaces on the target which significantly reduces the number of data files for your database. You can also enable AUTOEXTEND and specify the NEXT EXTENT size, and you can exclude specified tablespaces from being created automatically.
In addition, with automatic remap enabled, Zero Downtime Migration discovers any source tablespaces that must be remapped, instead of pre-creating the tablespace on the target.
Automatic Tablespace Creation and Automatic Tablespace Remap
-
Migration Customization Enhancements
You can edit a user action script locally and update the script in the user action metadata. If you do not have a local copy of the script, you can now locate the cached user action script in the user action metadata using the ZDMCLI command
query useraction
, and the full base path to the script is displayed in the output.You can now easily get data from one or more user actions and use that data to run another user action. For example, you can perform an action on the source database and use that data to perform an action on the target database.
Bugs Fixed
Zero Downtime Migration Release 21.2 incudes the bug fixes listed in the following table.
Table - Bugs Fixed In Zero Downtime Migration Release 21.2
Bug Number | Description |
---|---|
32749606 | ZDM PHYSICAL MIGRATION INTERMITTENTLY FAILS WITH PRGZ-3553 CHARACTER SET MISMATCH |
33013588 | ZDM: CONVERT NONCDB TO CDB FAILED - PRGO-1930 & PRCZ-2103 |
32808775 | ZDM PHYSICAL MIGRATION WITH NONCDB TO PDB CONVERSION FOR SRC AS SINGLE INSTANCE DB ISSUES |
33123365 | DB_DOMAIN OVERRIDEN BY SOURCE SPFILE CONTENT |
32767162 | ZDM: MIGRATION FROM NON CDB TO CDB IS FAILING, WHEN TARGET IS VMDB AND BACKUP MEDIUM IS NFS |
31507335 | ZDM: OFFLINE LOGICAL MIGRATION - ABORT SHOULD CLEAN THE DBLINK CREATED |
32100209 | ZDM : EXPORT PHASE FAILS WHEN THE EXPORT DIRECTORY IS IN ASM |
32960299 | ZDM SVC PREFERRED INSTANCES ATTRIBUTE IS NOT GETTING MAPPED CORRECTLY |
32447725 | ZDM : ABORTED ZDM JOB NOT ABORTING EXPORT/IMPORT JOBS |
32833720 | LOGICAL MIGRATION WORKFLOW FAILING WITH STANDARD EDITION SOURCE DATABASES |
33048353 | ZDM HIT ERROR: ASMCMD-9459 USER TRIED TO MOVE OR COPY A PASSWORD FILE WITHIN THE SAME DISK GROUP |
33032852 | ZDM: ZDM_UTILS REPORTED COREDUMP ERRORS DUE TO CREDENTIAL DOES NOT EXIST |
32464487 | ZDM: OSS FEDERATED USER ACCOUNT FAILING ZDM_VALIDATE_SRC |
32689878 | UNABLE TO START ZDM SERVICE NODE AFTER UPGRADE FROM 19.7 TO 21.1 -> CRS_ERROR:TCC-0004: THE CONTAINER WAS NOT ABLE TO START. |
32886747 | LOGICAL MIGRATION: ZDM DOESN'T DETECT LONG RUNNING TRANSACTION DURING SWITCHOVER PHASE |
Downloading the Zero Downtime Migration Installation Software
For a fresh installation of the latest Zero Downtime Migration software version, go to https://www.oracle.com/database/technologies/rac/zdm-downloads.html.
Downloading the Zero Downtime Migration Documentation
General Information
At the time of this release, there are some details and considerations about Zero Downtime Migration behavior that you should take note of.
Cloud Premigration Advisor Tool Support
Cloud Premigration Advisor Tool (CPAT) is supported with Zero Downtime Migration for the following use cases:
-
CPAT is run automatically on the source database environment during logical migration jobs from Oracle Cloud and on-premises Oracle Database sources (default behavior)
-
CPAT is run manually from a remote server against an Amazon Web Services RDS Oracle Database source; in other words, CPAT is not run by
ZDMCLI migrate database
(see Running CPAT with a Remote Connection)
CPAT is not supported in the following use cases:
-
Physical migration jobs
-
Generating fixup scripts for Amazon Web Services RDS Oracle Database sources
UNDO Tablespaces Added to the Source Database
Zero Downtime Migration adds UNDO tablespaces to the production database to match the target instance count if the production database has fewer instances.
To prevent Zero Downtime Migration from adding UNDO tablespaces to the source database, you can match the target database nodes count to that of the source database until the switchover, then you can add additional nodes to the target database after the switchover.
Cross-Edition Migration
Zero Downtime Migration cannot be used to migrate an Enterprise Edition database to a Standard Edition database. In the converse case, Standard Edition databases can be migrated to Enterprise Edition databases, but only using the logical migration work flow.
EXT3 File System Support
There is a known issue which prevents Zero Downtime Migration from being installed in EXT3 file systems. The root cause is MySQL bug 102384. This is not a limitation of Zero Downtime Migration, but a constraint from MySQL. When that bug is resolved, Zero Downtime Migration is expected to work on EXT3 file systems.
Known Issues
At the time of this release, the following are known issues with Zero Downtime Migration that could occur in rare circumstances. For each issue, a workaround is provided.
Logical Migration Using DBLINK Fails with PRGZ-1177
Issue: "PRGZ-1177 : Database link "dblink_name" is invalid and unusable" error causes failure in a logical migration using a database link in a PDB or multitenant database in version 12.1.0.x.
Solution: See 12c PDB or Multitenant Only: ORA-02085: Database Link "LINK_NAME_HERE" Connects To "TARGET_DB" (Doc ID 2344831.1)
PRGZ-1306 Thrown During Physical Migration from Single Instance Database with No Grid Infrastructure Configured
Issue: When attempting to migrate a single instance database with no Grid
Infrastructure configured using the -sourcesid
parameter, the migration
job fails with error PRGZ-1306.
Solution: This is an issue that is being tracked with Bug 33308423. As a
workaround, re-run the migration command substituting the -sourcesid
option with the -sourcedb
option. Otherwise do not perform the physical
migration on a single instance database with no GI configured until the bug is fixed and
applied to the Zero Downtime Migration server home.
ORA-01031 on Full Export from an Oracle 12.1 Source
Issue: When performing a full database export with Export Data Pump from an Oracle Database 12c (12.1) source database, the following errors occur:
05-AUG-21 10:36:12.483: ORA-31693: Table data object
"SYS"."TABLE" failed to load/unload and is being skipped due
to error: ORA-01031: insufficient privileges
Solution: See My Oracle Support document EXPDP - ORA-31693 ORA-01031 (Insufficient Privileges) On Some Tables When Exporting from 12cR1 (Doc ID 1676411.1)
Restore Fails When Source Uses WALLET_ROOT
Issue: Zero Downtime Migration does not currently handle the migration of the TDE
wallet from the source database to the target when the source database is using the
wallet_root
initialization parameter. Without the wallets available
on the target database, the restore phase fails with an error similar to the
following:
RMAN-00571: ===========================================================
RMAN-00569: =============== ERROR MESSAGE STACK FOLLOWS ===============
RMAN-00571: ===========================================================
RMAN-03002: failure of restore command at 06/15/2021 07:35:11
ORA-19870: error while restoring backup piece
/rman_PRD1/ZDM/IQPCZDM/c-3999816841-20210614-00
ORA-19913: unable to decrypt backup
Solution: Manually copy the wallet to the target and resume the job.
ORA-17503 on Dump Import from Amazon S3 to Autonomous Database Dedicated Infrastructure
Issue: When attempting to import a Data Pump dump file from an Amazon S3 bucket to an Oracle Autonomous Database Dedicated Infrastructure (ADB-D) target, an error occurs in ADB-D.
The ADB-D alert log shows an error similar to the following:
ORA-39000: bad dump file specification
ORA-31640: unable to open dump file
"https://s3-region.amazonaws.com/dump_file_path.dmp
for read
ORA-17503: target Failed to open file
https://s3-region.amazonaws.com/dump_file_path.dmp
ORA-17500: ODM err:ODM Client Error
Solution: Use the database link data transfer medium in Zero Downtime Migration to import Data Pump dumps to an ADB-D target.
DATA_TRANSFER_MEDIUM=DBLINK
SOURCEDATABASE_ENVIRONMENT_NAME=AMAZON
SOURCEDATABASE_ENVIRONMENT_DBTYPE=RDS_ORACLE
PRCZ-4026 Thrown During Migration to Oracle Database 19.10 Target
Issue: When attempting to migrate to an Oracle Database 19.10 home at target, the
migration job fails at phase ZDM_FINALIZE_TGT
with error PRCZ-4026,
because of Oracle Clusterware (OCW) Bug 31070231.
PRCZ-4026 : Resource ora.db_unique_name.db is already running
on nodes node.
Solution: Apply the Backport Label Request (BLR) for Bug#32646135 to the target 19.10 dbhome to avoid the reported issue. Once the BLR is applied, you can resume the failed migration job to completion.
Precaution: For physical migrations, you can avoid this issue by ensuring that your target database home is not on Oracle Database 19.10.
ORA-39006 Thrown During Logical Migration to Autonomous Database Dedicated Infrastructure Over Database Link
Issue: When attempting to migrate a database to an Autonomous Database Dedicated Infrastructure target over a database link, the migration job fails with error ORA-39006.
ORA-39006: internal error
Solution: This is a Data Pump issue that is being tracked with Bug 31830685. Do not perform logical migrations over a database link to Autonomous Database Dedicated Infrastructure targets until the bug is fixed and the fix is applied to the Autonomous target database.
Zero Downtime Migration Service Fails To Start After Upgrade
Issue: The following scenario occurs:
-
Perform migration jobs with Zero Downtime Migration 19.7
-
Response files used in those jobs are removed
-
Upgrade to Zero Downtime Migration 21.1
-
Attempt to run a migration
The following errors are seen.
CRS_ERROR:TCC-0004: The container was not able to start.
CRS_ERROR:One or more listeners failed to start. Full details will be found in
the appropriate container log fileContext [/rhp] startup failed due to previous
errors sync_start failed with exit code 1.
A similar error is found in the log files located in
zdm_installation_location/base/crsdata/hostname/rhp/logs/
.
Caused by: oracle.gridhome.container.GHException: Internal
error:PRGO-3003 : Zero downtime migration (ZDM) template file
/home/jdoe/zdm_mydb.rsp does not exist.
Solution: To recover, manually recreate the response files listed in the log, and place them in the location specified in the log.
Environments With Oracle 11.2.0.4 Must Apply Perl Patch
Issue: Before using Zero Downtime Migration, you must apply a PERL patch if your source database environment meets either of the following conditions.
- Clusterware environment with Oracle Grid Infrastructure 11.2.0.4
- Single instance environment with Oracle Database 11.2.0.4
Solution: Download and apply Perl patch version 5.28.2 or later. Ensure that both the source and target Oracle Database 11g home include the patch for BUG 30508206 - UPDATE PERL IN 11.2.0.4 DATABASE ORACLE HOME TO V5.28.2.
Oracle GoldenGate Hub Certificate Known Issues
Issue: Oracle Zero Downtime Migration leverages Oracle GoldenGate for its logical online migration work flow; an Oracle GoldenGate hub is set up on OCI compute for this purpose.
The Oracle GoldenGate hub NginX Reverse Proxy uses a self-signed certificate which will cause the following error:
SunCertPathBuilderException: unable to find valid certification path to requested
target when ZDM Server makes a REST API call.
Solution: See My Oracle Support document Zero Downtime Migration - GoldenGate Hub Certificate Known Issues (Doc ID 2768483.1)
Data Transfer Medium COPY Issues
Issue: Migrating data using logical migration with
DATA_TRANSFER_MEDIUM=COPY
set in the Zero Downtime Migration
response file fails.
Solution: When you specify DATA_TRANSFER_MEDIUM=COPY
you must also specify the following DUMPTRANSFERDETAILS_SOURCE_*
parameters.
DUMPTRANSFERDETAILS_TRANSFERTARGET_DUMPDIRPATH=<Target path to transferthe dumps to >
DUMPTRANSFERDETAILS_TRANSFERTARGET_HOST=<Target Db server or Target sidetransfer node >
DUMPTRANSFERDETAILS_TRANSFERTARGET_USER=<user having write access to specified path>
DUMPTRANSFERDETAILS_TRANSFERTARGET_USERKEY=<user authentication keypath on zdm node>
Source Discovery Does Not Find 'cut' in Default Location
Issue: Discovery at the source database server fails to find
cut
in the standard location.
The source database deployment's standard cut
location is
/bin/cut
. If cut
is not in the location, Zero
Downtime Migration cannot discover the source database information correctly, and the
migration fails in its initial phases.
Solution: To resolve the issue, ensure that cut
is
installed in the standard /bin/cut
path or create a symbolic link to
the installed location, for example:
ln -sf <installed_location_of_the_cut> /bin/cut
PRGG-1043 : No heartbeat table entries were found for Oracle GoldenGate Replicat process
Issue: An online logical migration job can report error PRGG-1043: No heartbeat table entries were found for Oracle GoldenGate Replicat process process_name due to one of the following causes:
-
Initialization parameter
job_queue_processes
was set to zero in the source or target database.Solution: Run the following statements on the database:
show parameter job_queue_processes; alter system set job_queue_processes=100 scope=both; exec dbms_scheduler.set_scheduler_attribute('SCHEDULER_DISABLED','FALSE');
-
Scheduled job
GG_UPDATE_HEARTBEATS
is not active in the source database. -
The server hosting Oracle GoldenGate deployments has a different time zone than the source database.
Solution: First, do one of the following solutions:
-
Modify the time zone for the server hosting Oracle GoldenGate deployments, OR
-
Use the web UI for the Oracle GoldenGate deployment to add Extract parameter
TRANLOGOPTIONS SOURCE_OS_TIMEZONE
and restart Extract.For example, if the source database time zone is UTC-5, then set parameter
TRANLOGOPTIONS SOURCE_OS_TIMEZONE -5
. For more information, see TRANLOGOPTIONS in Reference for Oracle GoldenGate.
Then, ensure that the
DST_PRIMARY_TT_VERSION
property in the source database is up to date. -
Troubleshooting
If you run into issues, check here in case a solution is published. For each issue, a workaround is provided.
General Connectivity Issues
Issue: If connectivity issues occur between the Zero Downtime Migration service host and the source or target environments, or between source and target environments, check the following areas.
Solution: Verify that the SSH configuration file
(/root/.ssh/config
) has the appropriate entries:
Host *
ServerAliveInterval 10
ServerAliveCountMax 2
Host ocidb1
HostName 192.0.2.1
IdentityFile ~/.ssh/ocidb1.ppk
User opc
ProxyCommand /usr/bin/nc -X connect -x www-proxy.example.com:80 %h %p
Note that the proxy setup might not be required when you are not using a
proxy server for connectivity. For example, when the source database server is on Oracle
Cloud Infrastructure Classic, you can remove or comment the line starting with
ProxyCommand
.
If the source is an Oracle RAC database, then make sure you copy the
~/.ssh/config
file to all of the source Oracle RAC servers. The SSH
configuration file refers to the first Oracle RAC server host name, public IP address,
and private key attributes.
Communications Link Failure
Issue: If the MySQL server crashes you will see errors such as this one for the ZDM operations:
$ ./zdmcli query job -jobid 6
Exception [EclipseLink-4002] (Eclipse Persistence Services -
2.7.7.qualifier): org.eclipse.persistence.exceptions.DatabaseException
Internal Exception: com.mysql.cj.jdbc.exceptions.CommunicationsException:
Communications link failure
The last packet sent successfully to the server was 0 milliseconds ago. The
driver has not received any packets from the server.
Error Code: 0
Query: ReadAllQuery(referenceClass=JobSchedulerImpl sql="SELECT
JOB_IDENTIFIER, M_ACELIST, ARGUMENTS, ATTRIBUTES, CLIENT_NAME,
COMMAND_PROVIDED, COMPARTMENT, CONTAINER_TYPE, CREATEDATE, CREATOR,
CURRENT_STATUS, DB_OCID, DBNAME, DEPLOYMENT_OCID, DISABLE_JOB_EXECUTION,
ELAPSED_TIME, END_TIME, EXECUTE_PHASES, EXECUTION_TIME, IS_EVAL, IS_PAUSED,
JOB_TYPE, METHOD_NAME, METRICS_LOCATION, OPERATION, PARAMETERS,
PARENT_JOB_ID, PAUSE_AFTER_PHASE, RESULT, PHASE, JOB_SCHEDULER_PHASES,
REGION, REST_USER_NAME, RESULT_LOCATION, SCHEDULED_TIME, SITE, SOURCEDB,
SOURCENODE, SOURCESID, SPARE1, SPARE2, SPARE3, SPARE_A, SPARE_B, SPARE_C,
START_TIME, STOP_AFTER_PHASE, TARGETNODE, JOB_THREAD_ID, UPD_DATE, USER_NAME,
ENTITY_VERSION, CUSTOMER FROM JOBSCHEDULER WHERE (PARENT_JOB_ID = ?)")
Solution: If such Communications errors are seen, restart the Zero Downtime Migration service so that the MySQL server is restarted, after which the pending jobs will resume automatically.
Stop the Zero Downtime Migration service:
zdmuser> $ZDM_HOME/bin/zdmservice stop
Start the Zero Downtime Migration service:
zdmuser> $ZDM_HOME/bin/zdmservice start
Evaluation Fails in Phase ZDM_GET_TGT_INFO
Issue: During the evaluation (-eval
) phase of the migration
process, the evaluation fails in the ZDM_GET_TGT_INFO
phase with the
following error for the Oracle RAC instance migration.
Executing phase ZDM_GET_TGT_INFO
Retrieving information from target node "trac11" ...
PRGZ-3130 : failed to establish connection to target listener from nodes [srac11, srac12]
PRCC-1021 : One or more of the submitted commands did not execute successfully.
PRCC-1025 : Command submitted on node srac11 timed out after 15 seconds.
PRCC-1025 : Command submitted on node srac12 timed out after 15 seconds.
Solution:
- Get the SCAN name of source database and add it to the
/etc/hosts
file on both target database servers, with the public IP address of the source database server and the source database SCAN name. For example:192.0.2.3 source-scan
- Get the SCAN name of the target database and add it to the
/etc/hosts
file on both source database servers, with the public IP address of the target database server and target database SCAN name. For example:192.0.2.1 target-scan
Note:
This issue, where the SCAN IP address is not added to/etc/hosts
file, might occur because in some cases the
SCAN IP address is assigned as a private IP address, so it might not be
resolvable.
Object Storage Is Not Accessible
About to connect() to swiftobjectstorage.xx-region-1.oraclecloud.com port 443 (#0)
Trying 192.0.2.1... No route to host
Trying 192.0.2.2... No route to host
Trying 192.0.2.3... No route to host
couldn't connect to host
Closing connection #0
curl: (7) couldn't connect to host
Solution: On the Zero Downtime Migration service host, in the response file
template ($ZDM_HOME/rhp/zdm/template/zdm_template.rsp
), set the Object
Storage Service proxy host and port parameters listed below, if a proxy is required to
connect to Object Storage from the source database server. For example:
SRC_OSS_PROXY_HOST=www-proxy-source.example.com
SRC_OSS_PROXY_PORT=80
In the response file template
($ZDM_HOME/rhp/zdm/template/zdm_template.rsp
), set the Object
Storage Service proxy host and port parameters listed below, if a proxy is required to
connect to Object Storage from the target database server. For example:
TGT_OSS_PROXY_HOST=www-proxy-target.example.com
TGT_OSS_PROXY_PORT=80
SSH Error "EdDSA provider not supported"
Issue:
The following error messages appear in $ZDM_BASE/crsdata/zdm service hostname/rhp/zdmserver.log.0
.
[sshd-SshClient[3051eb49]-nio2-thread-1] [ 2020-04-04 00:26:24.142 GMT ]
[JSChChannel$LogOutputStream.flush:1520] 2020-04-04: WARNING: org.apache.sshd.client.session.C:
globalRequest(ClientConnectionService[ClientSessionImpl[opc@samidb-db/140.238.254.80:22]])[hostkeys-00@openssh.com,
want-reply=false] failed (SshException) to process: EdDSA provider not supported
[sshd-SshClient[3051eb49]-nio2-thread-1] [ 2020-04-04 00:26:24.142 GMT ]
[JSChChannel$LogOutputStream.flush:1520] 2020-04-04: FINE : org.apache.sshd.client.session.C:
globalRequest(ClientConnectionService[ClientSessionImpl[opc@samidb-db/140.238.254.80:22]])[hostkeys-00@openssh.com,
want-reply=false] failure details
org.apache.sshd.common.SshException: EdDSA provider not supported
at org.apache.sshd.common.util.buffer.Buffer.getRawPublicKey(Buffer.java:446)
at org.apache.sshd.common.util.buffer.Buffer.getPublicKey(Buffer.java:420)
at org.apache.sshd.common.global.AbstractOpenSshHostKeysHandler.process(AbstractOpenSshHostKeysHandler.java:71)
at org.apache.sshd.common.global.AbstractOpenSshHostKeysHandler.process(AbstractOpenSshHostKeysHandler.java:38)
at org.apache.sshd.common.session.helpers.AbstractConnectionService.globalRequest(AbstractConnectionService.java:723)
at org.apache.sshd.common.session.helpers.AbstractConnectionService.process(AbstractConnectionService.java:363)
at org.apache.sshd.common.session.helpers.AbstractSession.doHandleMessage(AbstractSession.java:400)
at org.apache.sshd.common.session.helpers.AbstractSession.handleMessage(AbstractSession.java:333)
at org.apache.sshd.common.session.helpers.AbstractSession.decode(AbstractSession.java:1097)
at org.apache.sshd.common.session.helpers.AbstractSession.messageReceived(AbstractSession.java:294)
at org.apache.sshd.common.session.helpers.AbstractSessionIoHandler.messageReceived(AbstractSessionIoHandler.java:63)
at org.apache.sshd.common.io.nio2.Nio2Session.handleReadCycleCompletion(Nio2Session.java:357)
at org.apache.sshd.common.io.nio2.Nio2Session$1.onCompleted(Nio2Session.java:335)
at org.apache.sshd.common.io.nio2.Nio2Session$1.onCompleted(Nio2Session.java:332)
at org.apache.sshd.common.io.nio2.Nio2CompletionHandler.lambda$completed$0(Nio2CompletionHandler.java:38)
at java.security.AccessController.doPrivileged(Native Method)
at org.apache.sshd.common.io.nio2.Nio2CompletionHandler.completed(Nio2CompletionHandler.java:37)
at sun.nio.ch.Invoker.invokeUnchecked(Invoker.java:126)
at sun.nio.ch.Invoker$2.run(Invoker.java:218)
at sun.nio.ch.AsynchronousChannelGroupImpl$1.run(AsynchronousChannelGroupImpl.java:112)
at java.util.concurrent.ThreadPoolExecutor.runWorker(ThreadPoolExecutor.java:1149)
at java.util.concurrent.ThreadPoolExecutor$Worker.run(ThreadPoolExecutor.java:624)
at java.lang.Thread.run(Thread.java:748)
Caused by: java.security.NoSuchAlgorithmException: EdDSA provider not supported
at org.apache.sshd.common.util.security.SecurityUtils.generateEDDSAPublicKey(SecurityUtils.java:596)
at org.apache.sshd.common.util.buffer.keys.ED25519BufferPublicKeyParser.getRawPublicKey(ED25519BufferPublicKeyParser.java:45)
at org.apache.sshd.common.util.buffer.keys.BufferPublicKeyParser$2.getRawPublicKey(BufferPublicKeyParser.java:98)
at org.apache.sshd.common.util.buffer.Buffer.getRawPublicKey(Buffer.java:444)
... 22 more
[sshd-SshClient[3051eb49]-nio2-thread-1] [ 2020-04-04 00:26:24.142 GMT ]
[JSChChannel$LogOutputStream.flush:1520] 2020-04-04: FINE : org.apache.sshd.client.session.C:
sendGlobalResponse(ClientConnectionService[ClientSessionImpl[opc@samidb-db/140.238.254.80:22]])[hostkeys-00@openssh.com]
result=ReplyFailure, want-reply=false
[sshd-SshClient[3051eb49]-nio2-thread-2] [ 2020-04-04 00:26:24.182 GMT ]
[JSChChannel$LogOutputStream.flush:1520] 2020-04-04: FINE : org.apache.sshd.common.io.nio2.N:
handleReadCycleCompletion(Nio2Session[local=/192.168.0.2:41198, remote=samidb-db/140.238.254.80:22])
read 52 bytes
Solution: Zero Downtime Migration uses the RSA format.
Backup Fails with ORA-19836
Issue: Source database full backup fails with one of the following errors.
</ERRLINE><ERRLINE>ORA-19836: cannot use passphrase encryption for this backup
</ERRLINE><ERRLINE>RMAN-03009: failure of backup command on C8 channel at 04/29/2019
20:42:16
</ERRLINE><ERRLINE>ORA-19836: cannot use passphrase encryption for this backup
</ERRLINE><ERRLINE>RMAN-03009: continuing other job steps, job failed will not be
re-run
Solution 1: This issue can occur if you specify the
-sourcedb
value in the wrong case. For example, if the value
obtained from SQL command SHOW PARAMETER DB_UNIQUE_NAME
is
zdmsdb
, then you need to specify it as zdmsdb
in
lower case, and not as ZDMSDB
in upper case, as shown in the following
example.
zdmuser> $ZDM_HOME/bin/zdmcli migrate database -sourcedb zdmsdb -sourcenode ocicdb1 -srcroot
-targetnode ocidb1 -targethome /u01/app/oracle/product/12.1.0.2/dbhome_1
-backupuser backup_user@example.com -rsp /u01/app/zdmhome/rhp/zdm/template/zdm_template_zdmsdb.rsp
-tgtauth zdmauth -tgtarg1 user:opc
-tgtarg2 identity_file:/home/zdmuser/.ssh/zdm_service_host.ppk
-tgtarg3 sudo_location:/usr/bin/sudo
Solution 2: For Oracle Database 12c Release 1 and later, ensure that
$ORACLE_HOME/network/admin/sqlnet.ora
points to the correct
location of the TDE wallet, as shown here.
ENCRYPTION_WALLET_LOCATION=(SOURCE=(METHOD=FILE)(METHOD_DATA=(DIRECTORY=/opt/oracle/dcs/commonstore/wallets/tde/$ORACLE_UNQNAME)))
For Oracle Database 11g Release 2 (11.2.0.4) only, ensure that
$ORACLE_HOME/network/admin/sqlnet.ora
points to the correct
location of the TDE wallet as shown below, and replace the variable
$ORACLE_UNQNAME
with the value obtained with the SQL statement
SHOW PARAMETER DB_UNIQUE_NAME
.
ENCRYPTION_WALLET_LOCATION=(SOURCE=(METHOD=FILE)(METHOD_DATA=(DIRECTORY=/opt/oracle/dcs/commonstore/wallets/tde/$ORACLE_UNQNAME)))
For example:
SQL> show parameter db_unique_name
db_unique_name string oci112_region
ENCRYPTION_WALLET_LOCATION=(SOURCE=(METHOD=FILE)(METHOD_DATA=(DIRECTORY=/opt/oracle/dcs/commonstore/wallets/tde/oci112_region)))
Solution 3: Run the following query and make sure that the wallet status is
OPEN
.
SQL> select * from v$encryption_wallet
WRL_TYPE
-------------
WRL_PARAMETER
-------------
STATUS
-------------
file
/opt/oracle/dcs/commonstore/wallets/tde/abc_test
OPEN
Backup Fails with ORA-19914 and ORA-28365
Issue: Source database full backup fails with the following errors.
channel ORA_SBT_TAPE_3: backup set complete, elapsed time: 00:00:15
channel ORA_SBT_TAPE_3: starting compressed full datafile backup set
channel ORA_SBT_TAPE_3: specifying datafile(s) in backup set
input datafile file number=00005 name=+DATA/ODA122/7312FA75F2B202E5E053050011AC5977/DATAFILE/system.382.1003858429
channel ORA_SBT_TAPE_3: starting piece 1 at 25-MAR-19
RMAN-03009: failure of backup command on ORA_SBT_TAPE_3 channel at 03/25/2019 19:09:30
ORA-19914: unable to encrypt backup
ORA-28365: wallet is not open
continuing other job steps, job failed will not be re-run
channel ORA_SBT_TAPE_3: starting compressed full datafile backup set
channel ORA_SBT_TAPE_3: specifying datafile(s) in backup set
Solution: Ensure that the wallet is opened in the database, and in case of CDB, ensure that the wallet is opened in the CDB, all PDBs, and PDB$SEED. See Setting Up the Transparent Data Encryption Wallet in the Zero Downtime Migration documentation for information about setting up TDE.
Either the Bucket Named Object Storage Bucket Name Does Not Exist in the Namespace Namespace or You Are Not Authorized to Access It
See Oracle Support Knowledge Base article "Either the Bucket Named '<Object Storage Bucket Name>' Does not Exist in the Namespace '<Namespace>' or You are not Authorized to Access it (Doc ID 2605518.1)" for the desciption and workarounds for this issue.
Restore Database Fails With AUTOBACKUP does not contain an SPFILE
Issue: During the execution of phase ZDT_CLONE_TGT
,
restore database fails with the following error.
channel C1: looking for AUTOBACKUP on day: 20200427
channel C1: AUTOBACKUP found: c-1482198272-20200427-12
channel C1: restoring spfile from AUTOBACKUP c-1482198272-20200427-12
channel C1: the AUTOBACKUP does not contain an SPFILE
The source database is running using init.ora
file, but during the
restore target phase, the database is trying to restore the server parameter file
(SPFILE) from autobackup, therefore it fails.
Solution: Start the source database using an SPFILE and resubmit the migration job.
Restore Database Fails With ORA-01565
Issue: During the execution of phase ZDT_CLONE_TGT
,
restore database fails with the following error.
</ERRLINE><ERRLINE>With the Partitioning, Real Application Clusters, Automatic Storage Management, OLAP
</ERRLINE><ERRLINE>and Real Application Testing options
</ERRLINE><ERRLINE>
</ERRLINE><ERRLINE>CREATE PFILE='/tmp/zdm833428275/zdm/PFILE/zdm_tgt_mclone_nrt139.pfile' FROM SPFILE
</ERRLINE><ERRLINE>*
</ERRLINE><ERRLINE>ERROR at line 1:
</ERRLINE><ERRLINE>ORA-01565: error in identifying file '?/dbs/spfile@.ora'
</ERRLINE><ERRLINE>ORA-27037: unable to obtain file status
</ERRLINE><ERRLINE>Linux-x86_64 Error: 2: No such file or directory
</ERRLINE><ERRLINE>Additional information: 3
</ERRLINE><ERRLINE>
</ERRLINE><ERRLINE>
</ERRLINE><ERRLINE>Disconnected from Oracle Database 11g Enterprise Edition Release 11.2.0.4.0 - 64bit Production
</ERRLINE><ERRLINE>With the Partitioning, Real Application Clusters, Automatic Storage Management, OLAP
Solution: Start the target database using an SPFILE and resume the migration job.
Transparent Data Encryption General Information
Depending on your source database release, Transparent Data Encryption (TDE) wallet configuration may be required.
- Oracle Database 12c Release 2 and later
For Oracle Database 12c Release 2 and later releases, TDE wallet configuration is mandatory and must be enabled on the source database before migration begins.
If TDE is not enabled, the database migration will fail.
Upon restore, the database tablespaces are encrypted using the wallet.
- Oracle Database 12c Release 1 and earlier
On Oracle Database 12c Release 1 and Oracle Database 11g Release 2 (11.2.0.4), TDE configuration is not required.
For information about the behavior of TDE in an Oracle Cloud environment, see My Oracle Support document Oracle Database Tablespace Encryption Behavior in Oracle Cloud (Doc ID 2359020.1).
Job Fails in Phase ZDM_SETUP_TDE_TGT
Issue: The phase ZDM_SETUP_TDE_TGT
fails with one of the
following errors.
Executing phase ZDM_SETUP_TDE_TGT
Setting up Oracle Transparent Data Encryption (TDE) keystore on the target node oci1121 ...
oci1121: <ERR_FILE><Facility>PRGZ</Facility><ID>ZDM_KEYSTORE_NOT_SETUP_ERR</ID><ARGS><ARG>oci112_phx1z3</ARG></ARGS></ERR_FILE>
PRGO-3007 : failed to migrate database "db11204" with zero downtime
PRCZ-4002 : failed to execute command "/u01/app/18.0.0.0/grid/perl/bin/perl" using the privileged execution plugin "zdmauth" on nodes "oci1121"
PRCZ-2103 : Failed to execute command "/u01/app/18.0.0.0/grid/perl/bin/perl" on node "oci1121" as user "root". Detailed error:
<ERR_FILE><Facility>PRGZ</Facility><ID>ZDM_KEYSTORE_NOT_SETUP_ERR</ID><ARGS><ARG>oci112_phx1z3</ARG></ARGS></ERR_FILE>
Error at target server in /tmp/zdm749527725/zdm/log/mZDM_oss_standby_setup_tde_tgt_71939.log
2019-06-13 10:00:20: Keystore location /opt/oracle/dcs/commonstore/wallets/tde/$ORACLE_UNQNAME does not exists for database 'oci112_region'
2019-06-13 10:00:20: Reporting error:
<ERR_FILE><Facility>PRGZ</Facility><ID>ZDM_KEYSTORE_NOT_SETUP_ERR</ID><ARGS><ARG>oci112_region</ARG></ARGS></ERR_FILE>
Solution:
- Oracle Database 12c Release 1 and later
On the target database, make sure that
$ORACLE_HOME/network/admin/sqlnet.ora
points to the correct location of the TDE wallet. For exmaple:ENCRYPTION_WALLET_LOCATION=(SOURCE=(METHOD=FILE)(METHOD_DATA=(DIRECTORY=/opt/oracle/dcs/commonstore/wallets/tde/$ORACLE_UNQNAME)
- Oracle Database 11g Release 2 (11.2.0.4) only
On the target database, make sure that
$ORACLE_HOME/network/admin/sqlnet.ora
points to the correct location of the TDE wallet, and replace the$ORACLE_UNQNAME
variable with the value obtained from theSHOW PARAMETER DB_UNIQUE_NAME
SQL command.For example, run
SQL> show parameter db_unique_name db_unique_name string oci112_region
and replace
ENCRYPTION_WALLET_LOCATION=(SOURCE=(METHOD=FILE)(METHOD_DATA=(DIRECTORY=/opt/oracle/dcs/commonstore/wallets/tde/$ORACLE_UNQNAME)))
with
ENCRYPTION_WALLET_LOCATION=(SOURCE=(METHOD=FILE)(METHOD_DATA=(DIRECTORY=/opt/oracle/dcs/commonstore/wallets/tde/oci112_region)))
Troubleshooting Post Migration Automatic Backup Failures
Issue: Post migration, on the target database, Automatic Backup might fail.
You can verify the failure using the console in Bare Metal, VM and Exadata > DB Systems > DB System Details > Database Details > Backups.
Solution: Get the RMAN configuration settings from one of the following places.
- Zero Downtime Migration documentation in Target Database Prerequisites, if captured
- The log files at
/opt/oracle/dcs/log/hostname/rman/bkup/db_unique_name/
/tmp/zdmXXX/zdm/zdm_TDBNAME_rman.dat
For example, using the second option, you can get the RMAN configuration
settings from
/opt/oracle/dcs/log/ocidb1/rman/bkup/ocidb1_abc127/rman_configure*.log
,
then reset any changed RMAN configuration settings for the target database to ensure
that automatic backup works without any issues.
If this workaround does not help, then debug further by getting the RMAN job ID by
running the DBCLI command, list-jobs
, and describe the job details for
more error details by running the DBCLI command describe-job -i JOB
ID
from the database server as the root user.
For example, during the test, the following highlighted settings were modified to make Automatic Backup work.
rman target /
Recovery Manager: Release 12.2.0.1.0 - Production on Mon Jul 8 11:00:18 2019
Copyright (c) 1982, 2017, Oracle and/or its affiliates. All rights reserved.
connected to target database: ORCL (DBID=1540292788)
RMAN> show all;
using target database control file instead of recovery catalog
RMAN configuration parameters for database with db_unique_name OCIDB1_ABC127 are:
CONFIGURE RETENTION POLICY TO RECOVERY WINDOW OF 30 DAYS;
CONFIGURE BACKUP OPTIMIZATION OFF;
CONFIGURE DEFAULT DEVICE TYPE TO DISK; # default
CONFIGURE CONTROLFILE AUTOBACKUP ON;
CONFIGURE CONTROLFILE AUTOBACKUP FORMAT FOR DEVICE TYPE SBT_TAPE TO '%F'; # default
CONFIGURE CONTROLFILE AUTOBACKUP FORMAT FOR DEVICE TYPE DISK TO '%F'; # default
CONFIGURE DEVICE TYPE 'SBT_TAPE' PARALLELISM 4 BACKUP TYPE TO COMPRESSED BACKUPSET;
CONFIGURE DEVICE TYPE DISK PARALLELISM 1 BACKUP TYPE TO BACKUPSET; # default
CONFIGURE DATAFILE BACKUP COPIES FOR DEVICE TYPE SBT_TAPE TO 1; # default
CONFIGURE DATAFILE BACKUP COPIES FOR DEVICE TYPE DISK TO 1; # default
CONFIGURE ARCHIVELOG BACKUP COPIES FOR DEVICE TYPE SBT_TAPE TO 1; # default
CONFIGURE ARCHIVELOG BACKUP COPIES FOR DEVICE TYPE DISK TO 1; # default
CONFIGURE CHANNEL DEVICE TYPE DISK MAXPIECESIZE 2 G;
CONFIGURE CHANNEL DEVICE TYPE 'SBT_TAPE' MAXPIECESIZE 2 G FORMAT '%d_%I_%U_%T_%t' PARMS
'SBT_LIBRARY=/opt/oracle/dcs/commonstore/pkgrepos/oss/odbcs/libopc.so ENV=(OPC_PFILE=/opt/oracle/dcs/commonstore/objectstore/opc_pfile/1245080042/opc_OCIDB1_ABC127.ora)';
CONFIGURE MAXSETSIZE TO UNLIMITED; # default
CONFIGURE ENCRYPTION FOR DATABASE ON;
CONFIGURE ENCRYPTION ALGORITHM 'AES128'; # default
CONFIGURE COMPRESSION ALGORITHM 'MEDIUM' AS OF RELEASE 'DEFAULT' OPTIMIZE FOR LOAD TRUE;
CONFIGURE RMAN OUTPUT TO KEEP FOR 7 DAYS; # default
CONFIGURE ARCHIVELOG DELETION POLICY TO BACKED UP 1 TIMES TO 'SBT_TAPE';
CONFIGURE SNAPSHOT CONTROLFILE NAME TO '+RECO/ OCIDB1_ABC127/controlfile/snapcf_ocidb1_abc127.f';
CONFIGURE CONTROLFILE AUTOBACKUP FORMAT FOR DEVICE TYPE DISK clear;
RMAN>
Post Migration Automatic Backup Fails With DCS-10045
Issue: Post migration, Automatic Backup fails with the following error for non-TDE enabled migrated Oracle Database releases 11.2.0.4 and 12.1.0.2.
DCS-10045: Validation error encountered: Backup password is mandatory to take OSS backup for non-tde enabled database...
You can verify this error by getting the RMAN job ID by running DBCLI command
list-jobs
, and describe the job details to get the error details by
running DBCLI command describe-job -i JOB ID
from
the database server as the root user.
Solution:
- Find the TDE wallet location.
The Oracle Cloud Infrastructure provisioned database instance will have following entry in
sqlnet.ora
.ENCRYPTION_WALLET_LOCATION=(SOURCE=(METHOD=FILE)(METHOD_DATA=(DIRECTORY=/opt/oracle/dcs/commonstore/wallets/tde/$ORACLE_UNQNAME)))
- Remove the
cwallet.sso
file from the wallet location.For example,
/opt/oracle/dcs/commonstore/wallets/tde/$ORACLE_UNQNAME
. - For Oracle Database 11g Release 2, do the folowing steps.
- Connect to database using SQL*Plus as sysdba and verify the current wallet
location.
SQL> select * from v$encryption_wallet; WRL_TYPE WRL_PARAMETER STATUS file /opt/oracle/dcs/commonstore/wallets/tde/ocise112_region OPEN
- Close the wallet in the
database.
SQL> alter system set wallet close;
- Open the wallet using the wallet
password.
SQL> alter system SET WALLET open IDENTIFIED BY "walletpassword"
- Set the master encryption
key.
SQL> alter system set encryption key identified by "walletpassword"
- Recreate the autologin SSO
file.
/home/oracle>orapki wallet create -wallet /opt/oracle/dcs/commonstore/wallets/tde/$ORACLE_UNQNAME -auto_login Oracle PKI Tool : Version 11.2.0.4.0 - Production Copyright (c) 2004, 2013, Oracle and/or its affiliates. All rights reserved. Enter wallet password: #
- Retry Automatic Backup.
- Connect to database using SQL*Plus as sysdba and verify the current wallet
location.
- For Oracle Database 12c, do the folowing steps.
- Connect to database using SQL*Plus as sysdba and verify the current wallet
location and
status.
SQL> SELECT wrl_parameter, status, wallet_type FROM v$encryption_wallet; WRL_PARAMETER STATUS WALLET_TYPE /opt/oracle/dcs/commonstore/wallets/tde/ocise112_region OPEN_NO_MASTER_KEY OPEN
If the STATUS column contains a value of OPEN_NO_MASTER_KEY, you must create and activate the master encryption key.
- Close the wallet in the
database.
SQL> alter system set wallet close;
- Open the wallet-using
password.
SQL> ADMINISTER KEY MANAGEMENT SET KEYSTORE open IDENTIFIED BY "walletpassword" CONTAINER=all;
- Set the master encryption
key.
SQL> ADMINISTER KEY MANAGEMENT SET KEY IDENTIFIED BY "walletpassword" with backup;
Log in to each PDB and run
SQL> ALTER SESSION SET CONTAINER = PDB_NAME; SQL> ADMINISTER KEY MANAGEMENT SET KEY IDENTIFIED BY "walletpassword" with backup;
- Create the auto login
keystore.
SQL> ADMINISTER KEY MANAGEMENT CREATE AUTO_LOGIN KEYSTORE FROM KEYSTORE 'path to wallet directory' IDENTIFIED BY "walletpassword";
- Retry Automatic Backup.
- Connect to database using SQL*Plus as sysdba and verify the current wallet
location and
status.
Post Migration Automatic Backup Fails With DCS-10096
Issue: Post migration, Automatic Backup fails with the following error.
DCS-10096:RMAN configuration 'Retention policy' must be configured as 'configure retentio n
policy to recovery window of 30 days'
You can verify this error by getting the RMAN job ID by running DBCLI
command list-jobs
, and describe the job details for more error details
by running DBCLI command describe-job -i JOB
ID
from the database server as the root user.
Solution: Log in into RMAN prompt and configure the retention policy.
[oracle@racoci1 ~]$ rman target /
Recovery Manager: Release 12.2.0.1.0 - Production on Wed Jul 17 11:04:35 2019
Copyright (c) 1982, 2017, Oracle and/or its affiliates. All rights reserved.
connected to target database: SIODA (DBID=2489657199)
RMAN> CONFIGURE RETENTION POLICY TO RECOVERY WINDOW OF 30 DAYS;
old RMAN configuration parameters:
CONFIGURE RETENTION POLICY TO RECOVERY WINDOW OF 7 DAYS;
new RMAN configuration parameters:
CONFIGURE RETENTION POLICY TO RECOVERY WINDOW OF 30 DAYS;
new RMAN configuration parameters are successfully stored
Retry Automatic Backup.
INS-42505 Warning Shown During Installation
/stage/user/ZDM_KIT_relnumber>./zdminstall.sh setup
oraclehome=/stage/user/grid oraclebase=/stage/user/base
ziploc=/stage/user/ZDM_KIT_relnumber/rhp_home.zip -zdm
---------------------------------------
Unzipping shiphome to gridhome
---------------------------------------
Unzipping shiphome...
Shiphome unzipped successfully..
---------------------------------------
##### Starting GridHome Software Only Installation #####
---------------------------------------
Launching Oracle Grid Infrastructure Setup Wizard...
[WARNING] [INS-42505] The installer has detected that the Oracle Grid
Infrastructure home software at (/stage/user/grid) is not complete.
CAUSE: Following files are missing:
...
Solution: This warning message can be ignored. It does not affect the installation or cause any issues for migration.
Evaluation Fails in Phase ZDM_GET_SRC_INFO
Issue: During the evaluation (-eval
) phase of the migration
process, the evaluation fails in the ZDM_GET_SRC_INFO
phase with the
following error for the source single instance deployed without Grid infrastructure.
Executing phase ZDM_GET_SRC_INFO
retrieving information about database "zdmsidb" ...
PRCF-2056 : The copy operation failed on node: "zdmsidb".
Details: {1}
PRCZ-4002 : failed to execute command "/bin/cp" using the privileged
execution plugin "zdmauth" on nodes "zdmsidb"
scp: /etc/oratab: No such file or directory
Solution: Make an ORACLE_HOME
value entry in file
/etc/oratab
with value
db_name:$ORACLE_HOME:N
,
as shown in this example.
zdmsidb:/u01/app/oracle/product/12.2.0.1/dbhome_1:N
Unable to Rerun MIGRATE DATABASE Command
Issue:
Zero Downtime Migration blocks attempts to rerun the MIGRATE DATABASE
command for a specified database if that database is already part of an ongoing
migration job.
EXECUTING
or
PAUSED
state using the ZDMCLI ABORT JOB
command as
follows.-bash-4.2$ ./zdmcli abort job -jobid 70
server.example.com: Audit ID: 189
Migration Job Fails at ZDM_GET_SRC_INFO
Issue: A migration job fails with the following error.
[opc@zdm-server rhp]$ cat /home/opc/zdm_base/chkbase/scheduled/job-34-2021-01-23-14:10:32.log
zdm-server: 2021-01-23T14:10:32.155Z : Processing response file ...
zdm-server: 2021-01-23T14:10:32.262Z : Starting zero downtime migrate operation ...
PRCZ-4002 : failed to execute command "/bin/cp" using the privileged execution plugin "zdmauth" on nodes "PROD.compute-usconnectoneb95657.oraclecloud.internal"
Solution: You must set up SSH connectivity without a passphrase for the oracle user.
Migration Evaluation Failure with Java Exception Invalid Key Format
Issue: The following conditions are seen:
-
Zero Downtime Migration
migration -eval
command fails with the following error.Result file path contents: "/u01/app/zdmbase/chkbase/scheduled/job-19-2019-12-02-03:46:19.log" zdm-server.ocitoolingsn.ocitooling.oraclevcn.com: Processing response file ... null
-
The file
$ZDM_BASE/<zdm service host>/rhp/rhpserver.log.0
contains the following entry.Verify below error message observed in file $ZDM_BASE/<zdm service host>/rhp/rhpserver.log.0 rhpserver.log.7:[pool-58-thread-1] [ 2019-12-02 02:08:15.178 GMT ] [JSChChannel.getKeyPair:1603] Exception : java.security.spec.InvalidKeySpecException: java.security.InvalidKeyException: invalid key format
-
The Zero Downtime Migration installed user (For example: zdmuser) private key (id_rsa) file has the following entries.
-----BEGIN OPENSSH PRIVATE KEY---------- MIIEogIBAAKCAQEAuPcjftR6vC98fAbU4FhYVKPqc0CSgibtMSouo1DtQ06ROPN0 XpIEL4r8nGp+c5GSDONyhf0hiltBzg0fyqyurSw3XfGJq2Q6EQ61aL95Rt9CZh6b JSUwc69T4rHjvRnK824k4UpfUIqafOXb2mRgGVUkldo4yy+pLoGq1GwbsIYbS4tk uaYPKZ3A3H9ZA7MtZ5M0sNqnk/4Qy0d8VONWozxOLFC2A8zbbe7GdQw9khVqDb/x END OPENSSH PRIVATE KEY-----
Solution: Authentication key pairs (private and public key) are not
generated using the ssh-keygen
utility, so you must generate
authentication key pairs using steps in Generating a Private SSH Key Without a Passphrase.
After generating authentication key pairs, the private key file content looks like the following.
-----BEGIN RSA PRIVATE KEY-----
MIIEogIBAAKCAQEAuPcjftR6vC98fAbU4FhYVKPqc0CSgibtMSouo1DtQ06ROPN0
XpIEL4r8nGp+c5GSDONyhf0hiltBzg0fyqyurSw3XfGJq2Q6EQ61aL95Rt9CZh6b
JSUwc69T4rHjvRnK824k4UpfUIqafOXb2mRgGVUkldo4yy+pLoGq1GwbsIYbS4tk
uaYPKZ3A3H9ZA7MtZ5M0sNqnk/4Qy0d8VONWozxOLFC2A8zbbe7GdQw9khVqDb/x
-----END RSA PRIVATE KEY-----
Set up connectivity with the newly generated authentication key pairs and resume the migration job.
Migration Evaluation Fails with Error PRCG-1022
Issue: The following conditions are seen:
$ZDM_HOME/bin/zdmcli migrate database -sourcedb zdmsdb -sourcenode ocicdb1
-srcauth zdmauth -srcarg1 user:opc
-srcarg2 identity_file:/home/zdmuser/.ssh/zdm_service_host.ppk
-srcarg3 sudo_location:/usr/bin/sudo -targetnode ocidb1 -backupuser backup_user@example.com
-rsp /u01/app/zdmhome/rhp/zdm/template/zdm_template_zdmsdb.rsp -tgtauth zdmauth
-tgtarg1 user:opc -tgtarg2 identity_file:/home/zdmuser/.ssh/zdm_service_host.ppk
-tgtarg3 sudo_location:/usr/bin/sudo -eval
PRCG-1238 : failed to execute the Rapid Home Provisioning action for command 'migrate database'
PRCG-1022 : failed to connect to the Rapid Home Provisioning daemon for cluster anandutest
Failed to retrieve RMIServer stub: javax.naming.ServiceUnavailableException
[Root exception is java.rmi.ConnectException: Connection refused to host:
anandutest; nested exception is: java.net.ConnectException: Connection refused (Connection refused)]
Solution: Start the Zero Downtime Migration service using the
$ZDM_HOME/bin/zdmservice START
command, then run
any ZDMCLI
commands.
Unable to Resume a Migration Job
Issue: Zero Downtime Migration writes the
source and target log files to the /tmp/zdm-unique id
directory in the
respective source and target database servers.
If you pause a migration job and and then resume the job after several (sometimes 15-20
days), the /tmp/zdm-unique id
directory might be deleted or purged as
part of a clean up or server reboot that also cleans up /tmp.
Solution: After pausing a migration job, back up the
/tmp/zdm-unique id
directory. Before resuming the migration job,
check the /tmp
directory for /zdm-unique id
, and if it
is missing, restore the directory and its contents with your backup.
Migration Job Fails at ZDM_SWITCHOVER_SRC
Issue: A migration job fails at ZDM_SWITCHOVER_SRC
phase.
Solutions:
-
Ensure that there is connectivity from PRIMARY database nodes to STANDBY database nodes so the redo log are shipped as expected.
-
A job will fail at
ZDM_SWITCHOVER_SRC
if the recovery process (MRP0) is not running at the target. The recovery process reason for failure should be corrected if MRP0 is not running at Oracle Cloud Database Standby Instance, and then the process should be started manually at Oracle Cloud Database Standby Instance before the migration job can be resumed.
Additional Information for Migrating to Exadata Cloud Service
Read the following for general information, considerations, and links to more information about using Zero Downtime Migration to migrate your database to Exadata Cloud Service.
Considerations for Migrating to Exadata Cloud Service
For this release of Zero Downtime Migration be aware of the following considerations.
- If the source database is release 18c, then the target home should be at release 18.6 or later to avoid issues such as Bug 29445548 Opening Database In Cloud Environment Fails With ORA-600.
- If a backup was performed when one of the configured instances is down, you will encounter Bug 29863717 - DUPLICATING SOURCE DATABASE FAILED BECAUSE INSTANCE 1 WAS DOWN.
- The TDE keystore password must be set in the credential wallet. To set
the password as part of the Zero Downtime Migration workflow, specify the
-tdekeystorewallet tde_wallet_path
or-tdekeystorepasswd
argument irrespective of whether the wallet usesAUTOLOGIN
orPASSWORD
. In either case the password is stored in the credential wallet. If the-tdekeystorepasswd
argument is not supplied, then Zero Downtime Migration skips the settingtde_ks_passwd
key in the credential wallet, and no error is thrown. - The target environment must be installed with latest DBaaS Tooling RPM
with
db_unique_name
change support to be installed. - Provision a target database from the console without enabling auto-backups. In the Configure database backups section do not select the Enable automatic backups option.
Exadata Cloud Service Database Registration
Post migration, register the Exadata Cloud Service database, and make sure its meets all of the requirements.
Run the following commands on the Exadata Cloud Service database server as the root user.
/root>dbaascli registerdb prereqs --dbname db_name --db_unique_name db_unique_name
/root>dbaascli registerdb begin --dbname db_name --db_unique_name db_unique_name
For example
/root>dbaascli registerdb prereqs --dbname ZDM122 --db_unique_name ZDM122_phx16n
DBAAS CLI version 18.2.3.2.0
Executing command registerdb prereqs --db_unique_name ZDM122_phx16n
INFO: Logfile Location: /var/opt/oracle/log/ZDM122/registerdb/registerdb_2019-08-14_05:35:31.157978280334.log
INFO: Prereqs completed successfully
/root>
/root>dbaascli registerdb begin --dbname ZDM122 --db_unique_name ZDM122_phx16n
DBAAS CLI version 18.2.3.2.0
Executing command registerdb begin --db_unique_name ZDM122_phx16n
Logfile Location: /var/opt/oracle/log/ZDM122/registerdb/registerdb_2019-08-14_05:45:27.264851309165.log
Running prereqs
DBAAS CLI version 18.2.3.2.0
Executing command registerdb prereqs --db_unique_name ZDM122_phx16n
INFO: Logfile Location: /var/opt/oracle/log/ZDM122/registerdb/registerdb_2019-08-14_05:45:29.000432309894.log
INFO: Prereqs completed successfully
Prereqs completed
Running OCDE .. will take time ..
OCDE Completed successfully.
INFO: Database ZDM122 registered as Cloud database
/root>
Exadata Cloud Service Automatic Backup Issues
Check the backup configuration before you enable automatic backup from the console. You
can use the get config
command as shown in the first step below. You
should see bkup_oss=no
before you enable automatic backup.
You might see the error message in the console, "A backup configuration exists for this database. You must remove the existing configuration to use Oracle Cloud Infrastructure's managed backup feature."
To fix this error, remove the existing configuration.
First, make sure the automatic backup is disabled from the UI, then follow these steps to remove the existing backup configuration.
- Generate a backup configuration
file.
/var/opt/oracle/bkup_api/bkup_api get config --file=/tmp/db_name.bk --dbname=db_name
For example:
/var/opt/oracle/bkup_api/bkup_api get config --file=/tmp/zdmdb.bk --dbname=zdmdb
- Open the /tmp/db_name.bk file you created in the previous
step.
For example: Open /tmp/zdmdb.bk
change bkup_oss=yes from bkup_oss=no
- Disable OSS backup by setting
bkup_oss=no
./var/opt/oracle/bkup_api/bkup_api set config --file=/tmp/db_name.bk --dbname=db_name
For example:
/var/opt/oracle/bkup_api/bkup_api set config --file=/tmp/zdmdb.bk --dbname=zdmdb
- Check reconfigure
status.
/var/opt/oracle/bkup_api/bkup_api configure_status --dbname=db_name
For example:
/var/opt/oracle/bkup_api/bkup_api configure_status --dbname=zdmdb
Now enable automatic backup from console.
Verify the backups from the console. Click Create Backup to create a manual backup, and a backup should be created without any issues. and also Automatic Backup should be successful.
Additional Information for Migrating to Exadata Cloud at Customer
Read the following for general information, considerations, and links to more information about using Zero Downtime Migration to migrate your database to Exadata Cloud at Customer.
Considerations for Migrating to Exadata Cloud at Customer
For this release of Zero Downtime Migration be aware of the following considerations.
- You must apply the regDB patch for Bug 29715950 - "modify regdb to
handle db_unique_name not same as db_name" on all Exadata Cloud at Customer nodes.
This is required for the
ZDM_MANIFEST_TO_CLOUD
phase. Please note that the regDB tool is part of DBaaS Tooling. - If the source database is release 18c, then the target home should be at release 18.6 or later to avoid issues such as Bug 29445548 Opening Database In Cloud Environment Fails With ORA-600.
- PDB conversion related phases are listed in
-listphases
and can be ignored. Those are no-op phases. - If the backup medium is Zero Data Loss Recovery Appliance, then all configured
instances should be up at the source when a
FULL
orINCREMENTAL
backup is performed. - If a backup was performed when one of the configured instances is down, you will encounter Bug 29863717 - DUPLICATING SOURCE DATABASE FAILED BECAUSE INSTANCE 1 WAS DOWN.
- The TDE keystore password must be set in the credential wallet. To set
the password as part of the Zero Downtime Migration workflow, specify the
-tdekeystorewallet tde_wallet_path
or-tdekeystorepasswd
argument irrespective of whether the wallet usesAUTOLOGIN
orPASSWORD
. In either case the password is stored in the credential wallet. If the-tdekeystorepasswd
argument is not supplied, then Zero Downtime Migration skips the settingtde_ks_passwd
key in the credential wallet, and no error is thrown. - The target environment must be installed with latest DBaaS Tooling RPM
with
db_unique_name
change support to be installed.
Documentation Accessibility
For information about Oracle's commitment to accessibility, visit the Oracle Accessibility Program website at http://www.oracle.com/pls/topic/lookup?ctx=acc&id=docacc.
Access to Oracle Support
Oracle customers that have purchased support have access to electronic support through My Oracle Support. For information, visit http://www.oracle.com/pls/topic/lookup?ctx=acc&id=info or visit http://www.oracle.com/pls/topic/lookup?ctx=acc&id=trs if you are hearing impaired.
Oracle Zero Downtime Migration Release Notes, Release 21c (21.2)
F43501-05
March 2022
Copyright © 2019, 2022, Oracle and/or its affiliates.