1. Release Notes
  2. Known Issues with Oracle Database Appliance in This Release

4 Known Issues with Oracle Database Appliance in This Release

The following are known issues deploying, updating, and managing Oracle Database Appliance in this release.

Known Issues When Patching Oracle Database Appliance

Understand the known issues when patching Oracle Database Appliance to this release.

Parent topic: Known Issues with Oracle Database Appliance in This Release

Errors when running oakcli commands on Oracle Database Appliance Virtualized Platform

You may encounter an error when you run oakcli commands on Oracle Database Appliance Virtualized Platforms.

For example, the following error is encountered when running the oakcli expand storage command on Oracle Database Appliance Virtualized Platforms.
sh: /opt/oracle/oak/bin/odaadmcli: No such file or directory

This error occurs in Oracle Database Appliance Virtualized Platforms with Oracle Autonomous Health Framework version 19.3 or earlier, where Oracle Autonomous Health Framework is installed in the DCS home directory (/opt/oracle/dcs) under /opt/oracle/dcs/oracle.ahf/.

You can check your AHF version:
# tfactl -version;orachk -v
TFA Version : 193000
TFA Build ID : 20200108023845
ORACHK  VERSION: 19.3.0_20200108
# cd /opt/oracle
# ls
dcs  extapi  oak
# cd dcs
# ls
oracle.ahf
#

Hardware Models

Oracle Database Appliance High-Availability hardware models Virtualized Platform deployments

Workaround

Uninstall the installed Oracle Autonomous Health Framework version and download the latest Oracle Autonomous Health Framework version on both nodes of your Oracle Database Appliance High-Availability deployment.

Run the following steps on both nodes:
  1. Uninstall Oracle Autonomous Health Framework.
    # tfactl uninstall -local -deleterepo

    For Oracle Autonomous Health Framework version 19.3, manually remove the installed version, by running the following command:

    # rpm -e oracle-ahf
  2. Check that no oracle-ahf rpm is installed on the system.
    # rpm -qi oracle-ahf
package oracle-ahf is not installed
#
  3. Remove the DCS home (/opt/oracle/dcs) and its sub-directories.
    # cd /opt/oracle
# rm -rf /opt/oracle/dcs
  4. Check and ensure the DCS home directory (/opt/oracle/dcs) does not exist under the /opt/oracle directory on the system.
     # cd /opt/oracle/
 # ls
 extapi  oak
 #
  5. Download the latest Oracle Autonomous Health Framework version from My Oracle Support.
  6. Copy the downloaded zip file to /root on your system, for example, AHF-LINUX_v20.1.1.zip. The zip file name for each Oracle Autonomous Health Framework version may change based on the latest release number.
     # /root/AHF-LINUX_v20.1.1.zip
  7. Create the ahfinstall directory under /tmp.
    # cd /tmp
# mkdir ahfinstall
  8. Unzip the Oracle Autonomous Health Framework zip file into /tmp/ahfinstall.
    # cd ahfinstall
# unzip /root/AHF-LINUX_v20.1.1.zip
  9. Run the ahf_setup installer from /tmp/ahfinstall. During installation, select the default location for installation.

    Note: Do not choose the DCS home directory as the default Oracle Autonomous Health Framework location for install. 

     # ./ahf_setup
  10. Check the Oracle TFA and ORAchk versions.
    #tfactl -version;orachk -v
TFA Version : 201100
TFA Build ID : 20200331131556
ORACHK  VERSION: 20.1.1_20200331
#
  11. Check that Oracle Autonomous Health Framework is installed in the /opt/oracle.ahf directory.
Example of Running the Workaround
#  tfactl -version;orachk -v
TFA Version : 193000
TFA Build ID : 20200108023845
ORACHK  VERSION: 19.3.0_20200108

# tfactl uninstall -local -deleterepo
Starting AHF Uninstall
AHF will be uninstalled on: node2

Do you want to continue with AHF uninstall ? [Y]|N : y

Stopping AHF service on local node node2...
Sleeping for 10 seconds...

Stopping TFA Support Tools...
TFA-00002 Oracle Trace File Analyzer (TFA) is not running
Stopping orachk scheduler ...
Removing orachk cache discovery....
Successfully completed orachk cache discovery removal.
Removed orachk from inittab
Removing AHF setup on node2:
Removing /etc/rc.d/rc0.d/K17init.tfa
Removing /etc/rc.d/rc1.d/K17init.tfa
Removing /etc/rc.d/rc2.d/K17init.tfa
Removing /etc/rc.d/rc4.d/K17init.tfa
Removing /etc/rc.d/rc6.d/K17init.tfa
Removing /etc/init.d/init.tfa...
Removing /opt/oracle.ahf/jre
Removing /opt/oracle.ahf/common
Removing /opt/oracle.ahf/bin
Removing /opt/oracle.ahf/python
Removing /opt/oracle.ahf/analyzer
Removing /opt/oracle.ahf/tfa
Removing /opt/oracle.ahf/orachk
Removing /opt/oracle.ahf/ahf
Removing /u01/app/grid/oracle.ahf/data/node2
Removing /opt/oracle.ahf/install.properties
Removing /u01/app/grid/oracle.ahf/data/repository
Removing /u01/app/grid/oracle.ahf/data
Removing /u01/app/grid/oracle.ahf
Removing AHF Home : /opt/oracle.ahf
# tfactl -version;orachk -v
-bash: /usr/bin/tfactl: No such file or directory
-bash: /usr/bin/orachk: No such file or directory
# rpm -e oracle-ahf
warning:    erase unlink of /opt/oracle.ahf failed: No such file or directory
# rpm -qi oracle-ahf
package oracle-ahf is not installed
#
# cd /opt/oracle
# ls
dcs  extapi  oak
# rm -rf dcs/
# ls
extapi  oak
#
#cd /tmp
#mkdir ahfinstall
#cd ahfinstall/
# pwd
/tmp/ahfinstall
# unzip /root/AHF-LINUX_v20.1.1.zip
Archive:  /root/AHF-LINUX_v20.1.1.zip
  inflating: README.txt            
  inflating: ahf_setup
 

# ls
ahf_setup  README.txt
# ./ahf_setup

AHF Installer for Platform Linux Architecture x86_64
AHF Installation Log : /tmp/ahf_install_85684_2020_04_01-00_15_38.log
Starting Autonomous Health Framework (AHF) Installation
AHF Version: 20.1.1 Build Date: 202003311315
Default AHF Location : /opt/oracle.ahf
Do you want to install AHF at [/opt/oracle.ahf] ? [Y]|N : y
AHF Location : /opt/oracle.ahf
AHF Data Directory stores diagnostic collections and metadata.
AHF Data Directory requires at least 5GB (Recommended 10GB) of free space.
Choose Data Directory from below options :

1. /u01/app/grid [Free Space : 51458 MB]
2. Enter a different Location

Choose Option [1 - 2] : 1

AHF Data Directory : /u01/app/grid/oracle.ahf/data

Do you want to add AHF Notification Email IDs ? [Y]|N :   xxxxxxxxxxxx

AHF will also be installed/upgraded on these Cluster Nodes :

1. node1

The AHF Location and AHF Data Directory must exist on the above nodes
AHF Location : /opt/oracle.ahf
AHF Data Directory : /u01/app/grid/oracle.ahf/data

Do you want to install/upgrade AHF on Cluster Nodes ? [Y]|N : N
Extracting AHF to /opt/oracle.ahf
Configuring TFA Services
Discovering Nodes and Oracle Resources
Not generating certificates as GI discovered
Starting TFA Services

.----------------------------------------------------------------------------------.
| Host          | Status of TFA | PID   | Port | Version    | Build ID             |
+---------------+---------------+-------+------+------------+----------------------+
| node2 | RUNNING       | 87939 | 5000 | 20.1.1.0.0 | 20110020200331131556 |
'---------------+---------------+-------+------+------------+----------------------'

Running TFA Inventory...

Adding default users to TFA Access list...

.--------------------------------------------------------------------.
|                    Summary of AHF Configuration                    |
+-----------------+--------------------------------------------------+
| Parameter       | Value                                            |
+-----------------+--------------------------------------------------+
| AHF Location    | /opt/oracle.ahf                                  |
| TFA Location    | /opt/oracle.ahf/tfa                              |
| Orachk Location | /opt/oracle.ahf/orachk                           |
| Data Directory  | /u01/app/grid/oracle.ahf/data                    |
| Repository      | /u01/app/grid/oracle.ahf/data/repository         |
| Diag Directory  | /u01/app/grid/oracle.ahf/data/node2/diag |
'-----------------+--------------------------------------------------'


Starting orachk daemon from AHF ...
AHF install completed on node1.
Installing AHF on Remote Nodes :
AHF will be installed on node1, Please wait.
Installing AHF on node1:
[node1] Copying AHF Installer
[node1] Running AHF Installer
Adding rpm Metadata to rpm database on ODA system
RPM File /opt/oracle.ahf/rpms/oracle-ahf-201100-20200331131556.x86_64.rpm
Preparing...                ########################################### [100%]
---------

   1:oracle-ahf             ########################################### [100%]
Upgrading oracle-ahf
warning:    erase unlink of /opt/oracle/dcs/oracle.ahf failed: No such file or directory

AHF binaries are available in /opt/oracle.ahf/bin
AHF is successfully installed
Moving /tmp/ahf_install_85684_2020_04_01-00_15_38.log to /u01/app/grid/oracle.ahf/data/node2/diag/ahf/

# tfactl -version;orachk -v
TFA Version : 201100
TFA Build ID : 20200331131556
ORACHK  VERSION: 20.1.1_20200331

# ls -l /opt/
........
drwxr-xr-x 11 root root 4096 Mar 31 13:16 oracle.ahf
........
#

This issue is tracked with Oracle bug 31014517.

Parent topic: Known Issues When Patching Oracle Database Appliance

Patching Oracle Database home fails with errors

When applying the patch for Oracle Database homes, an error is encountered.

Error Encountered When Patching Oracle Database Homes on Bare Metal Systems:

When patching Oracle Database homes on baremetal systems, the odacli update-dbhome command fails with an error similar to the following: 

Please stop TFA before dbhome patching.

To resolve this issue, follow the steps described in the Workaround.

Error Encountered When Patching Oracle Database Homes on Virtualized Platform:

When patching Oracle Database homes on Virtualized Platform, patching fails with an error similar to the following:

INFO: Running prepatching on local node
WARNING: errors seen during prepatch on local node
ERROR: Unable to apply the patch 1

Check the prepatch log file generated in the directory /opt/oracle/oak/log/hostname/patch/18.8.0.0.0. You can also view the prepatch log for the last run with the command ls -lrt prepatch_*.log. Check the last log file in the command output.

In the log file, search for entries similar to the following:

ERROR: date_time_stamp: TFA is running on one or more nodes.
WARNING: date_time_stamp: Shutdown TFA and then restart patching
INFO: date_time_stamp: Read the Release Notes for additional information.

To resolve this issue, follow the steps described in the Workaround.

Hardware Models

All Oracle Database Appliance hardware models

Workaround

On Oracle Database Appliance bare metal systems, do the following:
  1. Run tfactl stop on all the nodes in the cluster.
  2. Restart patching once Oracle TFA Collector has stopped on all nodes.
On Oracle Database Appliance Virtualized Platform, do the following:
  1. Run /etc/init.d/init.tfa stop on all the nodes in the cluster.
  2. Restart patching once Oracle TFA Collector has stopped on all nodes.

This issue is tracked with Oracle bugs 30799713 and 30892062.

Parent topic: Known Issues When Patching Oracle Database Appliance

Error in patching Oracle Database Appliance

When applying the server patch for Oracle Database Appliance, an error is encountered.

Error Encountered When Patching Bare Metal Systems:

When patching the appliance on bare metal systems, the odacli update-server command fails with the following error: 

Please stop TFA before server patching.

To resolve this issue, follow the steps described in the Workaround.

Error Encountered When Patching Virtualized Platform:

When patching the appliance on Virtualized Platform, patching fails with an error similar to the following:

INFO: Running prepatching on local node
WARNING: errors seen during prepatch on local node
ERROR: Unable to apply the patch 1

Check the prepatch log file generated in the directory /opt/oracle/oak/log/hostname/patch/18.8.0.0.0. You can also view the prepatch log for the last run with the command ls -lrt prepatch_*.log. Check the last log file in the command output.

In the log file, search for entries similar to the following:

ERROR: date_time_stamp: TFA is running on one or more nodes.
WARNING: date_time_stamp: Shutdown TFA and then restart patching
INFO: date_time_stamp: Read the Release Notes for additional information.

To resolve this issue, follow the steps described in the Workaround.

Hardware Models

All Oracle Database Appliance hardware models

Workaround

On Oracle Database Appliance bare metal systems, do the following:
  1. Run tfactl stop on all the nodes in the cluster.
  2. Restart patching once Oracle TFA Collector has stopped on all nodes.
On Oracle Database Appliance Virtualized Platform, do the following:
  1. Run /etc/init.d/init.tfa stop on all the nodes in the cluster.
  2. Restart patching once Oracle TFA Collector has stopped on all nodes.

This issue is tracked with Oracle bugs 30260318 and 30892062.

Parent topic: Known Issues When Patching Oracle Database Appliance

Patching errors on Oracle Database Appliance Virtualized Platform

When applying the server patch for Oracle Database Appliance Virtualized Platform, an error is encountered.

Error Encountered When Patching Virtualized Platform:

When patching the appliance on Virtualized Platform, patching fails with an error similar to the following:

ERROR: Unable to apply the GRID patch 
ERROR: Failed to patch server (grid) component

This error can occur even if you stopped Oracle TFA Collector before patching. During server patching on the node, Oracle TFA Collector is updated and this can restart the TFA processes, thus causing an error. To resolve this issue, follow the steps described in the Workaround.

Hardware Models

All Oracle Database Appliance hardware models

Workaround

On Oracle Database Appliance Virtualized Platform, do the following:
  1. Run /etc/init.d/init.tfa stop on all the nodes in the cluster.
  2. Run the command:
    /u01/app/18.0.0.0/grid/bin/cluutil -ckpt -oraclebase /u01/app/grid -chkckpt -name ROOTCRS_PREPATCH -status

    Verify that the command output is SUCCESS.

  3. If the command output was SUCCESS, then run the following commands on all the nodes:
    /u01/app/18.0.0.0/grid/crs/install/rootcrs.sh -prepatch -rollback 
/u01/app/18.0.0.0/grid/crs/install/rootcrs.sh -postpatch
  4. Restart patching.

This issue is tracked with Oracle bug 30886701.

Parent topic: Known Issues When Patching Oracle Database Appliance

Error encountered when running Oracle Database Appliance commands

If you submit an odacli create-prepatchreport job, and then run any command before the job is completed, an error is encountered.

If you issue an odacli command such as odacli create-appliance, odacli update-dbhome, odacli update-server, odacli create-database, or odacli update-repository, while an odacli create-prepatchreport job is running, then you may see an error.

Hardware Models

All Oracle Database Appliance bare metal deployments

Workaround

Wait for the odacli create-prepatchreport job to complete before issuing any other odacli commands.

This issue is tracked with Oracle bug 30892528.

Parent topic: Known Issues When Patching Oracle Database Appliance

Error in updating Oracle ILOM when patching the appliance

When patching the appliance, there may be errors in patching Oracle ILOM.

When patching Oracle Database Appliance, patching of Oracle ILOM fails. The odacli describe-component command output does not display Oracle ILOM as up-to-date.

Hardware Models

All Oracle Database Appliance bare metal deployments

Workaround

Run the command ln -s /usr/bin/ipmiflash /usr/sbin/ipmiflash and then run the odacli update-server command again.

This issue is tracked with Oracle bug 30619842.

Parent topic: Known Issues When Patching Oracle Database Appliance

Patching pre-checks do not complete with --local option during server patching

Server patching fails while running patching pre-checks with the --local option.

Hardware Models

All Oracle Database Appliance hardware models

Workaround

Do not run patching pre-checks on the server with the --local option.

This issue is tracked with Oracle bug 30255817.

Parent topic: Known Issues When Patching Oracle Database Appliance

Relocation of Oracle RAC One Database fails during patching

When relocating Oracle RAC One Database during patching, an error is encountered.

When patching a database home in which one or more Oracle RAC One Databases are running, the relocation of the Oracle RAC One Database may fail. This causes the Oracle Database home patching to fail.

Hardware Models

All Oracle Database Appliance hardware models

Workaround

Shut down the Oracle RAC One node manually and then patch the database home. After patching completes successfully, start Oracle Database.

This issue is tracked with Oracle bug 30187542.

Parent topic: Known Issues When Patching Oracle Database Appliance

Error when patching to 12.1.0.2.190716 Bundle Patch

When patching Oracle Database release 12.1.0.2 to Oracle Database 12.1.0.2.190716 Bundle Patch, an error is encountered.

The ODACLI job displays the following error:

DCS-10001:Internal error encountered: Failed to run SQL script: datapatch script.

The data patch log contains the entry "Prereq check failed, exiting without installing any patches.".

Hardware Models

All Oracle Database Appliance hardware models bare metal deployments

Workaround

Install the same patch again.

This issue is tracked with Oracle bugs 30026438 and 30155710.

Parent topic: Known Issues When Patching Oracle Database Appliance

Server status not set to Normal when patching

When patching Oracle Database Appliance, an error is encountered.

When patching the appliance, the odacli update-server command fails with the following error: 

DCS-10001:Internal error encountered: Server upgrade state is not NORMAL node_name

Hardware Models

All Oracle Database Appliance hardware models

Workaround

  1. Run the command:
    /u01/app/18.0.0.0/grid//bin/cluvfy stage -post crsinst -collect cluster -gi_upgrade -n all
  2. Ignore the following two warnings:
    Verifying OCR Integrity ...WARNING
PRVG-6017 : OCR backup is located in the same disk group "+DATA" as OCR.

Verifying Single Client Access Name (SCAN) ...WARNING
RVG-11368 : A SCAN is recommended to resolve to "3" or more IP

  3. Run the command again till the output displays only the two warnings above. The status of Oracle Custerware status should be Normal again.

  4. You can verify the status with the command:
    /u01/app/18.0.0.0/grid/bin/crsctl query crs activeversion -f

This issue is tracked with Oracle bug 30099090.

Parent topic: Known Issues When Patching Oracle Database Appliance

Error in patching NVMe disks to the latest version

Patching of NVMe disks to the latest version may not be supported on some Oracle Database Appliance hardware models.

On Oracle Database Appliance X8-2 hardware models, the NVMe controller 7361456_ICRPC2DD2ORA6.4T is installed with higher version VDV1RL01/VDV1RL02. Patching of this controller is not supported on Oracle Database Appliance X8-2 hardware models. For other platforms, if the installed version is QDV1RE0F, or QDV1RE13, or QDV1RD09, or QDV1RE14 then when you patch the storage, the NVMe controller version is updated to qdv1rf30.

Hardware Models

All Oracle Database Appliance hardware models

Workaround

None

This issue is tracked with Oracle bug 30287439.

Parent topic: Known Issues When Patching Oracle Database Appliance

Failure in patching Oracle Database Appliance Virtualized Platform

Server patching for Oracle Database Appliance may fail with errors.

Patching the appliance server fails with the following error:

Worker 0: IOError: [Errno 28] No space left on device

This can occur during server patching. The space issue may occur either on ODA_BASE or dom0. The issue occurs when the log files opensm.log on dom0 and ibacm.log on ODA_BASE increase in size and consume all free space on the volume.

Hardware Models

Oracle Database Appliance hardware models X6-2 and X5-2 Virtualized Platform with InfiniBand

Workaround

Follow these steps:

  1. On ODA_BASE, truncate /var/log/opensm.log.
  2. On dom0, truncate /var/log/ibacm.log.
  3. Stop Oracle Clusterware:
    /u01/app/18.0.0.0/grid/bin/crsctl stop crs -f
  4. After the cluster and the cluster resources are stopped, start Oracle Clusterware:
    /u01/app/18.0.0.0/grid/bin/crsctl start crs

Restart Oracle Database Appliance server patching.

This issue is tracked with Oracle bug 30327847.

Parent topic: Known Issues When Patching Oracle Database Appliance

Error in patching Oracle Database Appliance Virtualized Platform

When applying the server patch for Oracle Database Appliance Virtualized Platform, an error is encountered.

Patching the appliance server fails with the following error:

ERROR: Host 192.168.16.28 listed in file /opt/oracle/oak/temp_privips.txt is not pingable at /opt/oracle/oak/pkgrepos/System/18.7.0.0.0/bin/pkg_install.pl 
line 1806 
ERROR: Unable to apply the patch 2

This can occur during non-local (rolling) server patch. The error is seen on the first node after patching of ODA_BASE and dom0 is complete. This issue is caused because the remote node Node1 rebooted during patching.

Hardware Models

Oracle Database Appliance hardware models X6-2 and X5-2 Virtualized Platform with InfiniBand

Workaround

  1. Shut down Oracle TFA Collector.
    /u01/app/18.0.0.0/grid/bin/tfactl stop

  2. Restart Oracle Database Appliance server patching.

This issue is tracked with Oracle bug 30318927.

Parent topic: Known Issues When Patching Oracle Database Appliance

Patching of M.2 drives not supported

Patching of M.2 drives (local disks disks SSDSCKJB48 and SSDSCKJB480G7) is not supported.

These drives are displayed when you run the odacli describe-component command. Patching of neither of the two known versions 0112 and 0121 of the M.2 disk is supported. Patching the LSI controller version 13.00.00.00 to version 16.00.01.00 is also not supported. However, on some Oracle Database Appliance X8-2 models, the installed LSI controller version may be 16.00.01.00.

Hardware Models

Oracle Database Appliance bare metal deployments

Workaround

None

This issue is tracked with Oracle bug 30249232.

Parent topic: Known Issues When Patching Oracle Database Appliance

DATA disk group fails to start after upgrading Oracle Grid Infrastructure to 18.5

After upgrading Oracle Grid Infrastructure to 18.5, the DATA disk group fails to start.

The following error is reported in the log file:

ORA-15038: disk '/dev/mapper/HDD_E1_S13_1931008292p1' mismatch on 'Sector
Size' with target disk group [512] [4096]

Hardware Models

Oracle Database Appliance hardware models X5-2 or later, with mixed storage disks installed

Workaround

To start Oracle Clusterware successfully, connect to Oracle ASM as grid user, and run the following SQL commands: 

SQL> show parameter _disk_sector_size_override; 

NAME                                 TYPE        VALUE 
-------------------------------------------------------- 
_disk_sector_size_override           boolean     TRUE 

SQL> alter system set "_disk_sector_size_override" = FALSE scope=both; 
alter system set "_disk_sector_size_override" = FALSE scope=both 
* 
ERROR at line 1: 
ORA-32000: write to SPFILE requested but SPFILE is not modifiable 

SQL> alter diskgroup DATA mount; 

Diskgroup altered. 

SQL> alter system set "_disk_sector_size_override" = FALSE scope=both; 

System altered

This issue is tracked with Oracle bug 29220984.

Parent topic: Known Issues When Patching Oracle Database Appliance

Errors when deleting database storage after migration to DCS stack

After migrating to the DCS stack, some volumes in the database storage cannot be deleted.

Create an Oracle ACFS database storage using the oakcli create dbstorage command for multitenant environment (CDB) without database in the OAK stack and then migrate to the DCS stack. When deleting the database storage, only the DATA volume is deleted, and not the REDO and RECO volumes.

Hardware Models

All Oracle Database Appliance high-availability hardware models for bare metal deployments

Workaround

Create a database on Oracle ACFS database storage with the same name as the database for which you want to delete the storage volumes, and then delete the database. This cleans up all the volumes and file systems.

This issue is tracked with Oracle bug 28987135.

Parent topic: Known Issues When Patching Oracle Database Appliance

Repository in offline or unknown status after patching

After rolling or local patching of both nodes to 18.8, repositories are in offline or unknown state on node 0 or 1.

The command oakcli start repo <reponame> fails with the error: 

OAKERR8038 The filesystem could not be exported as a crs resource  
OAKERR:5015 Start repo operation has been disabled by flag

Models

Oracle Database Appliance X7-2-HA, X6-2-HA, X5-2, X4-2, X3-2, and V1.

Workaround

Log in to oda_base of any node and run the following two commands: 

oakcli enable startrepo -node 0  
oakcli enable startrepo -node 1

The commands start the repositories and enable them to be available online.

This issue is tracked with Oracle bug 27539157.

Parent topic: Known Issues When Patching Oracle Database Appliance

Versions of some components not updated after cleaning up and reprovisioning Oracle Database Appliance

Oracle Auto Service Request (ASR), or Oracle TFA Collector, or Oracle ORAchk versions are not updated after cleaning up and reprovisioning Oracle Database Appliance.

When cleaning up and reprovisioning Oracle Database Appliance with release 18.8, the Oracle Auto Service Request (ASR), or Oracle TFA Collector, or Oracle ORAchk RPMs may not be updated to release 18.8. The components are updated when you apply the patches for Oracle Database Appliance release 18.8.

Hardware Models

All Oracle Database Appliance deployments

Workaround

Update to the latest server patch for the release.

This issue is tracked with Oracle bugs 28933900 and 30187516.

Parent topic: Known Issues When Patching Oracle Database Appliance

11.2.0.4 databases fail to start after patching

After patching Oracle Database Appliance to release 18.3, databases of version 11.2.0.4 fail to start.

Hardware Models

All Oracle Database Appliance Hardware models

Workaround

Databases of versions 11.2.0.4.170814 and 11.2.0.4.171017 must be manually started after patching to Oracle Database Appliance release 18.3.

Start the databases with the command: 
srvctl start database -db db_unique_name

This issue is tracked with Oracle bug 28815716.

Parent topic: Known Issues When Patching Oracle Database Appliance

FLASH disk group is not mounted when patching or provisioning the server

The FLASH disk group is not mounted after a reboot, including after provisioning, reimaging, or patching the server with Oracle Database Appliance 12.2.1.2.

This issue occurs when the node reboots and then you attempt to create an Oracle Automatic Storage Management Cluster File System (Oracle ACFS) database. When patching or provisioning a server with Oracle Database Appliance 12.2.1.2, you will encounter an SSH disconnect issue and an error.
# oakcli update -patch 12.2.1.2 --server

**************************************************************************** 
*****   For all X5-2 customers with 8TB disks, please make sure to     *****
*****   run storage patch ASAP to update the disk firmware to "PAG1".  *****
**************************************************************************** 
INFO: DB, ASM, Clusterware may be stopped during the patch if required 
INFO: Both Nodes may get rebooted automatically during the patch if required 
Do you want to continue: [Y/N]?: y 
INFO: User has confirmed for the reboot 
INFO: Patch bundle must be unpacked on the second Node also before applying the patch 
Did you unpack the patch bundle on the second Node? : [Y/N]? : y  
Please enter the 'root'  password :  
Please re-enter the 'root' password:  
INFO: Setting up the SSH 
..........Completed .....  
... ...
INFO: 2017-12-26 00:31:22: -----------------Patching ILOM & BIOS----------------- 
INFO: 2017-12-26 00:31:22: ILOM is already running with version 3.2.9.23r116695 
INFO: 2017-12-26 00:31:22: BIOS is already running with version 30110000 
INFO: 2017-12-26 00:31:22: ILOM and BIOS will not be updated  
INFO: 2017-12-26 00:31:22: Getting the SP Interconnect state... 
INFO: 2017-12-26 00:31:44: Clusterware is running on local node 
INFO: 2017-12-26 00:31:44: Attempting to stop clusterware and its resources locally 
Killed 
# Connection to server.example.com closed.

The Oracle High Availability Services, Cluster Ready Services, Cluster Synchronization Services, and Event Manager are online. However, when you attempt to create an Oracle Automatic Storage Management Cluster File System (Oracle ACFS) database, you receive an error: flash space is 0.

Hardware Models

Oracle Database Appliance X5-2, X6-2-HA, and X7-2 HA SSD systems.

Workaround

Manually mount FLASH disk group before creating an Oracle ACFS database.

Perform the following steps as the GRID owner:

  1. Set the environment variables as grid OS user:

    on node0 
export ORACLE_SID=+ASM1 
export ORACLE_HOME= /u01/app/12.2.0.1/grid

  2. Log on to the ASM instance as sysasm

    $ORACLE_HOME/bin/sqlplus / as sysasm

  3. Execute the following SQL command:

    SQL> ALTER DISKGROUP FLASH MOUNT

This issue is tracked with Oracle bug 27322213.

Parent topic: Known Issues When Patching Oracle Database Appliance

Known Issues When Deploying Oracle Database Appliance

Understand the known issues when provisioning or deploying Oracle Database Appliance.

Parent topic: Known Issues with Oracle Database Appliance in This Release

Incorrect RAID configuration of some Oracle Database Appliance X8-2 Systems

Oracle Database Appliance X8-2 Systems shipped before December 20, 2019 may have local system or boot disks with incorrect RAID configuration, and hence require reimaging the system.

Hardware Models

Oracle Database Appliance X8-2 Systems shipped before December 20, 2019

Workaround

See My Oracle Support Note 2622035.1 for more details.

https://support.oracle.com/rs?type=patch&id=2622035.1

This issue is tracked with Oracle bug 30651492.

Parent topic: Known Issues When Deploying Oracle Database Appliance

Failure in creating RECO disk group during provisioning

When provisioning Oracle Database Appliance X8-2-HA with High Performance configuration containing default storage and expansion shelf, creation of RECO disk group fails.

Hardware Models

All Oracle Database Appliance X8-2-HA with High Performance configuration

Workaround

  1. Power off storage expansion shelf.
  2. Reboot both nodes.
  3. Proceed with provisioning the default storage shelf (first JBOD).
  4. After the system is successfully provisioned with default storage shelf (first JBOD), check that oakd is running on both nodes in foreground mode.
     # ps -aef | grep oakd
  5. Check that all first JBOD disks have the status online, good in oakd, and CACHED in Oracle ASM.
  6. Power on the storage expansion shelf (second JBOD), wait for a few minutes for the oeprating system and other subsystems to recognize it.
  7. Run the following command from the master node to add the storage expansion shelf disks (two JBOD setup) to oakd and Oracle ASM. 
    #odaadmcli show ismaster 
      OAKD is in Master Mode 

      # odaadmcli expand storage -ndisk 24 -enclosure 1 
       Skipping precheck for enclosure '1'... 
       Check the progress of expansion of storage by executing 'odaadmcli  
show disk' 
       Waiting for expansion to finish ... 
      #
  8. Check that the storage expansion shelf disks (two JBOD setup) are added to oakd and Oracle ASM.

Replace odaadmcli with oakcli commands on Oracle Database Appliance Virtualized Platform in the procedure.

For more information, see the chapter Managing Storage in the Oracle Database Appliance X8-2 Deployment Guide.

This issue is tracked with Oracle bug 30839054.

Parent topic: Known Issues When Deploying Oracle Database Appliance

Simultaneous creation of two Oracle ACFS Databases fails

If you try to create two Oracle ACFS databases on a system where there is no database or database storage already created, then database creation fails for one of the databases with an error.

DCS-10001:Internal error encountered: Fail to run command Failed to create  
volume.

Hardware Models

All Oracle Database Appliance bare metal deployments

Workaround

Manually delete the DATA volume (and REDO volume, in case of Oracle Database Appliance X8-2) from the system.

For High Perfomance configuration, run the following commands:
su - GRID_USER 
export ORACLE_SID=+ASM1(in case of first node) /+ASM2(in case of second 
node); 
export ORACLE_HOME=GRID_HOME; 
GRID_HOME/bin/asmcmd --nocp voldelete -G Data datdbname
For Oracle Database Appliance X8-2 High Perfomance configuration, remove the REDO volume as follows:
su - GRID_USER 
export ORACLE_SID=+ASM1(in case of first node) /+ASM2(in case of second 
node); 
export ORACLE_HOME=GRID_HOME; 
GRID_HOME/bin/asmcmd --nocp voldelete -G Reco rdodbname
For High Capacity configuration, run the following commands:
su - GRID_USER 
export ORACLE_SID=+ASM1(in case of first node) /+ASM2(in case of second 
node); 
export ORACLE_HOME=GRID_HOME; 
GRID_HOME/bin/asmcmd --nocp voldelete -G Flash datdbname (if volume exists in FLASH disk group)
GRID_HOME/bin/asmcmd --nocp voldelete -G data datdbname (if volume exists in DATA disk group)
For Oracle Database Appliance X8-2 High Capacity configuration, remove the REDO volume as follows:
su - GRID_USER 
export ORACLE_SID=+ASM1(in case of first node) /+ASM2(in case of second 
node); 
export ORACLE_HOME=GRID_HOME; 
GRID_HOME/bin/asmcmd --nocp voldelete -G Flash rdodbname

This issue is tracked with Oracle bug 30750497.

Parent topic: Known Issues When Deploying Oracle Database Appliance

Shared repositories are not online after patching Virtualized Platform

The oakcli show repo command displays only the local repository or the shared repository status is unknown.

Hardware Models

All Oracle Database Appliance Virtualized Platform

Workaround

After patching Oracle Database Appliance Virtualized Platform, run the following commands to reenable the shared repository:
# oakcli enable startrepo -node 0 
# oakcli enable startrepo -node 1 
# oakcl restart oak

This issue is tracked with Oracle bug 30325619.

Parent topic: Known Issues When Deploying Oracle Database Appliance

Creation of CDB for 12.1.0.2 databases may fail

Creation of multitenant container database (CDB) for 12.1.0.2 databases on Virtualized Platform may fail.

If the database name (db_name) and database unique name (db_unique_name) are different when creating snahsot database, then the following error is encountered: 

WARNING: 2018-09-13 12:47:18: Following data files are not on SNAP location

Hardware Models

All Oracle Database Appliance hardware models for Virtualized Platform

Workaround

None.

This issue is tracked with Oracle bug 29231958.

Parent topic: Known Issues When Deploying Oracle Database Appliance

Database creation hangs when using a deleted database name for database creation

The accelerator volume for data is not created on flash storage, for database created during provisioning of appliance.

If you delete a 11.2.0.4 database, and then create a new database with same name as the deleted database, database creation hangs while unlocking the DBSNMP user for the database.

Hardware Models

All Oracle Database Appliance high-availability environments

Workaround

Before creating the 11.2.0.4 database with the same name as the deleted database, delete the DBSNMP user, if the user exists.

For example, the following command creates a database testdb with user DBSNMP. 

/u01/app/18.0.0.0/grid/bin/crsctl delete wallet -type CVUDB -name testdb -user DBSNMP

This issue is tracked with Oracle bug 28916487.

Parent topic: Known Issues When Deploying Oracle Database Appliance

Error encountered after running cleanup.pl

Errors encountered in running odacli commands after running cleanup.pl.

After running cleanup.pl, when you try to use odacli commands, the following error is encountered: 

DCS-10042:User oda-cliadmin cannot be authorized.

Hardware Models

All Oracle Database Appliance hardware models for bare metal deployments

Workaround

Run the following commands to set up the credentials for the user oda-cliadmin on the agent wallet: 

# rm -rf /opt/oracle/dcs/conf/.authconfig  
# /opt/oracle/dcs/bin/setupAgentAuth.sh

This issue is tracked with Oracle bug 29038717.

Parent topic: Known Issues When Deploying Oracle Database Appliance

Accelerator volume for data is not created on flash storage

The accelerator volume for data is not created on flash storage, for databases created during provisioning of appliance.

Hardware Models

Oracle Database Appliance high capacity environments with HDD disks

Workaround

Do not create the database when provisioning the appliance. This creates all required disk groups, including flash. After provisioning the appliance, create the database. The accelerator volume is then created.

This issue is tracked with Oracle bug 28836461.

Parent topic: Known Issues When Deploying Oracle Database Appliance

Errors in clone database operation

Clone database operation fails due to errors.

If the source database is single-instance or Oracle RAC One Node, or running on the remote node, the clone database operation fails, because the paths are not created correctly in the control file.

Clone database operation may also fail with errors if the source database creation time stamp is too close to the clone operation (at least within 60 minutes).

Hardware Models

All Oracle Database Appliance high-availability hardware models for bare metal deployments

Workaround

Create the clone database from the source database instance that is running on the same node from which the clone database creation is triggered.

For Oracle Database 12c and later, synchronize the source database before the clone operation, by running the command:
SQL> alter system checkpoint;

This issue is tracked with Oracle bugs 29002563, 29002004, 29001906, 29001855, 29001631, 28995153, 28986643, 30309971, and 30228362.

Parent topic: Known Issues When Deploying Oracle Database Appliance

Clone database operation fails

For Oracle Database release 12.1 databases, the database clone creation may fail because the default compatible version from Oracle binaries was set to 12.0.0.0.0

Hardware Models

All Oracle Database Appliance high-availability hardware models for bare metal deployments

Workaround

Set the compatible value to that of the source database. Follow these steps:
  1. Change the parameter value.
    SQL> ALTER SYSTEM SET COMPATIBLE = '12.1.0.2.0' SCOPE=SPFILE;
  2. Shut down the database.
    SQL> SHUTDOWN IMMEDIATE
  3. Start the database.
    SQL> Startup
  4. Verify the parameter for the new value.
    SQL> SELECT name, value, description FROM v$parameter WHERE name ='compatible';

This issue is tracked with Oracle bug 30309914.

Parent topic: Known Issues When Deploying Oracle Database Appliance

Errors after restarting CRS

If the Cluster Ready Services (CRS) are stopped or restarted, before stopping the repository and virtual machines, then this may cause errors.

Repository status is unknown and High Availability Virtual IP is offline if the Cluster Ready Services (CRS) are stopped or restarted before stopping the repository and virtual machines.

Hardware Models

Oracle Database Appliance HA models X7-2-HA, X6-2-HA, X5-2, X4-2, X3-2, V1

Workaround

Follow these steps:

  1. Start the High Availability Virtual IP on node1. 
    # /u01/app/GI_version/grid/bin/srvctl start havip -id havip_0

  2. Stop the oakVmAgent.py process on dom0.

  3. Run the lazy unmount option on the dom0 repository mounts:
    umount -l mount_points

This issue is tracked with Oracle bug 20461930.

Parent topic: Known Issues When Deploying Oracle Database Appliance

Unable to create an Oracle ASM Database for Release 12.1

Known issues with Oracle Automatic Storage Management (Oracle ASM) are preventing the REDO diskgroup from mounting for Oracle Database Release 12.1.

Unable to create an Oracle ASM database lower than 12.1.0.2.17814 PSU (12.1.2.12).

Hardware Models

Oracle Database Appliance X6-2-HA, X5-2, X4-2, X3-2, and V1.

Workaround

There is not a workaround. If you have Oracle Database 11.2 or 12.1 that is using Oracle Automatic Storage Management (Oracle ASM) and you want to upgrade to a higher release of Oracle Database, then you must be on at least Oracle Database Appliance 12.1.2.12.0 and Database Home 12.1.0.2.170814.

The upgrade path for Oracle Database 11.2 or 12.1 Oracle ASM is as follows:

  • If you are on Oracle Database Appliance version 12.1.2.6.0 or later, then upgrade to 12.1.2.12 or higher before upgrading your database.

  • If you are on Oracle Database Appliance version 12.1.2.5 or earlier, then upgrade to 12.1.2.6.0, and then upgrade again to 12.1.2.12 or higher before upgrading your database.

This issue is tracked with Oracle bug 21626377, 27682997, 27250552, and 21780146. The issues are fixed in Oracle Database 12.1.0.2.170814.

Parent topic: Known Issues When Deploying Oracle Database Appliance

Database creation fails for odb-01s DSS databases

When attempting to create an DSS database with shape odb-01s, the job may fail with errors.

CRS-2674: Start of 'ora.test.db' on 'example_node' failed
CRS-5017: The resource action "ora.test.db start" encountered the following
error:
ORA-03113: end-of-file on communication channel
Process ID: 0
Session ID: 0 Serial number: 0
. For details refer to "(:CLSN00107:)" in
"/u01/app/grid/diag/crs/example_node/crs/trace/crsd_oraagent_oracle.trc".

Hardware Models

Oracle Database Appliance X6-2-HA, X5-2, X4-2, X3-2, and V1

Workaround

There is no workaround. Select an alternate shape to create the database.

This issue is tracked with Oracle bug 27768012.

Parent topic: Known Issues When Deploying Oracle Database Appliance

Known Issues When Managing Oracle Database Appliance

Understand the known issues when managing or administering Oracle Database Appliance.

Parent topic: Known Issues with Oracle Database Appliance in This Release

Error when expanding storage on Oracle Database Appliance Virtualized Platform

When you run the oakcli expand storage command on Oracle Database Appliance Virtualized Platforms, an error is encountered.

The following error is encountered when running the oakcli expand storage command on Oracle Database Appliance Virtualized Platforms.
sh: /opt/oracle/oak/bin/odaadmcli: No such file or directory
For example, following is the error when you expand storage with 5 disks:
# oakcli expand storage -ndisk 5 -enclosure 0
sh: /opt/oracle/oak/bin/odaadmcli: No such file or directory
sh: /opt/oracle/oak/bin/odaadmcli: No such file or directory
sh: /opt/oracle/oak/bin/odaadmcli: No such file or directory
sh: /opt/oracle/oak/bin/odaadmcli: No such file or directory
sh: /opt/oracle/oak/bin/odaadmcli: No such file or directory
The disk(s) [e0_pd_00][e0_pd_01][e0_pd_02][e0_pd_03][e0_pd_04] is/are not
online in oakd
Aborting ...

This error occurs in Oracle Database Appliance Virtualized Platforms with Oracle Autonomous Health Framework version 19.3, where Oracle Autonomous Health Framework is installed in the DCS home directory (/opt/oracle/dcs) under /opt/oracle/dcs/oracle.ahf/.

You can check your AHF version:
# tfactl -version;orachk -v
TFA Version : 193000
TFA Build ID : 20200108023845
ORACHK  VERSION: 19.3.0_20200108
# cd /opt/oracle
# ls
dcs  extapi  oak
# cd dcs
# ls
oracle.ahf
#

Hardware Models

Oracle Database Appliance High-Availability hardware models Virtualized Platform deployments

Workaround

Uninstall the installed Oracle Autonomous Health Framework version and download the latest Oracle Autonomous Health Framework version on both nodes of your Oracle Database Appliance High-Availability deployment.

Run the following steps on both nodes:
  1. Uninstall Oracle Autonomous Health Framework.
    # tfactl uninstall -local -deleterepo

    For Oracle Autonomous Health Framework version 19.3, manually remove the installed version, by running the following command:

    # rpm -e oracle-ahf
  2. Check that no oracle-ahf rpm is installed on the system.
    # rpm -qi oracle-ahf
package oracle-ahf is not installed
#
  3. Remove the DCS home (/opt/oracle/dcs) and its sub-directories.
    # cd /opt/oracle
# rm -rf /opt/oracle/dcs
  4. Check and ensure the DCS home directory (/opt/oracle/dcs) does not exist under the /opt/oracle directory on the system.
     # cd /opt/oracle/
 # ls
 extapi  oak
 #
  5. Download the latest Oracle Autonomous Health Framework version from My Oracle Support.
  6. Copy the downloaded zip file to /root on your system, for example, AHF-LINUX_v20.1.1.zip. The zip file name for each Oracle Autonomous Health Framework version may change based on the latest release number.
     # /root/AHF-LINUX_v20.1.1.zip
  7. Create the ahfinstall directory under /tmp.
    # cd /tmp
# mkdir ahfinstall
  8. Unzip the Oracle Autonomous Health Framework zip file into /tmp/ahfinstall.
    # cd ahfinstall
# unzip /root/AHF-LINUX_v20.1.1.zip
  9. Run the ahf_setup installer from /tmp/ahfinstall. During installation, select the default location for installation.

    Note: Do not choose the DCS home directory as the default Oracle Autonomous Health Framework location for install. 

     # ./ahf_setup
  10. Check the Oracle TFA and ORAchk versions.
    #tfactl -version;orachk -v
TFA Version : 201100
TFA Build ID : 20200331131556
ORACHK  VERSION: 20.1.1_20200331
#
  11. Check that Oracle Autonomous Health Framework is installed in the /opt/oracle.ahf directory.
Example of Running the Workaround
#  tfactl -version;orachk -v
TFA Version : 193000
TFA Build ID : 20200108023845
ORACHK  VERSION: 19.3.0_20200108

# tfactl uninstall -local -deleterepo
Starting AHF Uninstall
AHF will be uninstalled on: node2

Do you want to continue with AHF uninstall ? [Y]|N : y

Stopping AHF service on local node node2...
Sleeping for 10 seconds...

Stopping TFA Support Tools...
TFA-00002 Oracle Trace File Analyzer (TFA) is not running
Stopping orachk scheduler ...
Removing orachk cache discovery....
Successfully completed orachk cache discovery removal.
Removed orachk from inittab
Removing AHF setup on node2:
Removing /etc/rc.d/rc0.d/K17init.tfa
Removing /etc/rc.d/rc1.d/K17init.tfa
Removing /etc/rc.d/rc2.d/K17init.tfa
Removing /etc/rc.d/rc4.d/K17init.tfa
Removing /etc/rc.d/rc6.d/K17init.tfa
Removing /etc/init.d/init.tfa...
Removing /opt/oracle.ahf/jre
Removing /opt/oracle.ahf/common
Removing /opt/oracle.ahf/bin
Removing /opt/oracle.ahf/python
Removing /opt/oracle.ahf/analyzer
Removing /opt/oracle.ahf/tfa
Removing /opt/oracle.ahf/orachk
Removing /opt/oracle.ahf/ahf
Removing /u01/app/grid/oracle.ahf/data/node2
Removing /opt/oracle.ahf/install.properties
Removing /u01/app/grid/oracle.ahf/data/repository
Removing /u01/app/grid/oracle.ahf/data
Removing /u01/app/grid/oracle.ahf
Removing AHF Home : /opt/oracle.ahf
# tfactl -version;orachk -v
-bash: /usr/bin/tfactl: No such file or directory
-bash: /usr/bin/orachk: No such file or directory
# rpm -e oracle-ahf
warning:    erase unlink of /opt/oracle.ahf failed: No such file or directory
# rpm -qi oracle-ahf
package oracle-ahf is not installed
#
# cd /opt/oracle
# ls
dcs  extapi  oak
# rm -rf dcs/
# ls
extapi  oak
#
#cd /tmp
#mkdir ahfinstall
#cd ahfinstall/
# pwd
/tmp/ahfinstall
# unzip /root/AHF-LINUX_v20.1.1.zip
Archive:  /root/AHF-LINUX_v20.1.1.zip
  inflating: README.txt            
  inflating: ahf_setup
 

# ls
ahf_setup  README.txt
# ./ahf_setup

AHF Installer for Platform Linux Architecture x86_64
AHF Installation Log : /tmp/ahf_install_85684_2020_04_01-00_15_38.log
Starting Autonomous Health Framework (AHF) Installation
AHF Version: 20.1.1 Build Date: 202003311315
Default AHF Location : /opt/oracle.ahf
Do you want to install AHF at [/opt/oracle.ahf] ? [Y]|N : y
AHF Location : /opt/oracle.ahf
AHF Data Directory stores diagnostic collections and metadata.
AHF Data Directory requires at least 5GB (Recommended 10GB) of free space.
Choose Data Directory from below options :

1. /u01/app/grid [Free Space : 51458 MB]
2. Enter a different Location

Choose Option [1 - 2] : 1

AHF Data Directory : /u01/app/grid/oracle.ahf/data

Do you want to add AHF Notification Email IDs ? [Y]|N :   xxxxxxxxxxxx

AHF will also be installed/upgraded on these Cluster Nodes :

1. node1

The AHF Location and AHF Data Directory must exist on the above nodes
AHF Location : /opt/oracle.ahf
AHF Data Directory : /u01/app/grid/oracle.ahf/data

Do you want to install/upgrade AHF on Cluster Nodes ? [Y]|N : N
Extracting AHF to /opt/oracle.ahf
Configuring TFA Services
Discovering Nodes and Oracle Resources
Not generating certificates as GI discovered
Starting TFA Services

.----------------------------------------------------------------------------------.
| Host          | Status of TFA | PID   | Port | Version    | Build ID             |
+---------------+---------------+-------+------+------------+----------------------+
| node2 | RUNNING       | 87939 | 5000 | 20.1.1.0.0 | 20110020200331131556 |
'---------------+---------------+-------+------+------------+----------------------'

Running TFA Inventory...

Adding default users to TFA Access list...

.--------------------------------------------------------------------.
|                    Summary of AHF Configuration                    |
+-----------------+--------------------------------------------------+
| Parameter       | Value                                            |
+-----------------+--------------------------------------------------+
| AHF Location    | /opt/oracle.ahf                                  |
| TFA Location    | /opt/oracle.ahf/tfa                              |
| Orachk Location | /opt/oracle.ahf/orachk                           |
| Data Directory  | /u01/app/grid/oracle.ahf/data                    |
| Repository      | /u01/app/grid/oracle.ahf/data/repository         |
| Diag Directory  | /u01/app/grid/oracle.ahf/data/node2/diag |
'-----------------+--------------------------------------------------'


Starting orachk daemon from AHF ...
AHF install completed on node1.
Installing AHF on Remote Nodes :
AHF will be installed on node1, Please wait.
Installing AHF on node1:
[node1] Copying AHF Installer
[node1] Running AHF Installer
Adding rpm Metadata to rpm database on ODA system
RPM File /opt/oracle.ahf/rpms/oracle-ahf-201100-20200331131556.x86_64.rpm
Preparing...                ########################################### [100%]
---------

   1:oracle-ahf             ########################################### [100%]
Upgrading oracle-ahf
warning:    erase unlink of /opt/oracle/dcs/oracle.ahf failed: No such file or directory

AHF binaries are available in /opt/oracle.ahf/bin
AHF is successfully installed
Moving /tmp/ahf_install_85684_2020_04_01-00_15_38.log to /u01/app/grid/oracle.ahf/data/node2/diag/ahf/

# tfactl -version;orachk -v
TFA Version : 201100
TFA Build ID : 20200331131556
ORACHK  VERSION: 20.1.1_20200331

# ls -l /opt/
........
drwxr-xr-x 11 root root 4096 Mar 31 13:16 oracle.ahf
........
#

This issue is tracked with Oracle bug 31014517.

Parent topic: Known Issues When Managing Oracle Database Appliance

Inconsistency in available and current system firmware

The current system firmware may be different from the available firmware after applying the latest patch.

Oracle Database Appliance X8-2 with expander model ORACLE/DE3-24C are at version 0309 but patching of expander firmware from earlier versions to 0309 is not supported in this release. Oracle Database Appliance Release 18.8 contains the patch for expander version 0306, so when you run odacli describe-component command, the available expander version is displayed as 0306.

Oracle Database Appliance X8-2 with controller model LSI Logic/0x0097 are at version 16.00.00.00 but patching of controller firmware from earlier versions to 16.00.00.00 is not supported in this release. Oracle Database Appliance Release 18.8 contains the patch for controller version 13.00.00.00, so when you run odacli describe-component command, the available expander version is displayed as 13.00.00.00.

Hardware Models

Oracle Database Appliance X8-2 hardware models

Workaround

Ignore this inconsistency, since this is a display issue and does not affect the installed firmware version.

This issue is tracked with Oracle bug 30787910.

Parent topic: Known Issues When Managing Oracle Database Appliance

DCS logs not collected by the odaadmcli manage diagcollect command

By default, the DCS logs are not collected when you run the odaadmcli manage diagcollect command.

Hardware Models

Oracle Database Appliance hardware models bare metal deployments

Workaround

Use the CLI command odaadmcli manage diagcollect --components dcs to collect DCS logs.

This issue is tracked with Oracle bug 30760941.

Parent topic: Known Issues When Managing Oracle Database Appliance

Inconsistency in ORAchk summary and details report page

ORAChk report summary on the Browser User Interface may show different counts of Critical, Failed, and Warning issues than the report detail page.

Hardware Models

Oracle Database Appliance hardware models bare metal deployments

Workaround

Ignore counts of Critical, Failed, and Warning issues in the ORAchk report summary on the Browser User Interface. Check the report detail page.

This issue is tracked with Oracle bug 30676674.

Parent topic: Known Issues When Managing Oracle Database Appliance

Error in attaching vdisk to guest VM

The current system firmware may be different from the available firmware after applying the latest patch.

When multiple vdisks from the oda_base driver_domain are attached to the guest VM, their entries are not written on the xenstore, vdisks are not attached to the VM, and the VM may not start.

The following errors are logged on xen-hotplug.log in ODA_BASE:
xenstore-write: could not write path backend/vbd/6/51728/node 
xenstore-write: could not write path backend/vbd/6/51728/hotplug-error

Hardware Models

Oracle Database Appliance Virtualized Platform

Workaround

  1. Add the followng into the /etc/sysconfig/xencommons file in dom0:
    XENSTORED_ARGS="--entry-nb=4096 --transaction=512"
  2. Reboot dom0.

This issue is tracked with Oracle bug 30886365.

Parent topic: Known Issues When Managing Oracle Database Appliance

Extensive tracing generated for server processes

Extensive tracing files for the server processes are generated with DRM messages.

2019-08-07 03:35:33.498*:example1():   
[0x3fc1001c][0xf02],[TX][ext0x0,0x0][domid 0x0] 
  maxnodes 16, key 2663540594, node 2 (inst 3), member_node 0 
2019-08-07 03:35:33.498*:example1():   delta 15 
2019-08-07 03:35:33.498*:example2():   
[0x3fc1001c][0xf11],[TX][ext0x0,0x0][domid 0x0] 
  maxnodes 16, key 2663540609, node 1 (inst 2), member_node 1

Hardware Models

All Oracle Database Appliance hardware models

Workaround

Disable tracing:
alter system set event='trace [rac_enq] disk disable' scope=spfile;

This issue is tracked with Oracle bug 30166512.

Parent topic: Known Issues When Managing Oracle Database Appliance

Missing DATA, RECO, and REDO entries when dbstorage is rediscovered

Running the odacli update-registry command with -n all --force or -n dbstorage --force option can result in metadata corruption.

Hardware Models

All Oracle Database Appliance hardware models bare metal deployments

Workaround

Run the -all option when all the databases created in the system use OAKCLI in migrated systems. On other systems that run on DCS stack, update all components other than dbstorage individually, using the odacli update-registry -n component_name_to_be_updated_excluding_dbstorage.

This issue is tracked with Oracle bug 30274477.

Parent topic: Known Issues When Managing Oracle Database Appliance

Incorrect Aura8 firmware value displayed

The Aura8 firmware version displayed in the components list is incorrect.

Models

Oracle Database Appliance X8-2S and X8-2M

Workaround

None.

This issue is tracked with Oracle bug 30340410.

Parent topic: Known Issues When Managing Oracle Database Appliance

iRestore, recovery, and update operations on a database fail

iRestore, recovery, and update operations on a database fail, if the ObjectStore Container used by the database already has a copy.

Hardware Models

All Oracle Database Appliance hardware models

Workaround

When performing iRestore, recovery, and update operations on a database, ensure that files are not copied to the ObjectStore Container.

This issue is tracked with Oracle bug 30529607.

Parent topic: Known Issues When Managing Oracle Database Appliance

ODA_BASE is in read-only mode or cannot start

The /OVS directory is full and ODA_BASE is in read-only mode.

The vmcore file in the /OVS/ var directory can cause the /OVS directory (Dom 0) to become 100% used. When Dom 0 is full, ODA_BASE is in read-only mode or cannot start.

Hardware Models

Oracle Database Appliance X6-2-HA, X5-2, X4-2, X3-2, and V1.

Oracle Database Appliance X7-2-HA Virtualized Platform.

Workaround

Perform the following to correct or prevent this issue:

  • Periodically check the file usage on Dom 0 and clean up the vmcore file, as needed.

  • Edit the oda_base vm.cfg file and change the on_crash = 'coredump-restart' parameter to on_crash = 'restart'. Especially when ODA_BASE is using more than 200 GB (gigabytes) of memory.

This issue is tracked with Oracle bug 26121450.

Parent topic: Known Issues When Managing Oracle Database Appliance

The odaeraser tool does not work if oakd is running in non-cluster mode

After cleaning up the deployment, the Secure Eraser tool does not work if oakd is running in non-cluster mode.

Hardware Models

All Oracle Database Appliance Hardware bare metal systems

Workaround

After cleanup of the deployment, oakd is started in the non-cluster mode, and it cannot be stopped using "odaadmcli stop oak" command. In such a case, if the Secure Erase tool is run, then the odaeraser command fails.

Use the command odaadmcli shutdown oak to stop oakd.

This issue is tracked with Oracle bug 28547433.

Parent topic: Known Issues When Managing Oracle Database Appliance

Issues with the Web Console on Microsoft web browsers

Oracle Database Appliance Web Console has issues on Microsoft Edge and Microsoft Internet Explorer web browsers.

Following are issues with Microsoft web browsers:
  • Oracle Database Appliance Web Console does not display correctly on Microsoft Edge and Microsoft Internet Explorer web browsers.
  • Advanced Information for the appliance does not display on Microsoft Internet Explorer web browser.
  • Job activity status does not refresh in the Web Console on Microsoft Internet Explorer web browser.
  • After configuring the oda-admin password, the following error is displayed:
    Failed to change the default user (oda-admin) account password. 
Status Code: 500 DCS-10001: DCS-10001:Internal error encountered: User not authorized

    Workaround: Close the Microsoft Internet Explorer browser session and open another browser session.

Models

All Oracle Database Appliance Hardware Models bare metal deployments

Workaround

To access the Web Console, use either Google Chrome or Firefox.

This issue is tracked with Oracle bugs 30077007, 30099089, and 29887027.

Parent topic: Known Issues When Managing Oracle Database Appliance

Disk space issues due to Zookeeper logs size

The Zookeeper log files, zookeeper.out and /opt/zookeeper/log/zkMonitor.log, are not rotated, when new logs are added. This can cause disk space issues.

Hardware Models

All Oracle Database Appliance hardware models for bare metal deployments

Workaround

Rotate the zookeeper log file manually, if the log file size increases, as follows:

  1. Stop the DCS-agent service for zookeeper on both nodes. 

    initctl stop initdcsagent

  2. Stop the zookeeper service on both nodes.

    /opt/zookeeper/bin/zkServer.sh stop

  3. Clean the zookeeper logs after taking the backup, by manually deleting the existing file or by following steps 4 to 10.

  4. Set the ZOO_LOG_DIR as an environment variable to a different log directory, before starting the zookeeper server. 

    export ZOO_LOG_DIR=/opt/zookeeper/log

  5. Switch to ROLLINGFILE, to set the capability to roll. 

    export ZOO_LOG4J_PROP="INFO, ROLLINGFILE"
    Restart the zookeeper server, for the changes to take effect.

  6. Set the following parameters in the /opt/zookeeper/conf/log4j.properties file, to limit the number of backup files, and the file sizes. 

    zookeeper.log.dir=/opt/zookeeper/log
zookeeper.log.file=zookeeper.out
log4j.appender.ROLLINGFILE.MaxFileSize=10MB
log4j.appender.ROLLINGFILE.MaxBackupIndex=10

  7. Start zookeeper on both nodes.

    /opt/zookeeper/bin/zkServer.sh start

  8. Check the zookeeper status, and verify that zookeeper runs in leader/follower/standalone mode. 

    /opt/zookeeper/bin/zkServer.sh status
ZooKeeper JMX enabled by default
Using config: /opt/zookeeper/bin/../conf/zoo.cfg
Mode: follower

  9. Start the dcs agent on both nodes.

    initctl start initdcsagent

  10. Purge the zookeeper monitor log, zkMonitor.log, in the location /opt/zookeeper/log. You do not have to stop the zookeeper service.

This issue is tracked with Oracle bug 29033812.

Parent topic: Known Issues When Managing Oracle Database Appliance

Error after running the cleanup script

After running the cleanup.pl script, the following error message appears: DCS-10001:Internal error encountered: Fail to start hand shake.

The error is causes when you run the following steps:

  1. Run cleanup.pl on the first node (Node0). Wait until the cleanup script finishes, then reboot the node.

  2. Run cleanup.pl on the second node (Node1). Wait until the cleanup script finishes, then reboot the node.

  3. After both nodes are started, use the command-line interface to list the jobs on Node0. An internal error appears.

    # odacli list-jobs
DCS-10001:Internal error encountered: Fail to start hand shake to localhost:7070

Hardware Models

Oracle Database Appliance X7-2-HA

Workaround

  1. Verify the zookeeper status on the both nodes before starting dcsagent: 

    /opt/zookeeper/bin/zkServer.sh status

    For a single-node environment, the status should be: leader, or follower, or standalone.

  2. Restart the dcsagent on Node0 after running the cleanup.pl script. 

    # initctl stop initdcsagent 
# initctl start initdcsagent

Parent topic: Known Issues When Managing Oracle Database Appliance

Old configuration details persisting in custom environment

The configuration file /etc/security/limits.conf contains default entries even in the case of custom environments.

On custom environments, when a single user is configured for both grid and oracle, the default grid user entries for the image are not removed from the /etc/security/limits.conf file.

Models

Oracle Database Appliance X7-2-HA, X7-2S, and X7-2M

Workaround

This issue does not affect the functionality. Manually edit the /etc/security/limits.conf file and remove invalid entries.

This issue is tracked with Oracle bug 26978354.

Parent topic: Known Issues When Managing Oracle Database Appliance

Incorrect SGA and PGA values displayed

For online transaction processing (OLTP), In-Memory (IMDB), and decision support services (DSS) databases created with odb36 database shape, the PGA and SGA values are displayed incorrectly.

For OLTP databases created with odb36 shape, following are the issues:

  • sga_target is set as 128 GB instead of 144 GB

  • pga_aggregate_target is set as 64 GB instead of 72 GB

For DSS databases created with with odb36 shape, following are the issues:

  • sga_target is set as 64 GB instead of 72 GB

  • pga_aggregate_target is set as 128 GB instead of 144 GB

For IMDB databases created with Odb36 shape, following are the issues:

  • sga_target is set as 128 GB instead of 144 GB

  • pga_aggregate_target is set as 64 GB instead of 72 GB

  • inmmory_size is set as 64 GB instead of 72 GB

Models

Oracle Database Appliance X7-2-HA, X7-2S, and X7-2M

Workaround

Reset the PGA and SGA sizes manually

This issue is tracked with Oracle bug 27036374.

Parent topic: Known Issues When Managing Oracle Database Appliance

Error in node number information when running network CLI commands

Network information for node0 is always displayed for some odacli commands, when the -u option is not specified.

If the -u option is not provided, then the describe-networkinterface, list-networks and the describe-network odacli commands always display the results for node0 (the default node), irrespective of whether the command is run from node0 or node1.

Hardware Models

Oracle Database Appliance X7-2-HA, X6-2-HA, X5-2, X4-2, X3-2, and V1

Workaround

Specify the -u option in the odacli command, for details about the current node.

This issue is tracked with Oracle bug 27251239.

Parent topic: Known Issues When Managing Oracle Database Appliance

Unrecognized Token Messages Appear in /var/log/messages

After updating Oracle Database Appliance, unrecognized token messages appear in /var/log/messages.

Updating to Oracle Database Appliance 12.1.2.11.0 updates the Oracle VM Server version to 3.4.3. After updating, the following messages appear in /var/log/messages: 

Unrecognized token: "max_seq_redisc"
Unrecognized token: "rereg_on_guid_migr"
Unrecognized token: "aguid_inout_notice"
Unrecognized token: "sm_assign_guid_func"
Unrecognized token: "reports"
Unrecognized token: "per_module_logging"
Unrecognized token: "consolidate_ipv4_mask"

You can ignore the messages for these parameters, they do not impact the InfiniBand compliant Subnet Manager and Administration (opensm) functionality. However, Oracle recommends removing the parameters to avoid flooding /var/log/messages.

Hardware Models

Oracle Database Appliance X6-2-HA and X5-2 with InfiniBand

Workaround

Perform the following to remove the parameters:

  1. After patching, update the /etc/opensm/opensm.conf file in bare metal deployments and in Dom0 in virtualized platform environment to remove the parameters. 

    cat /etc/opensm/opensm.conf  | egrep -w
'max_seq_redisc|rereg_on_guid_migr|aguid_inout_notice|sm_assign_guid_func|repo
rts|per_module_logging|consolidate_ipv4_mask' | grep -v ^#
max_seq_redisc 0
rereg_on_guid_migr FALSE
aguid_inout_notice FALSE
sm_assign_guid_func uniq_count
reports 2
per_module_logging FALSE
consolidate_ipv4_mask 0xFFFFFFFF

  2. Reboot. The messages will not appear after rebooting the node.

This issue is tracked with Oracle bug 25985258.

Parent topic: Known Issues When Managing Oracle Database Appliance