A.5.1 Upgrade Oracle Guest

The procedures in this section perform the IDIH Oracle Guest upgrade.

A.5.1 IDIH CLI: Perform a System Health Check on the Oracle Guest

1. Log in to the Oracle guest as the admusr user.

   ssh <IDIH IP address>

   login as: admusr

   password: <enter password>

2. Execute the analyze_server.sh script.

   $ sudo /usr/TKLC/xIH/plat/bin/analyze_server.sh -i

   Sample output:

   [admusr@cat-ora ~]$ /usr/TKLC/xIH/plat/bin/analyze_server.sh -i

   13:24:52: STARTING HEALTHCHECK PROCEDURE

   13:24:52: date: 03-17-15, hostname: cat-ora

   13:24:52: TPD VERSION: 7.8.3.0.0-89.21.0

   13:24:52: -------------------------------------------

   13:24:52: Checking disk free space

   13:24:52: No disk space issues found

   :

   13:25:02: All tests passed!

   13:25:02: ENDING HEALTHCHECK PROCEDURE WITH CODE 0

   If the output indicates the following error, ignore the error and continue the upgrade. This error indicates the target release and the running release are the same.

   00:47:29: Checking runlevel

   00:47:29: >>> Error: Runlevel value "3 4" is different from "N 4"

   If the output indicates any other failure, do not proceed with the upgrade. It is recommended to contact My Oracle Support (MOS) for guidance.

A.5.1 IDIH CLI: Shut Down the Mediation Guest

Follow this procedure to shut down the mediation guest to prepare for the Oracle Guest Upgrade.

1. Log in to the Mediation guest as admusr user.

   ssh <IDIH IP address>

   login as: admusr

   password: <enter password>

2. Shut down the Mediation guest.

   $ sudo init 0

   The active SOAM server may have some or all of the following expected alarms:

   Alarm ID = 19800 Communication Agent Connection Down

   Alarm ID = 11511 Unable to connect via Comagent to remote DIH server with hostname

   The active NOAM server may have some or all of the following expected alarms:

   Alarm ID = 19800 Communication Agent Connection Down

A.5.1 IDIH CLI: Shut Down the Application Guest

Follow this procedure to shut down the application guest to prepare for the Oracle Guest Upgrade.

1. Log in to the Application guest as admusr user.

   ssh <IDIH IP address>

   login as: admusr

   password: <enter password>

2. Shut down the Application guest.

   $ sudo init 0

   The active SOAM server may have some or all of the following expected alarms:

   Alarm ID = 19800 Communication Agent Connection Down

   Alarm ID = 11511 Unable to connect via Comagent to remote DIH server with hostname

   The active NOAM server may have some or all of the following expected alarms:

   Alarm ID = 19800 Communication Agent Connection Down

A.5.1 IDIH Application Guest CLI: Increase Size of /var/TKLC

1. It is seen that space available in /var/TKLC directory is less than the ISO size. So, , so there is need to increase the space of this directory.
2. Log in to the Application guest as admusr user.

   ssh <IDIH IP address>

   login as: admusr

   password: <enter password>

3. Check the space

   df -kh /var/TKLC

4. Note down the current space. Available space should be more than 6 GB space for this. In case sufficient space is already there, skip the next sub-steps.
5. Increase the space

   sudo lvresize -L +6G /dev/mapper/vgroot-plat_var_tklc

6. Resize the space

   sudo resize2fs /dev/mapper/vgroot-plat_var_tklc

7. Check the space again

   df -kh /var/TKLC

8. Available space should be more than 6 GB space for this.

A.5.1 Move Oracle ISO

1. Use a file transfer tool to copy the Oracle ISO to the Oracle guest as admusr.

   Example:

   $ scp oracle- DSR-8.6.0.1.0_96.15.0-x86_64.iso

   admusr@<ora-guest-ip>:/var/TKLC/upgrade

A.5.1 IDIH CLI: Start Oracle Guest Upgrade

The Oracle guest is upgraded using the Platform Configuration utility.

1. Launch the platform configuration utility.

   $ sudo su - platcfg

2. In the resulting menu, navigate to Maintenance, then Upgrade, and then Initiate Upgrade.
3. At the ISO selection menu, select the target release Oracle ISO and press Enter.

A.5.1 IDIH CLI: Monitor Upgrade Progress

1. To view the detailed progress of the upgrade, access the server command line (via SSH or Console), and enter:

   $ tail -f /var/TKLC/log/upgrade/upgrade.log

   Once the server has is upgraded, it reboots. It takes a couple of minutes for the Oracle processes to start up.

   Note:

   The ISO file for DB upgrade contains the default ASMSetup file and it is possible for it to overwrite “sd” definition with vd for disks. This affects the upgrade on setups that has sd disks, like sda, sdb and so on. In this scenario, just after the upgrade, when a reboot is triggered, it is possible that ASM will not be able to assign sdb disk to be used for Oracle DB. This can be verified by checking /etc/udev/rules.d/98-asm.rules file. This would contain KERNEL=="dm-7" SYMLINK+="asm/ASM0" OWNER="grid" GROUP="oinstall" MODE="0660" entry and KERNEL=="sdb" SYMLINK+="asm/ASM1" OWNER="grid" GROUP="oinstall" MODE="0660" would be removed. If sdb is missing from this file, it is recommended to edit ASMSteup file, $sudo vi /opt/xIH/oracle/instances/ASMSetup. Locate line 94, modify the expression ^vd by ^sd and save the file. Reboot the VM achange after making the changes. This would resolve the problem.

A.5.1 IDIH CLI: Perform a System Health Check on the Oracle Guest

1. Wait a few minutes to allow the Oracle guest to stabilize after the reboot, and repeat step 1 to perform the post-upgrade system health check.

   Note:

   The following cautions are expected due to the mediation and app servers being shut down:

   Caution:

   Mediation server is not reachable (or ping response exceeds 3 seconds).

   Caution:

   App server is not reachable (or ping response exceeds 3 seconds).