Skip to main content

We've Moved!

Product Documentation has moved to docs.hitachivantara.com
Hitachi Vantara Knowledge

Troubleshooting Oracle Database

This section provides guidelines for how to troubleshoot issues that might occur when using Oracle Databases.

An online backup or mount fails

Problem:

In the event of a failure, Protector attempts to revert the operations performed up to the point of the failure, but success cannot be guaranteed.

Solution:

If an Oracle online backup fails, then check that the operations described in the attachments to the log entry backupOnline Worksteps are reverted. It is especially important that the step endOnlineBackup is performed. If not then execute the steps described in the attachment to the log entry endOnlineBackup.

If an Oracle offline backup fails, check that the database is started.

If a mount fails then check if the status of the snapshot is mounted. If so, then perform an un-mount. In addition to this, check that an eventual ASM Diskgroup (asmcmd lsdg) is un-mounted and removed from the resource group (asmcmd umount ASM_DISKGROUP; srvctl remove diskgroup -g ASM_DISKGROUP). Do not try to issue a mount operation to another server, prior a successful un-mount operation.

Failed to discover Oracle environment when creating application node

Problem:

The following error message is seen when attempting to discover an Oracle Environment in the ‘Create Node – Oracle Database’ section:

Failed to discover Oracle Environment

Solution:

  1. Click Rediscover Oracle Environment
    NoteIf click fails, continue with the below steps.
  2. Ensure that Oracle is running. If not, start Oracle and repeat the discover process.
  3. Edit the file /etc/nsswitch.conf
    1. Using a suitable editor, open the file /etc/nsswitch.conf
    2. Locate the line that starts with hosts and comment it out with a #
    3. Copy the line and re-order it so that dns is first and files is last. E.g.:
      #hosts: files dns myhostname
      hosts: dns myhostname files
    4. Save the file.
    5. Attempt to discover the Oracle Environment again. The edited file should be picked up straight away and thus it should not be necessary to restart or reboot.

Oracle database snapshot fails to mount

Problem:

The following error message is logged if you attempt to mount an Oracle database snapshot to a location where one is already mounted:

Handler call failed: [...] ASM Diskgroups [...] mounted

Solution:

You cannot mount a second snapshot of the database to a location where one is already mounted.

RAC lock on database failed

The following error message is logged, during online snapshot operations, for all but one node in a RAC environment:

Handler call failed: [...] RAC Lock Database failed, check if [...]

Only one RAC node will succeed to lock the database prior to online snapshotting, all other nodes in the RAC will fail.

Solution:

Check that the operation has succeeded on exactly one of the RAC nodes.

Cannot find Oracle database metadata files on mount

Problem:

The metadata files relating to the mounted Oracle database files cannot be located.

Solution:

The following information message is logged when an Oracle database is mounted:

OracleHandler [...]: mount for 'Oracle DB: [...]' finished [...]

The attachment lists the destination paths where the Oracle metadata files have been placed.

Error when reverting on ASM in normal/high redundancy mode

Problem:

If you attempt to revert a database running on ASM in normal or high redundancy mode, the following warning message is logged during backup operations, followed by the error message:

Handler call failed: [...] normal or high redundancy used; revert not possible with one mirror

Solution:

ASM normal/high redundancy modes are currently not supported.

Warning during backup if data/redo files in the same directory

Problem:

The following warning message is logged during backup operations when the data and redo files of the Oracle database are located in the same directory:

OracleHandler [...]: Oracle Database Files and Redo in the [...]

Solution:

It is not recommended to have redo logs and data on the same physical disks, as this may cause unusable backups for anything other than crash consistent backups. Please refer to Oracle application software prerequisites for more information.

Oracle RAC RPO based policy creates more snapshots than expected

Problem:

When using the 'All Day' schedule option for an Oracle RAC policy and relying on the RPO setting alone, more snapshots than expected may be created. This can happen if rules are activated while one node in the RAC is busy or restarting, causing the activation time to be offset for that node.

Solution:

Try using the ‘Scheduled Time’ option rather than an RPO. Also check that the RAC nodes’ times are synchronized.

Listing Oracle RMAN channel configurations with schedulershow

schedulershow is a CLI tool that is used to aid the Oracle database administrator to configure Oracle RMAN to save and restore data using a datastore managed by Ops Center Protector. It provides the following functionality.

  • List the Oracle RMAN dataflows active on the current node
  • Create a sample RMAN channel definition for a dataflow operation

In order for a node to store or access any data in a Protector managed datastore using Oracle RMAN, the node has to be part of a dataflow and granted access using the access operation. Once the dataflow is compiled and distributed, it can be listed with schedulershow and Oracle RMAN can access it using an SBT channel.

Command line options
Abbreviated ArgumentFull Argument Description
-h--helpDisplay help.
-c---config <trigger> Display RMAN channel configuration for a dataflow. The command provides a list of valid triggers when running the it without parameters..
Usage Example

List all Oracle RMAN dataflows active on this node

app/bin/schedulershow

Application DataFlow Source OperationName Storage Trigger

Oracle mysampledf1 myOracleNode Access myRepository 1f04eeec33b10190b79ea6d47a65f3943f5a9de0c1a5d1557afeb420ea604046

Oracle Demo2 myOracleNode Access myRepository 6e7b941d5e73ec5a3ce72b64e70c36c56e6d12f76a379d0982d130d1b505a895

Get the Oracle RMAN channel definition for a dataflow

app/bin/schedulershow -c 1f04eeec33b10190b79ea6d47a65f3943f5a9de0c1a5d1557afeb420ea604046

# Application = Oracle

# DataFlow=mysampledf1

# Source=myOracleNode

# OperationName=Access

# Storage=myRepository

# Trigger=1f04eeec33b10190b79ea6d47a65f3943f5a9de0c1a5d1557afeb420ea604046

DEVICE TYPE 'SBT_TAPE' PARMS

'SBT_LIBRARY=/opt/hitachi/protector/app/lib/libhsbt.so,SBT_PARMS=(TRIGGER=1f04eeec33b10190b79ea6d47a65f3943f5a9de0c1a5d1557afeb420ea604046)'

 

  • Was this article helpful?