Storage Foundation for Oracle® RAC 7.3.1 Administrator's Guide - Linux
- Section I. SF Oracle RAC concepts and administration
- Overview of Storage Foundation for Oracle RAC
- About Storage Foundation for Oracle RAC
- How SF Oracle RAC works (high-level perspective)
- Component products and processes of SF Oracle RAC
- Periodic health evaluation of SF Oracle RAC clusters
- About Virtual Business Services
- About Veritas InfoScale Operations Manager
- About Veritas Services and Operations Readiness Tools (SORT)
- Administering SF Oracle RAC and its components
- Administering SF Oracle RAC
- Setting the environment variables for SF Oracle RAC
- Starting or stopping SF Oracle RAC on each node
- Applying Oracle patches on SF Oracle RAC nodes
- Migrating Pluggable Databases (PDB) between Container Databases (CDB)
- Installing Veritas Volume Manager, Veritas File System, or ODM patches on SF Oracle RAC nodes
- Applying operating system updates on SF Oracle RAC nodes
- Adding storage to an SF Oracle RAC cluster
- Recovering from storage failure
- Backing up and restoring Oracle database using Veritas NetBackup
- Enhancing the performance of SF Oracle RAC clusters
- Administering SmartIO
- Creating snapshots for offhost processing
- Managing database storage efficiently using SmartTier
- Optimizing database storage using Thin Provisioning and SmartMove
- Scheduling periodic health checks for your SF Oracle RAC cluster
- Using environment variables to start and stop VCSMM modules
- Verifying the nodes in an SF Oracle RAC cluster
- Administering VCS
- About managing VCS modules
- Viewing available Veritas device drivers
- Starting and stopping VCS
- Environment variables to start and stop VCS modules
- Adding and removing LLT links
- Configuring aggregated interfaces under LLT
- Displaying the cluster details and LLT version for LLT links
- Configuring destination-based load balancing for LLT
- Enabling and disabling intelligent resource monitoring for agents manually
- Administering the AMF kernel driver
- Administering I/O fencing
- About administering I/O fencing
- About the vxfentsthdw utility
- General guidelines for using the vxfentsthdw utility
- About the vxfentsthdw command options
- Testing the coordinator disk group using the -c option of vxfentsthdw
- Performing non-destructive testing on the disks using the -r option
- Testing the shared disks using the vxfentsthdw -m option
- Testing the shared disks listed in a file using the vxfentsthdw -f option
- Testing all the disks in a disk group using the vxfentsthdw -g option
- Testing a disk with existing keys
- About the vxfenadm utility
- About the vxfenclearpre utility
- About the vxfenswap utility
- Enabling or disabling the preferred fencing policy
- About I/O fencing log files
- Migrating from disk-based fencing to server-based fencing using the installer
- Migrating from server-based fencing to disk-based fencing using the installer
- Administering the CP server
- Administering CFS
- Administering CVM
- Listing all the CVM shared disks
- Establishing CVM cluster membership manually
- Changing the CVM master manually
- Importing a shared disk group manually
- Deporting a shared disk group manually
- Starting shared volumes manually
- Verifying if CVM is running in an SF Oracle RAC cluster
- Verifying CVM membership state
- Verifying the state of CVM shared disk groups
- Verifying the activation mode
- Administering Flexible Storage Sharing
- About Flexible Storage Sharing disk support
- About the volume layout for Flexible Storage Sharing disk groups
- Setting the host prefix
- Exporting a disk for Flexible Storage Sharing
- Setting the Flexible Storage Sharing attribute on a disk group
- Using the host disk class and allocating storage
- Administering mirrored volumes using vxassist
- Displaying exported disks and network shared disk groups
- Tuning LLT for memory and performance in FSS environments
- Backing up and restoring disk group configuration data
- Administering SF Oracle RAC global clusters
- Administering SF Oracle RAC
- Overview of Storage Foundation for Oracle RAC
- Section II. Performance and troubleshooting
- Troubleshooting SF Oracle RAC
- About troubleshooting SF Oracle RAC
- Restarting the installer after a failed network connection
- Installer cannot create UUID for the cluster
- Troubleshooting SF Oracle RAC pre-installation check failures
- Troubleshooting LLT health check warning messages
- Troubleshooting I/O fencing
- SCSI reservation errors during bootup
- The vxfentsthdw utility fails when SCSI TEST UNIT READY command fails
- Node is unable to join cluster while another node is being ejected
- System panics to prevent potential data corruption
- Cluster ID on the I/O fencing key of coordinator disk does not match the local cluster's ID
- Fencing startup reports preexisting split-brain
- Registered keys are lost on the coordinator disks
- Replacing defective disks when the cluster is offline
- Troubleshooting I/O fencing health check warning messages
- Troubleshooting CP server
- Troubleshooting server-based fencing on the SF Oracle RAC cluster nodes
- Issues during online migration of coordination points
- Troubleshooting Cluster Volume Manager in SF Oracle RAC clusters
- Restoring communication between host and disks after cable disconnection
- Shared disk group cannot be imported in SF Oracle RAC cluster
- Error importing shared disk groups in SF Oracle RAC cluster
- Unable to start CVM in SF Oracle RAC cluster
- CVM group is not online after adding a node to the SF Oracle RAC cluster
- CVMVolDg not online even though CVMCluster is online in SF Oracle RAC cluster
- Shared disks not visible in SF Oracle RAC cluster
- Troubleshooting CFS
- Troubleshooting interconnects
- Troubleshooting Oracle
- Error when starting an Oracle instance in SF Oracle RAC
- Clearing Oracle group faults
- Oracle log files show shutdown called even when not shutdown manually
- DBCA fails while creating an Oracle RAC database
- Oracle's clusterware processes fail to start
- Oracle Clusterware fails after restart
- Troubleshooting the Virtual IP (VIP) configuration in an SF Oracle RAC cluster
- Troubleshooting Oracle Clusterware health check warning messages in SF Oracle RAC clusters
- Troubleshooting ODM in SF Oracle RAC clusters
- Troubleshooting Flex ASM in SF Oracle RAC clusters
- Prevention and recovery strategies
- Tunable parameters
- Troubleshooting SF Oracle RAC
- Section III. Reference
About GAB run-time or dynamic tunable parameters
You can change the GAB dynamic tunable parameters while GAB is configured and while the cluster is running. The changes take effect immediately on running the gabconfig command. Note that some of these parameters also control how GAB behaves when it encounters a fault or a failure condition. Some of these conditions can trigger a PANIC which is aimed at preventing data corruption.
You can display the default values using the gabconfig -l command. To make changes to these values persistent across reboots, you can append the appropriate command options to the /etc/gabtab file along with any existing options. For example, you can add the -k option to an existing /etc/gabtab file that might read as follows:
gabconfig -c -n4
After adding the option, the /etc/gabtab file looks similar to the following:
gabconfig -c -n4 -k
Table: GAB dynamic tunable parameters describes the GAB dynamic tunable parameters as seen with the gabconfig -l command, and specifies the command to modify them.
Table: GAB dynamic tunable parameters
GAB parameter | Description and command |
|---|---|
Control port seed | This option defines the minimum number of nodes that can form the cluster. This option controls the forming of the cluster. If the number of nodes in the cluster is less than the number specified in the Use the following command to set the number of nodes that can form the cluster: gabconfig -n count Use the following command to enable control port seed. Node can form the cluster without waiting for other nodes for membership: gabconfig -x |
Halt on process death | Default: Disabled This option controls GAB's ability to halt (panic) the system on user process death. If _had and _hashadow are killed using kill -9, the system can potentially lose high availability. If you enable this option, then the GAB will PANIC the system on detecting the death of the client process. The default behavior is to disable this option. Use the following command to enable halt system on process death: gabconfig -p Use the following command to disable halt system on process death: gabconfig -P |
Missed heartbeat halt | Default: Disabled If this option is enabled then the system will panic on missing the first heartbeat from the VCS engine or the vxconfigd daemon in a CVM environment. The default option is to disable the immediate panic. This GAB option controls whether GAB can panic the node or not when the VCS engine or the vxconfigd daemon miss to heartbeat with GAB. If the VCS engine experiences a hang and is unable to heartbeat with GAB, then GAB will NOT PANIC the system immediately. GAB will first try to abort the process by sending SIGABRT (kill_ntries - default value 5 times) times after an interval of "iofence_timeout" (default value 15 seconds). If this fails, then GAB will wait for the "isolate timeout" period which is controlled by a global tunable called isolate_time (default value 2 minutes). If the process is still alive, then GAB will PANIC the system. If this option is enabled GAB will immediately HALT the system in case of missed heartbeat from client. Use the following command to enable system halt when process heartbeat fails: gabconfig -b Use the following command to disable system halt when process heartbeat fails: gabconfig -B |
Halt on rejoin | Default: Disabled This option allows the user to configure the behavior of the VCS engine or any other user process when one or more nodes rejoin a cluster after a network partition. By default GAB will not PANIC the node running the VCS engine. GAB kills the userland process (the VCS engine or the vxconfigd process). This recycles the user port (port h in case of the VCS engine) and clears up messages with the old generation number programmatically. Restart of the process, if required, must be handled outside of GAB control, e.g., for hashadow process restarts _had. When GAB has kernel clients (such as fencing, VxVM, or VxFS), then the node will always PANIC when it rejoins the cluster after a network partition. The PANIC is mandatory since this is the only way GAB can clear ports and remove old messages. Use the following command to enable system halt on rejoin: gabconfig -j Use the following command to disable system halt on rejoin: gabconfig -J |
Keep on killing | Default: Disabled If this option is enabled, then GAB prevents the system from PANICKING when the VCS engine or the vxconfigd process fail to heartbeat with GAB and GAB fails to kill the VCS engine or the vxconfigd process. GAB will try to continuously kill the VCS engine and will not panic if the kill fails. Repeat attempts to kill process if it does not die gabconfig -k |
Quorum flag | Default: Disabled This is an option in GAB which allows a node to IOFENCE (resulting in a PANIC) if the new membership set is < 50% of the old membership set. This option is typically disabled and is used when integrating with other products Enable iofence quorum gabconfig -q Disable iofence quorum gabconfig -d |
GAB queue limit | Default: Send queue limit: 128 Default: Recv queue limit: 128 GAB queue limit option controls the number of pending message before which GAB sets flow. Send queue limit controls the number of pending message in GAB send queue. Once GAB reaches this limit it will set flow control for the sender process of the GAB client. GAB receive queue limit controls the number of pending message in GAB receive queue before GAB send flow control for the receive side. Set the send queue limit to specified value gabconfig -Q sendq:value Set the receive queue limit to specified value gabconfig -Q recvq:value |
IOFENCE timeout | Default: 15000(ms) This parameter specifies the timeout (in milliseconds) for which GAB will wait for the clients to respond to an IOFENCE message before taking next action. Based on the value of kill_ntries , GAB will attempt to kill client process by sending SIGABRT signal. If the client process is still registered after GAB attempted to kill client process for the value of kill_ntries times, GAB will halt the system after waiting for additional isolate_timeout value. Set the iofence timeout value to specified value in milliseconds. gabconfig -f value |
Stable timeout | Default: 5000(ms) Specifies the time GAB waits to reconfigure membership after the last report from LLT of a change in the state of local node connections for a given port. Any change in the state of connections will restart GAB waiting period. Set the stable timeout to specified value gabconfig -t stable |
Isolate timeout | Default: 120000(ms) This tunable specifies the timeout value for which GAB will wait for client process to unregister in response to GAB sending SIGKILL signal. If the process still exists after isolate timeout GAB will halt the system gabconfig -S isolate_time:value |
Kill_ntries | Default: 5 This tunable specifies the number of attempts GAB will make to kill the process by sending SIGABRT signal. gabconfig -S kill_ntries:value |
Driver state | This parameter shows whether GAB is configured. GAB may not have seeded and formed any membership yet. |
Partition arbitration | This parameter shows whether GAB is asked to specifically ignore jeopardy. See the gabconfig (1M) manual page for details on the -s flag. |