qrmutil

(Cluster Edition only) qrmutil allows you to back up, restore, and reconfigure the quorum device.

The utility is located in $SYBASE/$SYBASE_ASE/bin.

Syntax

--additional-run-parameters=parameter_list
--ase-config-extract=file_name
--ase-config-info
--ase-config-store=file_name
--ase-config-version=version_number
--buildquorum=[force]--cluster-take-over
--config-file=file_name
--diag={all | boot | toc | nodes | locks | config | cms}
--display={boot | nodes | heartbeat | master | cluster |
		instance | config | state}
--drop-cluster=[force]
--drop-instance=instance_name
--errorlog=file_name
--extract-config=file_name
-h, --help
-F, --cluster-input=file_name
--fence-capable=device_path
--installation=installation_mode
-s, --instance=instance_name
--instance-node=node_name
--interfaces-dir=path_to_interfaces_file
--max-instances=number_of_instances
--master-dev=master_device
--membership-mode=membership_mode
--primary-address=interconnect_address
--primary-port=port_number
--primary-protocol=protocol
-Q, --quorum-dev=quorum_device
--register-node=node_name
--secondary-address=interconnect_address
--secondary-port=port_number
--secondary-protocol=protocol
--traceflags=traceflag_list
--unregister-node=node_name
--verify-node=node_name
-v, --version]

Parameters

--additional-run-parameters=parameter_list – parameters that Unified Agent uses to start the data server. Unlike other settings, dataserver does not read additional run parameters. They are read by the unified agent and passed to the dataserver command line. If you include the --instance parameter, the additional run parameters apply to the specified instance. Otherwise, the additional run parameters apply to all instances in the cluster.
--ase-config-extract=file_name – extracts the Adaptive Server configuration file stored on the quorum device to the named file.
--ase-config-info – displays information about the Adaptive Server configuration file stored on the quorum device.
--ase-config-store=file_name – stores the named file in the quorum device as the Adaptive Server configuration file.
--ase-config-version=[version_number] – displays or sets the version of the master Adaptive Server configuration file stored on the quorum device.
--buildquorum[=force] – builds a new quorum device. Use =force to overwrite an exiting file or an existing quorum device on a raw partition. You must include the --cluster-input parameter with --buildquorum.
--config-file=config_file_name – if used with -instance, sets this path to the Adaptive Server configuration file for the specified instance. If you do not include -instance, sets the path to the cluster-wide configuration file.
--diag={all | boot | toc | nodes | locks | config | cms} – for internal use only.
--display={boot | nodes | heartbeat | master | cluster | instance | config | state} – displays the current state of cluster or instance:
- boot – displays start-up information for the cluster, including the version of the quorum device, any trace flags issued at start-up, the boot ID of the cluster, and any messages displayed at start-up.
- nodes – displays the registered management nodes.
- heartbeat – displays heartbeat information for all nodes in the cluster.
- master – displays master device information.
- cluster – displays the cluster configuration.
- instance – displays the instance configuration. You must include --instance=instance_name with this parameter.
- config – displays configuration for the cluster and for all instances in the cluster.
- state – displays the current state for the cluster and for all instances in the cluster.
--drop-cluster=[force] – drops a cluster and removes the quorum device. Use =force to force the drop if the quorum indicates the cluster is running.
Warning! --drop-cluster removes the cluster.
--drop-instance=instance_name – do not use; Sybase internal use only.
Warning! Use the sybcluster utility to drop an instance from the cluster.
--errorlog=log_file_name – full path to the error log for the specified instance. You must include the -instance-name parameter. Takes effect at next restart of the instance.
--extract-config=file_name – extracts the configuration area of the quorum device to the specified file.
-h | --help – displays the full syntax of qrmutil.
-F | cluster-input=file_name – loads the cluster configuration from the specified cluster input file.
--fence-capable=device_path – tests if specified device can be used for I/O fencing. Returns either “Device is fence capable” or “Device is not fence capable”.
--installation=installation_mode – changes the installation mode for the cluster. Values are:
- shared (default)
- private
-instance=instance_name – applies qrmutil parameters to a specified instance.
--interfaces-dir=interfaces_path – the path to a directory that contains a file named interfaces. If this parameter is used with --instance, it sets the path to the interfaces file for the specified instance. If --instance is not included, sets the path to the cluster-wide interfaces file.
--max-instances=number_of _instances – sets the maximum number of instances for the cluster configuration.
--master-dev=master_device_name – changes the master device the cluster uses.
--membership-mode=membership_mode – sets the membership mode. Values are:
- native (default)
- vcs
--primary-address=interconnect_address – changes the primary interconnect address for a given instance.
--primary-port=port_number – changes the starting port number for the primary interconnect for a given instance.
--primary-protocol=protocol – changes the protocol used for the primary cluster interconnect.
-Q | --quorum-dev=quoum_path – specifies the full path to the quorum device.
--register-node=node_name – registers a node for quorum management.
--secondary-address=inteconnect_address – changes the secondary interconnect address for a given instance.
--secondary-port=port_number – changes the starting port number for the secondary interconnect for a given instance.
--secondary-protocol=protocol – changes the protocol used for the secondary cluster interconnect.
--traceflags=trace_flag, trace_flag – changes the cluster-wide or the instance-specific trace flags for start-up. If you do not include a list of trace flags, qrmutil clears the trace flags for the cluster instance.
--unregister-node=node_name – unregisters a node from quorum management.
--verify-node=node_name – indicates that the specified node is registered on the quorum device.
-v | --version – displays the version information for the qrmutil utility.

Examples

Change path – Changes the path to the error log to /sybase/opt/cluster/ASE-15_0/ase1.log:

qrmutil --quorum_dev=/dev/raw/raw101 --instance=ase1 
--errorlog=/sybase/cluster/ASE-15_0/ASE-15_0/ase1.log

Register node – Registers the node “blade5” for mycluster:
```
qrmutil --quorum_dev=/dev/raw/raw101 --register-node=blade5
```

Create quorum device – Creates a new quorum device for the cluster “mycluster”:

qrmutil --quorum-dev=/dev/raw/raw101 --cluster-input=/sybase/cluster/ase1.inp -buildquorum

Back up device – Backs up the quorum device to /sybase/cluster_bak/quorum.bak:

qrmutil --quorum-dev=/dev/raw/raw101 
--extract-config=/sybase/cluster_bak/quorum.bak

Restore device – Restores the quorum device from the backup created in /sybase/cluster_bak/quorum.bak:

qrmutil --quorum-dev=/dev/raw/raw101 --cluster-input=/sybase/cluster_bak/quorum.bak --buildquorum=force

Display configuration – Displays the cluster configuration stored on the quorum device:
```
qrmutil --quorum-dev=/dev/raw/raw101 --display=config
```

Test – Tests whether the named device can be fenced:

qrmutil --quorum-dev=/dev/raw/raw101 --fence-capable=/dev/raw/raw106

Usage

qrmutil is primarily a diagnostic utility. SAP recommends that you use sybcluster to make configuration changes to the cluster.
You may pass as many as 20 commands to qrmutil. However, you can specify the --instance= parameter only once.
If you specify --buildquorum, the quorum is built and qrmutil exits without running any commands other than --cluster-input.
qrmutil exits after it executes the --drop-cluster parameter.

This is an example of using multiple commands:

qrmutil --quorum-dev=/dev/raw/raw101 --display=cluster 
--register-node=blade1 --unregister-node=blade2 --verify-node=blade3

Permissions

To run qrmutil, you must be the same sybase user that started the instance, with execute priveledges on the qrmutil binary, have direct access to the quorum device, and at least read permissions on the quorum file.