HP Serviceguard Metrocluster specs

6 Configuring complex workloads in a Continentalclusters environment using SADTA

Site Aware Disaster Tolerant Architecture (SADTA) enables automatic recovery of an entire application stack that is protected using physical data replication. The application stack can be packaged using mulit-node packages and failover packages with dependencies among them. SADTA also provides a single interface for manual failover of all the packages configured for an application stack.

Figure 2 SADTA Configuration in Continentalclusters

Continentalclusters

Site A				Site B
App. Pkg				App. Pkg
Site A Mount Point MNP		Site Safety	Site Safety	Site B Mount Point
Site A Disk Group MNP		Latch	Latch	Site B Disk Group
Site A CFS Sub Cluster		SG CFS SMNP		Site B CFS Sub Cluster
	Site				Site
	Controller				Controller
Node 1	Node 2			Node 1	Node 2

Data Replication
Application	Application
Data Disk	Data Disk
Disk Array	Disk Array
Site A	Site B
Active Application	Passive Application
Conﬁguration	Conﬁguration
Primary Cluster	Recovery Cluster

This section lists and describes the procedures for configuring a complex workload in Continentalclusters using SADTA.

To configure a complex workload in Continentalclusters:

1.Set up the replication between the arrays in the primary cluster and the recovery cluster.

2.Configure a primary cluster with a single site defined in the Serviceguard cluster configuration file.

3.Configure a recovery cluster with a single site defined in the Serviceguard cluster configuration file.

4.Set up the complex workload in the primary cluster.

43

Image 43

HP Serviceguard Metrocluster manual Sadta Configuration in Continentalclusters

Contents

HP Part Number Published February Legal Notices Contents Administering Continentalclusters Disaster recovery rehearsal in Continentalclusters Troubleshooting Continentalclusters 114 Sample Continentalclusters Ascii configuration file 115Package attributes Legacy packages 100 Glossary 145 Index 150 Introduction Sample Continentalclusters Configuration Traditional deployment method Building the Continentalclusters configurationCreating the Serviceguard clusters at both the sites Easy deployment method # passwd conclusr Setting up security# cmapplyconf -v -C /etc/cmcluster/cluster.config Csshsetup -r -k rsa -f host-list-file Creating data replication between the clusters Userhost Anyserviceguardnode Userrole MonitorUsername conclusr Using software based logical replication Using any other array based physical replication technology # vgexport -s -p -m mapfilename vgname Creating and Exporting LVM Volume Groups# mkdir /dev/vgxx # mknod /dev/vgxx/group c 64 0xnn0000 # vgchange -a n vgname Creating VxVM Disk Groups Devicegroup # cmmakepkg -m dts/ccxpca temp.configDts/xpca/dtspkgdir Dts/xpca/dtspkgdir pkgdirname Fence # cmcheckconf -P temp.config# cmapplyconf -P temp.config # cmmakepkg -m dts/cccaeva temp.config DC1SMISLIST DtapplicationstartuppolicyDrgroupname DC1STORAGEWORLDWIDENAME # cmhaltpkg packagename # cmmakepkg -m dts/ccsrdf temp.configDts/dts/dtspkgdir Rdfmode # cmrunpkg -n nodename packagename # cmmakepkg -m dts/cc3parrc pkgName.config# cmmodpkg -e packagename # cmcheckconf -P pkgName.config Configuring the monitor package# cmapplyconf -P pkgName.config # cmcheckconf -P ccmonpkg.conf Creating a Continentalclusters configuration# cd /etc/cmcluster # cmqueryconcl -C cmconcl.config Cluster information Recovery groups, the following parameters are available Recovery groups Monitoring definitions has the following parameters Monitoring definitions # cmapplyconcl -v -C cmconcl.config Checking and applying the Continentalclusters configurationVerify the Continentalclusters configuration # cmcheckconcl -v -C cmconcl.config Starting the Continentalclusters monitor package Testing the ContinentalclustersTesting Individual Packages Testing Continentalclusters Operations # cmviewconcl Testing the Continentalclusters Building the Continentalclusters configuration Receiving notification Performing recovery in case of disasterVerifying that recovery is required Preparing the storage manually in the recovery cluster # cmrecovercl Recovering the entire cluster after a cluster alarm# vxdg deport disk group name Viewing the Continentalclusters status Recovering the entire cluster after a cluster alertRecovering a single cluster in an N-1 configuration # cmrecovercl -f Switching the Primary and Recovery Cluster Roles Restoring disaster recovery cluster after a disaster Run this command Check and apply the Continentalclusters configurationRestart the monitor packages on every cluster Creating a new Primary Cluster # cmrunpkg ccmonpkg Creating a new Recovery Cluster# cmapplyconf -p ccmonpkg.config Restart the monitor package on the recovery cluster Configuring maintenance mode in Continentalclusters Disaster recovery rehearsal in ContinentalclustersOverview of Disaster Recovery rehearsal Configuring Continentalclusters Disaster Recovery rehearsal # mkdir directorypath # cmgetconf -p ccmonpkg ccnew.config# vgcreate /dev/vgcc -f /dev/sda1 # lvcreate -L size vgname Mke2fs -j Lvol # cmcheckconf -P packagename Configuring Continentalclusters rehearsal packagesModifying Continentalclusters configuration # cmhaltpkg ccmonpkg # cmapplyconf -P ccnew.config Client access IP address at recovery cluster Precautions to be taken while performing DR RehearsalCluster role switch during rehearsal Start rehearsal Move the recovery group into maintenance modeVerify data replication environment Prepare the replication environment for DR rehearsal Move the recovery group out of maintenance mode Cleanup of secondary mirror copyRecovering the primary cluster disaster during DR Rehearsal Restore replication environment for recovery Run cmrrcovercl command Limitations of DR rehearsal feature Sadta Configuration in Continentalclusters Configuring the primary cluster with a single site Setting up replication # vxdg -g cvmdgname set activation=sw Configuring the recovery cluster with a single siteSetting up the complex workload in the primary cluster # vxdg -s init cvmdgname replicateddisk1 replicateddisk2 # cmapplyconf -P cfspkg1.ascii # newfs -F vxfs /dev/vx/rdsk/cvmdgname/cvmdgvolname# cmmakepkg -m sg/cfsall /etc/cmcluster/cfspkg1.ascii # cmcheckconf -P cfspkg1.ascii # vgimport -s -m mapfilename vgname # cmapplyconf -v -P cvmdgpkgname.conf# mkdir /dev/vgname # mknod /dev/vgname/group c 64 0xnn0000 # lvcreate -L Xxxx /dev/vgname Configuring the complex workload at the primary cluster # cmmakepkg -m dts/sc -m dts/cc3parrc cwsc.config # cmmakepkg -m dts/sc -m dts/ccxpca cwsc.config# cmmakepkg -m dts/sc -m dts/cccaeva cwsc.config # cmmakepkg -m dts/sc -m dts/ccsrdf cwsc.config # cmapplyconf -v -P /etc/cmcluster/cwsc/cwsc.config Suspending the replication to the recovery cluster# cmviewcl -v -p pkgname # vxdg -stfC import cvmdgname Where node3 and node4 are the nodes at the recovery cluster# cmrunpkg siteBcfspkgname # vgchange -c n vgname # vgchange -c y vgname Configuring Site Safety Latch dependencies# vgchange -c n -S n vgname # vgchange -c y -S y vgname Configuring Continentalclusters Resuming the replication to the recovery cluster Checking the status of clusters, nodes, and packages Administering Continentalclusters Configured Event Status Duration Last Notification Sent Package Status State Pkgswitch Node Network Attributes Enabling and disabling maintenance modeStartup and Switching Characteristics # cmrecovercl -d recoverygroup1 # cmrecovercl -g RecoveryGroupName Recovering a cluster when the storage array or disks failStarting a recovery package forcefully # cmrecovercl -e recoverygroup1 # cmforceconcl cmrunpkg -n node3 Pkg1 Adding a Recovery Group to Continentalclusters# cmapplyconf -C cluster.config Adding or Removing a Node from a Cluster Modifying Continentalclusters configuration Modifying a package in a recovery groupRemoving a recovery group from the Continentalclusters Changing monitoring definitions Removing a rehearsal package from a recovery groupModifying a recovery group with a new rehearsal package Behavior of Serviceguard commands in Continentalclusters # ps -ef grep cmclapplyd Verifying the status of Continentalclusters daemonsRenaming the Continentalclusters # ps -ef grep cmclsentryd Maintaining the data replication environment Deleting the Continentalclusters configuration# what /usr/sbin/cmviewconcl Using the pairresync command Continuous Access EVA Link Suspend and Resume Modes Maintaining 3PAR Remote Copy data replication environment Maintaining EMC Srdf data replication environment Maintaining a Node Administering Continentalclusters using Sadta configurationMaintaining the Site Maintaining Site Controller Package Start the Site Controller package by enabling it Deleting the Site Controller PackageStarting a Complex Workload # cfsdgadm deleteems pkg1dg /dts/mcsc/cwsc # cmviewcl -l package Moving a Complex Workload to the Recovery ClusterShutting Down a Complex Workload Restarting a Failed Site Controller Package Reviewing Messages and Log Files of Monitoring Daemon Troubleshooting ContinentalclustersTroubleshooting Continentalclusters Error Messages Reviewing Messages and Log Files Troubleshooting Continentalclusters Error Messages PackageName on cluster ClusterName Migrating to Continentalclusters A.08.00 Verify the configuration and the status of the cluster Continentalclusters Worksheets Recovery Group WorksheetData Center Worksheet Cluster Event Worksheet Recovery Checklist Replication configuration Continentalclusters Site configuration Configuring a CRS sub-cluster using CFS CRS Sub-cluster configuration using CFS RAC database configuration RAC database configuration RAC MNP Site Controller package configuration Site Controller package configuration Configuration file parameters for Continentalclusters Datasenderpackage clustername/packagename Primarypackage Clustername/Packagename Continentalclusters Command and Daemon Reference Cmdeleteconcl -f Cmrunpkg or cmmodpkg commandCmcluster/cmoncl.config Cmrecovercl -f Recovery group name Cmrecovercl -r-gCmviewconcl Autofencedatasplit Package attributes Autopsuepsus Autopsusssws Autosvolpsue Autosvolpfus Clustertype Autosvolpsus Continuous Access link timeout value HorctimeoutHorctakeover command, -t timeout. Pkg Startup Timeout Horctimeout Monnotificationfrequency DTS PKG DIRWaittime Monpollinterval Monnotificationconsole MonnotificationsyslogAutoresync Fails to start with an error Before startup of the package on destination volumeStart with an error If Waittime is 0 default value, and if the state of DR Package Attributes for Continentalcluster with EMC Srdf Querytimeout DefaultEtc/cmcluster/ packagename/FORCEFLAG Consistencygroups is set to Supported mode are sync for synchronous and async for ConfigurationsGroup in an M by N configuration Where the Metrocluster Srdf environment file is located Migrating legacy to modular packages Legacy packagesMigrating legacy monitor package # cmmigratepkg -p packagename -s -o modularsgconf # cmmigratepkg -p packagename -o modularccmonpkg.conf# cmcheckconf -P modularccmonpkg.conf # cmapplyconf -P modularccmonpkg.conf # cmapplyconf -P modularsgecmcc.conf Validate the new modular package configuration filePathtoenvfile modularsgecmcc.conf # cmcheckconf -P modularsgecmcc.conf # cmmigratepkg -p packagename -s -o modularsg.conf # mkdir /etc/cmcluster/ccmonpkg Configuring legacy packagesConfiguring the monitor package in legacy style Validate the new package configuration file # cd /etc/cmcluster/pkgname # cmmakepkg -p pkgname.ascii Use the cmcheckconf command to validate the package# cmcheckconf -P ccmonpkg.config # mkdir /etc/cmcluster/pkgname # cmmakepkg -s pkgname.cntl Create a package control script EVA Legacy packages # mkdir /etc/cmcluster/pkgname Set Pkgdir to /etc/cmcluster/pkgname Example 1 Example # cfsdgadm add cvmdgnamecvmdgpkgname all=sw\ node1node2 Configuring storage devices for complex workloadEtc/cmcluster/pkgname # vxdg -s init cvmdgname replicateddisk1\ replicateddisk2 # mkdir /cfs # chmod 775 /cfs # mkdir /cfs/cvmdgname Where node1 and node2 are the nodes in the Source Disk Site# cfsdgadm activate cvmdgname # newfs -F vxfs \ Dev/vx/rdsk/cvmdgname/cvmdgvolname # cfsmount /cfs/cvmdgname EVA Continentalclusters Ascii configuration file Sample Continentalclusters Ascii configuration fileSample Continentalclusters Ascii configuration file Clustername #### #### Recoverygroupname Datasenderpackage Notification Console #### #### Clusterevent # Cluster Information Clusteralarm Seconds Notification Recoverygroupname RG2 Sample input and output files for cmswitchconcl command Recoverygroupname RG2 Primarypackage Oracle RAC instances in a Continentalclusters environment Supported Continentalclusters and RAC configuration # cfsdgadm activate diskgroup # vxdctl -c mode# cfsdgadm add disk group name all=sw # cfsmntadm add racdgl vol4 /cfs/mntl all=rwPage # cd your own directory # cp /opt/cmconcl/scripts/ccrac.config ccrac.config.mycopyCopy the file Copy the edited file to the final directory # cp ccrac.config.mycopy /etc/cmconcl/ccrac/ccrac.configPage Failover of Oracle RAC instances to the recovery site Oracle RAC instance package can be started in sequence Failback of Oracle RAC instances after a failover # /opt/cmconcl/bin/ccracmgmt.ksh -i indexNumber \ start Rehearsing Oracle RAC databases in Continentalclusters# cfsumount Mount Point Name # cfsdgadm deactivate Disk Group Name Sample Oracle RAC database with ASM in Sadta Configure a primary cluster with a single site Configuring the ASM disk group in the primary cluster Configure a recovery cluster with a single siteInstalling and configuring Oracle Clusterware Installing Oracle Real Application Clusters RAC software Halting the RAC database in the primary cluster Creating the Oracle RAC database in the primary cluster # chown -R oracleoinstall /opt/app/oracle/admin Page # cmviewcl -v -p ASMDGpkgname Run the following command at the remote site# chown -R oracleoinstall /opt/app/oracle/admin/hrdb # srvctl status Site cc2site1 # cmapplyconf -v -P sitecontrollerconfigurationfile # cmrunpkg siteController1 See also quorum server Glossary Span 100s or 1000s of kilometers Or system availability Format /dev/devtype/filename Svol Index Symbols 151