Red Hat ENTERPRISE LINUX 5 - GLOBAL NETWORK BLOCK DEVICE Guide d'installation

Configuring and Managing a Red HatClusterRed Hat Cluster for RedHat Enterprise Linux5.2Cluster_AdministrationISBN: N/APublication date: May 2008

If you spot a typo, or if you have thought of a way to make this manual better, we would love tohear from you. Please submit a report in Bugzilla (htt

Field DescriptionIP Address The IP address assigned to the PAP console.Login The login name used to access the PAP console.Password The password used

Table B.6. GNBD (Global Network Block Device)Field DescriptionName A name for the server with HP iLO support.Hostname The hostname assigned to the dev

Field DescriptionLogin The login name of a user capable of issuing power on/offcommands to the given IPMI port.Password The password used to authentic

Field DescriptionPort The switch outlet number.Table B.13. RPS-10 Power Switch (two-node clusters only)Field DescriptionName A name for the SANBox2 de

Field DescriptionName A name for the WTI power switch connected to the cluster.IP Address The IP address assigned to the device.Password The password

Appendix C. HA ResourceParametersThis appendix provides descriptions of HA resource parameters. You can configure theparameters with Luci, system-conf

Resource Resource Agent Reference to ParameterDescriptionmachines.Table C.1. HA Resource SummaryField DescriptionName The name of the Apache Service.S

Field Descriptionunmount Force Unmount kills all processes using the mount point to free up the mountwhen it tries to unmount.Reboot hostnode ifunmoun

Field DescriptionIP Address The IP address for the resource. This is a virtual IP address. IPv4 and IPv6addresses are supported, as is NIC link monito

Field Descriptionrights. For more information, refer to the exports (5) man page, GeneralOptions.Table C.8. NFS ClientField DescriptionNameDescriptive

Red Hat Cluster Configuration andManagement OverviewRed Hat Cluster allows you to connect a group of computers (called nodes or members) to worktogeth

Field DescriptionName Specifies a service name for logging and other purposes.Config File Specifies an absolute path to a configuration file. The defa

Field DescriptionWait(seconds)Table C.13. PostgreSQL 8Field DescriptionSAPDatabaseNameSpecifies a unique SAP system identifier. For example, P01.SAPex

Field DescriptionprofileName of theSAP STARTprofileSpecifies name of the SAP START profile.Table C.15. SAP® InstanceNoteRegarding Table C.16, “Samba S

Field Descriptionrunning on it. If no nodes are available for a service to run exclusively, theservice is not restarted after a failure. Additionally,

Field DescriptionName Specifies a service name for logging and other purposes.Config File Specifies the absolute path to the configuration file. The d

Field DescriptionDomainRecoverypolicyRecovery policy provides the following options:• Disable — Disables the virtual machine if it fails.• Relocate —

106

Appendix D. HA Resource BehaviorThis appendix describes common behavior of HA resources. It is meant to provide ancillaryinformation that may be helpf

• Section 5, “Debugging and Testing Services and Resource Ordering”NoteThe sections that follow present examples from the cluster configuration file,/

• Children must all stop cleanly before a parent may be stopped.• For a resource to be considered in good health, all its children must be in good hea

Other options are available for storage according to the type of storage interface; for example,iSCSI or GNBD. A Fibre Channel switch can be configure

Resource Child Type Start-order Value Stop-order ValueLVM lvm 1 9File System fs 2 8GFS File System clusterfs 3 7NFS Mount netfs 4 6NFS Export nfsexpor

Example D.3. Ordering Within a Resource TypeTyped Child Resource Starting OrderIn Example D.3, “Ordering Within a Resource Type”, the resources are st

stopped in the reverse order listed in the Service foo portion of/etc/cluster/cluster.conf.5. lvm:1 — This is an LVM resource. All LVM resources are s

3. fs:1 — This is a File System resource. If there were other File System resources in Servicefoo, they would start in the order listed in the Service

4. ip:10.1.1.1 — This is an IP Address resource. If there were other IP Address resources inService foo, they would stop in the reverse order listed i

<ip address="10.2.13.20"/></service>Example D.5. NFS Service Set Up for Resource Reuse and InheritanceIf the service were flat (

<service name="foo"><script name="script_one" ...><script name="script_two" .../></script><sc

Action Syntaxresourceagents.Display thestart andstopordering ofa service.Display start order:rg_test noop /etc/cluster/cluster.conf start service serv

118

Appendix E. Upgrading A Red HatCluster from RHEL 4 to RHEL 5This appendix provides a procedure for upgrading a Red Hat cluster from RHEL 4 to RHEL 5.T

relationship among the cluster components. Figure 1.2, “Cluster Configuration Structure” showsan example of the hierarchical relationship among cluste

f. Run service ccsd stop.3. Disable cluster software from starting during reboot. At each node, run /sbin/chkconfig asfollows:# chkconfig --level 2345

You shouldn't change any of these values if the filesystem is mounted.Are you sure? [y/n] ycurrent lock protocol name = "lock_gulm"new

122

IndexAACPIconfiguring, 17Apache HTTP Serverhttpd.conf, 84setting up service, 83Bbehavior, HA resources, 107Cclusteradministration, 13, 47, 75diagnosin

Ggeneralconsiderations for cluster administration, 25Hhardwarecompatible, 13HTTP servicesApache HTTP Serverhttpd.conf, 84setting up, 83Iintegrated fen

A brief overview of each configuration tool is provided in the following sections:• Section 2, “Conga”• Section 3, “system-config-cluster Cluster Admi

To administer a cluster or storage, an administrator adds (or registers) a cluster or a computerto a luci server. When a cluster or a computer is regi

Figure 1.4. luci cluster TabChapter 1. Red Hat Cluster Configuration and Management Overview6

Figure 1.5. luci storage Tab3. system-config-cluster Cluster Administration GUIThis section provides an overview of the cluster administration graphic

While system-config-cluster provides several convenient tools for configuringand managing a Red Hat Cluster, the newer, more comprehensive tool, Conga

The Cluster Configuration Tool represents cluster configuration components in theconfiguration file (/etc/cluster/cluster.conf) with a hierarchical gr

Configuring and Managing a Red Hat Cluster describes the configuration and management ofRed Hat cluster systems for Red Hat Enterprise Linux 5.2 It do

Services. Using configuration buttons at the bottom of the right frame (below Properties),you can create services (when Services is selected) or edit

The nodes and services displayed in the Cluster Status Tool are determined by the clusterconfiguration file (/etc/cluster/cluster.conf). You can use t

12

Before Configuring a Red HatClusterThis chapter describes tasks to perform and considerations to make before installing andconfiguring a Red Hat Clust

Cluster Nodes” lists the IP port numbers, their respective protocols, the components to whichthe port numbers are assigned, and references to iptables

If a cluster node is running luci, port 11111 should already have been enabled.IP PortNumberProtocol Component Reference to Example ofiptables Rules80

10.10.10.0/24 -d 10.10.10.0/24 --dports 11111 -j ACCEPTExample 2.3. Port 11111: ricci (Cluster Node and Computer Running luci)-A INPUT -i 10.10.10.200

1http://www.redhat.com/cluster_suite/hardware/10.10.10.0/24 -d 10.10.10.0/24 --dports 50007 -j ACCEPTExample 2.9. Port 50007: ccsd (UDP)3. Configuring

• Changing the BIOS setting to "instant-off" or an equivalent setting that turns off the nodewithout delayNoteDisabling ACPI Soft-Off with t

management.— OR —• chkconfig --level 2345 acpid off — This command turns off acpid.2. Reboot the node.3. When the cluster is configured and running, v

Configuring and Managing a Red Hat Cluster: Red Hat Clusterfor Red Hat Enterprise LinuxCopyright © 2008 Red Hat, Inc.Copyright © 2008 Red Hat, Inc. Th

may vary among computers. However, the objective of this procedure is toconfigure the BIOS so that the computer is turned off via the power buttonwith

3.3. Disabling ACPI Completely in the grub.conf FileThe preferred method of disabling ACPI Soft-Off is with chkconfig management (Section 3.1,“Disabli

title Red Hat Enterprise Linux Server (2.6.18-36.el5)root (hd0,0)kernel /vmlinuz-2.6.18-36.el5 ro root=/dev/VolGroup00/LogVol00console=ttyS0,115200n8

Resource Behavior.5. Configuring max_lunsIt is not necessary to configure max_luns in Red Hat Enterprise Linux 5.In Red Hat Enterprise Linux releases

CMAN membership timeout valueThe CMAN membership timeout value (the time a node needs to be unresponsive beforeCMAN considers that node to be dead, an

multicast addressing and IGMP are enabled. Without multicast and IGMP, not all nodes canparticipate in a cluster, causing the cluster to fail.NoteProc

and corrupting it. It is strongly recommended that fence devices (hardware or softwaresolutions that remotely power, shutdown, and reboot cluster node

Configuring Red Hat Cluster WithCongaThis chapter describes how to configure Red Hat Cluster software using Conga, and consists ofthe following sectio

2. Starting luci and ricciTo administer Red Hat Clusters with Conga, install and run luci and ricci as follows:1. At each node to be administered by C

5. Start luci using service luci restart. For example:# service luci restartShutting down luci: [ OK ]Starting luci: generating https SSL certificates

Configuring and Managing a Red Hat Cluster

4. Global Cluster PropertiesWhen a cluster is created, or if you select a cluster to configure, a cluster-specific page isdisplayed. The page provides

3. Multicast tab — This tab provides an interface for configuring these MulticastConfiguration parameters: Let cluster choose the multicast address an

Parameter DescriptionMinimum Score The minimum score for a node to be considered "alive". If omitted orset to 0, the default function, floor

The following shared fence devices are available:• APC Power Switch• Brocade Fabric Switch• Bull PAP• Egenera SAN Controller• GNBD• IBM Blade Center•

The starting point of each procedure is at the cluster-specific page that you navigate to fromChoose a cluster to administer displayed on the cluster

Figure 3.1. Fence Device Configuration3. At the Add a Sharable Fence Device page, click the drop-down box under Fencing Typeand select the type of fen

5.2. Modifying or Deleting a Fence DeviceTo modify or delete a fence device, follow these steps:1. At the detailed menu for the cluster (below the clu

Creating a cluster consists of selecting a set of nodes (or members) to be part of the cluster.Once you have completed the initial step of creating a

4. Click Submit. Clicking Submit causes the following actions:a. Cluster software packages to be downloaded onto the added node.b. Cluster software to

1. Click the link of the node to be deleted. Clicking the link of the node to be deleted causes apage to be displayed for that link showing how that n

Introduction ... vii1. Document Conventions

be started (either manually or by the cluster software).• Unordered — When a cluster service is assigned to an unordered failover domain, themember on

7.1. Adding a Failover DomainTo add a failover domain, follow the steps in this section. The starting point of the procedure isat the cluster-specific

displayed on the cluster tab.1. At the detailed menu for the cluster (below the clusters menu), click Failover Domains.Clicking Failover Domains cause

9. To make additional changes to the failover domain, continue modifications at the FailoverDomain Form page and click Submit when you are done.8. Add

TipUse a descriptive name that clearly distinguishes the service from other servicesin the cluster.4. Add a resource to the service; click Add a resou

inet 127.0.0.1/8 scope host loinet6 ::1/128 scope hostvalid_lft forever preferred_lft forever2: eth0: <BROADCAST,MULTICAST,UP> mtu 1356 qdisc pf

• Hard Drives• Partitions• Volume GroupsEach section is set up as an expandable tree, with links to property sheets for specific devices,partitions, a

Managing Red Hat Cluster WithCongaThis chapter describes various administrative tasks for managing a Red Hat Cluster andconsists of the following sect

• Delete this cluster — Selecting this action halts a running cluster, disables clustersoftware from starting automatically, and removes the cluster c

Selecting Have node leave cluster shuts down cluster software and makes the nodeleave the cluster. Making a node leave a cluster prevents the node fro

10. Configuring Cluster Storage ...454. Managing Red Hat Cluster With Conga ...

• If service is running — Configure this service, Restart this service, and Stop thisservice.• If service is not running — Configure this service, Sta

Configuring Red Hat Cluster Withsystem-config-clusterThis chapter describes how to configure Red Hat Cluster software usingsystem-config-cluster, and

3. Creating fence devices. Refer to Section 4, “Configuring Fence Devices”.4. Creating cluster members. Refer to Section 5, “Adding and Deleting Membe

Figure 5.1. Starting a New Configuration FileNoteThe Cluster Management tab for the Red Hat Cluster Suite management GUI isavailable after you save th

dialog box if you enable Use a Quorum disk: Interval, TKO, Votes, Minimum Score,Device, Label, and Quorum Disk Heuristic. Table 5.1, “Quorum-Disk Para

Figure 5.2. Creating A New Configuration4. When you have completed entering the cluster name and other parameters in the NewConfiguration dialog box,

Figure 5.3. The Cluster Configuration ToolParameter DescriptionUse a Quorum Disk Enables quorum disk. Enables quorum-disk parameters in the NewConfigu

Parameter DescriptionDevice The storage device the quorum daemon uses. The device must be thesame on all nodes.Label Specifies the quorum disk label c

5. Specify the Fence Daemon Properties parameters: Post-Join Delay and Post-Fail Delay.a. The Post-Join Delay parameter is the number of seconds the f

Figure 5.4. Fence Device Configuration2. At the Fence Device Configuration dialog box, click the drop-down box under Add a NewFence Device and select

IntroductionThis document provides information about installing, configuring and managing Red Hat Clustercomponents. Red Hat Cluster components are pa

2. At the bottom of the right frame (labeled Properties), click the Add a Cluster Node button.Clicking that button causes a Node Properties dialog box

box to be displayed.c. At the Fence Configuration dialog box, bottom of the right frame (below Properties),click Add a New Fence Level. Clicking Add a

nodes, follow these steps:1. Add the node and configure fencing for it as inSection 5.1, “Adding a Member to a Cluster”.2. Click Send to Cluster to pr

2. Click Send to Cluster to propagate the updated configuration to other running nodes in thecluster.3. Use the scp command to send the updated /etc/c

Figure 5.6. Confirm Deleting a Memberd. At that dialog box, click Yes to confirm deletion.e. Propagate the updated configuration by clicking the Send

• Unrestricted — Allows you to specify that a subset of members are preferred, but that acluster service assigned to this domain can run on any availa

• Section 6.1, “Adding a Failover Domain”• Section 6.2, “Removing a Failover Domain”• Section 6.3, “Removing a Member from a Failover Domain”6.1. Addi

Figure 5.7. Failover Domain Configuration: Configuring a FailoverDomain4. Click the Available Cluster Nodes drop-down box and select the members for t

Figure 5.8. Failover Domain Configuration: Adjusting Priorityb. For each node that requires a priority adjustment, click the node listed in the Member

6.2. Removing a Failover DomainTo remove a failover domain, follow these steps:1. At the left frame of the Cluster Configuration Tool, click the failo

environment.• Global File System: Configuration and Administration — Provides information about installing,configuring, and maintaining Red Hat GFS (R

• New cluster — If this is a new cluster, choose File => Save to save the changes to thecluster configuration.• Running cluster — If this cluster i

Figure 5.9. Adding a Cluster Service4. If you want to restrict the members on which this cluster service is able to run, choose afailover domain from

types of services you can leave the Run Exclusive unchecked.NoteCircumstances that require enabling Run Exclusive are rare. Enabling RunExclusive can

NoteTo verify the existence of the IP service resource used in a cluster service, youmust use the /sbin/ip addr list command on a cluster node. The fo

each node or start the cluster software on each cluster node by running the followingcommands at each node in this order:1. service cman start2. servi

Managing Red Hat Cluster Withsystem-config-clusterThis chapter describes various administrative tasks for managing a Red Hat Cluster andconsists of th

3. service clvmd stop, if CLVM has been used to create clustered volumes4. service cman stopStopping the cluster services on a member causes its servi

You can use the Cluster Status Tool to enable, disable, restart, or relocate a high-availabilityservice. The Cluster Status Tool displays the current

3. Modifying the Cluster ConfigurationTo modify the cluster configuration (the cluster configuration file(/etc/cluster/cluster.conf), use the Cluster

3. Clicking Send to Cluster causes a Warning dialog box to be displayed. Click Yes to saveand propagate the configuration.4. Clicking Yes causes an In

Italic Courier font represents a variable, such as an installation directory:install_dir/bin/bold fontBold font represents application programs and te

9. Propagate the updated configuration file throughout the cluster by clicking Send to Cluster.NoteThe Cluster Configuration Tool does not display the

in the order shown to restart cluster software:1. service cman start2. service clvmd start, if CLVM has been used to create clustered volumes3. servic

82

Appendix A. Example of Setting UpApache HTTP ServerThis appendix provides an example of setting up a highly available Apache HTTP Server on aRed Hat C

systems from accessing the same data simultaneously, which may result in data corruption.Therefore, do not include the file systems in the /etc/fstab

your configuration. For example:• Specify the directory that contains the HTML files. Also specify this mount point whenadding the service to the clus

Before the service is added to the cluster configuration, ensure that the Apache HTTP Serverdirectories are not mounted. Then, on one node, invoke the

or leave it as None.• Click the Add a Shared Resource to this service button. From the available list, chooseeach resource that you created in the pre

88

Appendix B. Fence DeviceParametersThis appendix provides tables with parameter descriptions of fence devices.NoteCertain fence devices have an optiona