HPUX 专区 High Availability MC/ServiceGuard Installation

ZYY0104

High Availability MC/ServiceGuard Installation
Contents
1 MC/ServiceGuard Installation
2 MC/ServiceGuard Configuration
2.1 Basic System Configuration
2.2 Cluster Configuration
2.3 Sample Package and Service Configuration
3 MC/ServiceGuard Installation Acceptance Test
3.1 MC/ServiceGuard Package/Service Failure Scenarios
Appendix A Sample cluster configuration file - /etc/cmcluster/cluster.cfg
Appendix B Sample package configuration file - /etc/cmcluster/spkg/spkg.cfg
Appendix C Sample package control file - /etc/cmcluster/spkg/spkg.ctl (configuration portion)
Appendix D Sample service script file - /etc/cmcluster/spkg/sscript
Appendix E Sample cluster configuration diagram
Appendix F MC/ServiceGuard Quick Reference
Reference Documentation
a MC/ServiceGuard Version A.10.06 Release Notes (B3936-90009)
b Managing MC/ServiceGuard (B3936-90005)1 MC/ServiceGuard Installation
MC/ServiceGuard Cluster configuration is based on two nodes machines (namely primary and standby), single LAN connection (data and heartbeat signals), single shared SCSI bus, one shared SCSI data disk as lock disk.  Appendix E shows sample MC/ServiceGuard configuration diagram.
Full Installation on MC/ServiceGuard, including cluster and package manager.
command swinstall on node primary and node standby
2 MC/ServiceGuard Configuration
2.1 Basic System Configuration
  i Update hosts table (on primary and standby)
   Edit file /etc/hosts to include all configured network inte**ces of all hosts
  ii Setup trusted system (on primary and standby)
   Edit file /.rhosts with examples:
    primary root
    standby root
  iii Configure shared volume groups and logical volumes
   On primary, create shared volume group(s) and export map file
    primary# mkdir /dev/vgshare
    primary# mknod /dev/vgshare/group c 64 0x090000
     * choose unused minor number eg 0x090000 as necessary
    primay# pvcreate -f /dev/rdsk/cXtYd0
     * Lock disk controller number X and SCSI address Y
    primary# vgcreate /dev/vgshare /dev/dsk/cXtYd0
    primary# lvcreate -l 100 /dev/vgshare
    primary# newfs -F vxfs /dev/vgshare/rlvol1
    primary# mkdir /shared
    primary# mount /dev/vgshare/lvol1 /shared
    primary# bdf   
     * checkpoint for filesystem mounting
    primary# umount /shared
    primary# vgchange -a n /dev/vgshare
    primary# vgexport -p -v -s -m /tmp/vgshare.map /dev/vgshare
    primary# rcp /tmp/vgshare.map standby:/tmp/vgshare.map
   On standby, import shared volume group(s)
    standby# mkdir /dev/vgshare
    standby# mknod /dev/vgshare/group c 64 0x090000
     * choose unused minor number eg 0x090000 as necessary
standby# vgimport -s -v -m /tmp/vgshare.map /dev/vgshare /dev/dsk/cXtYd0
     * Lock disk controller number X and SCSI address Y
     ** may be different from the ones in primary
    standby# vgchange -a y /dev/vgshare
    standby# mkdir /shared
    standby# mount /dev/vgshare/lvol1 /shared
    standby# bdf    # checkpoint
    standby# vgcfgbackup /dev/vgshare
    standby# umount /shared
    standby# vgchange -a n /dev/vgshare
   On both primary and standby, activate local volume group(s) only
    Edit file /etc/lvmrc
    a. AUTO_VG_ACTIVATE=1 change to AUTO_VG_ACTIVATE=0
    b. under function custom_vg_activation, add local VGs
      /sbin/vgchange -a y -s /dev/vg00 # all local VGs
   On primary, get ready for MC/ServiceGuard configuration
    primary# vgchange -a y /dev/vgshare

2.2 Cluster Configuration
  All time intervals, parameters are set as MC/ServiceGuard default. Please notice that the configuration is for functional test purpose only, not a HP supported configuration.
  i Query existing hardware & system configuration (on primary)
   primary# cd /etc/cmcluster
   primary# cmquerycl -v -C /etc/cmcluster/cluster.cfg -n primary -n standby
  ii Modify cluster configuration, check and apply (on primary)
   primary# vi /etc/cmcluster/cluster.cfg
    * refer to appendix A for sample cluster configuration file
   primary# cmcheckconf -v -C /etc/cmcluster/cluster.cfg
   primary# cmapplyconf -v -C /etc/cmcluster/cluster.cfg
  iii Startup cluster (on primary)
   primary# cmruncl -v
   primary# cmviewcl -v   # check point, cluster status is up
  iv Auto Startup (on both primary and standby)
   primary/standby# vi /etc/rc.config.d/cmcluster
    AUTOSTART_CMCLD=0 changed to AUTOSTART_CMCLD=1
2.3 Sample Package and Service Configuration
  A sample package (spkg) is configured with single service (ssvr).  The service is designed to run a sample service script (sscript, refer to appendix D) which is able to display current time and
hostname on console device every ten seconds.
  i Configure package (on primary)
   primary# mkdir /etc/cmcluster/spkg
   primary# cmmakepkg -p /etc/cmcluster/spkg/spkg.cfg
   primary# cmmakepkg -s /etc/cmcluster/spkg/spkg.ctl
   primary# chmod 755 /etc/cmcluster/spkg/spkg.ctl
  ii Modify package configuration (on primary)
   primary# vi /etc/cmcluster/spkg/spkg.cfg
    * refer to appendix B for sample package configuration file
   primary# vi /etc/cmcluster/spkg/spkg.ctl
    * refer to appendix C for sample package control file
  iii Duplicate configuration and control files to other sites (on primary)
   primary# cd /etc/cmcluster
   primary# tar cvf - * | remsh standby '(cd /etc/cmcluster; tar xvpf -)'
   primary# cmhaltcl -fv
   primary# cmcheckconf -v -C /etc/cmcluster/cluster.cfg -P
     /etc/cmcluster/spkg/spkg.cfg
   primary# cmapplyconf -v -C /etc/cmcluster/cluster.cfg -P
    /etc/cmcluster/spkg/spkg.cfg
  iv Startup cluster (on either primary or standby)
   primary/standby# cmruncl -v
   primary/standby# cmviewcl -v   # check point
3 MC/ServiceGuard Installation Acceptance Test
3.1 MC/ServieGuard Package/Service Failure Scenarios
  By killing service script, service and causing package fail.  The package will be restarted at standby node.
  i Startup MC/SG cluster and nodes (on either primary or standby)
   primary/standby# cmruncl -v
  ii Startup standard package and service on primary
   automatically start with cluster, or
   primary# cmrunpkg -v
  iii Verify cluster, nodes, package and services status on primary
   primary# cmviewcl -v
   * make sure package auto-switch flag is enable.
  iv Terminate service manually on primary
   eg by killing process sscript
  v Verify cluster, nodes, package and services status on standby
   standby# cmviewcl -v
  vi Transfer package from standby back to primary manually by shutting down on standby and restarting package on primary
   standby# cmhaltpkg -v spkg
   standby# cmmodpkg -e -n primary spkg
   standby# cmrunpkg -v -n primary spkg
   standby# cmmodpkg -e spkgAppendix A Sample cluster configuration file - /etc/cmcluster/cluster.cfg
CLUSTER_NAME   sample  # fill-in cluster name
FIRST_CLUSTER_LOCK_VG  /dev/vgshare # auto-detected lock VG
        # change if necessary
NODE_NAME    primary  # auto-detected hostname
  NETWORK_INTE**CE  lan0   # auto-detected LAN inte**ce
    HEARTBEAT_IP   192.9.28.1  # auto-detected IP address
  FIRST_CLUSTER_LOCK_PV  /dev/dsk/cXtYd0 # auto-detected lock PV
        # change if necessary
NODE_NAME    standby
  NETWORK_INTE**CE  lan0
    HEARTBEAT_IP   192.9.28.2
  FIRST_CLUSTER_LOCK_PV  /dev/dsk/cXtYd0
HEARTBEAT_INTERVAL  1000000  # default timing parameters
NODE_TIMEOUT   2000000
AUTO_START_TIMEOUT  600000000
NETWORK_POLLING_INTERVAL 2000000
VOLUME_GROUP   /dev/vgshare # auto-detected shared VG
        # change if necessary
Appendix B Sample package configuration file - /etc/cmcluster/spkg/spkg.cfg
PACKAGE_NAME   spkg   # fill-in package name
NODE_NAME    primary  # fill-in primary hostname
NODE_NAME    standby  # fill-in standby hostname
RUN_SCRIPT    /etc/cmcluster/spkg/spkg.ctl
        # fill-in package start
        # control script name
RUN_SCRIPT_TIMEOUT  NO_TIMEOUT  # default timing parameter
HALT_SCRIPT    /etc/cmcluster/spkg/spkg.ctl
        # fill-in package start
        # control script name
HALT_SCRIPT_TIMEOUT  NO_TIMEOUT  # default timing parameter
SERVICE_NAME   ssvr   # fill-in service name
SERVICE_FAIL_FAST_ENABLED NO   # default parameter
SERVICE_HALT_TIMEOUT  300   # default timing parameter
SUBNET    192.9.28.0  # fill-in subnetwork address
PKG_SWITCHING_ENABLED  YES   # default parameter
NET_SWITCHING_ENABLED  YES   # default parameter
NODE_FAIL_FAST_ENABLED  NO   # default parameter
Appendix C Sample package control file - /etc/cmcluster/spkg/spkg.ctl (configuration portion)
PATH=/sbin:/usr/bin:/usr/sbin:/etc:/bin  # default search path
VGCHANGE="vgchange -a e"    # default exclusive VG option
      
VG[0]=/dev/vgshare     # fill-in shared Volume Group
LV[0]="/dev/vgshare/lvol1"; FS[0]="/shared" # fill-in LV and mount point
IP[0]=192.9.28.3      # fill-in relocateable IP
SUBNET[0]=192.9.28.0     # fill-in subnetwork address
SERVICE_NAME[0]=ssvr     # fill-in service name
SERVICE_CMD[0]=/etc/cmcluster/spkg/sscript # fill-in service script name
SERVICE_RESTART[0]=""     # default parameter      
Appendix D Sample service script file - /etc/cmcluster/spkg/sscript
#!/usr/bin/ksh
trap "exit " 15
while true
do
echo "`uname -n` `date`" | tee -a /shared/sscript.log > /dev/console
sleep 10

zfzhuman123

哥 以我们当前的知识水平  还不能理解你这么深奥的知识！给点注释吧哥！

binboys

这种系统现在都是银行使用得多