Josh-D. S. Davis

Xaminmo / Omnimax / Max Omni / Mad Scientist / Midnight Shadow / Radiation Master

Previous Entry Share Next Entry
PowerHA quickbuild
Josh 2014
joshdavis
Because Facebook notes editor has zero formatting functionality in the new version.


####################################
### POWERHA QUICKBUILD - SANITIZED
####################################

This is a list of all the commands I'm using to build this cluster.
It's been sanitized of any customer information.


####################################
### Cleanup
####################################

clrmclstr
rmcluster -n MYCLUSTER
y | rmcluster -r hdisk2
rmdev -Rdl cluster0
/usr/sbin/rsct/bin/cthagsctrl -z
/usr/sbin/rsct/bin/cthagsctrl -d
echo "cthags 12348/udp" >> /etc/services
/usr/sbin/rsct/bin/cthagsctrl -a
/usr/sbin/rsct/bin/cthagsctrl -s
stopsrc -s clcomd ; sleep 2 ; startsrc -s clcomd
rm /var/hacmp/adm/* /var/hacmp/log/* /var/hacmp/clverify/* /usr/es/sbin/cluster/etc/auto_versync.pid
no -po nonlocsrcroute=1
no -po ipsrcrouterecv=1
shutdown -Fr now


####################################
### System config
####################################

# oslevel -s
7100-04-01-1543

# halevel -s
7.1.3 SP4

# emgr -P
PACKAGE INSTALLER LABEL
======================================================== =========== ==========
openssl.base installp 101a_fix
bos.net.tcp.client installp IV79944s1a
openssh.base.server installp IV80743m9a
openssh.base.client installp IV80743m9a
bos.net.tcp.client installp IV80191s1a
bos.rte.control installp IV80586s1a

# cat /etc/hosts
127.0.0.1 localhost
10.0.0.1 gateway
10.0.0.10 mycluster MYCLUSTER
10.0.0.11 node1
10.0.0.12 node2


####################################
### Cluster communication
####################################

echo node1 > /etc/cluster/rhosts
echo node2 >> /etc/cluster/rhosts
cat /etc/cluster/chosts > /usr/es/sbin/cluster/etc/rhosts
echo 10.0.0.11 >> /usr/es/sbin/cluster/etc/rhosts
echo 10.0.0.12 >> /usr/es/sbin/cluster/etc/rhosts
echo 10.0.0.1 > /usr/es/sbin/cluster/netmon.cf
stopsrc -s clcomd ; sleep 2 ; startsrc -s clcomd
sleep 10
cl_rsh -n node1 date
cl_rsh -n node2 date

####################################
### Basic cluster build
####################################

export CLUSTER=MYCLUSTER
export NODES="node2 node1"
export HBPVID=deadbeefcafe1234
clmgr add cluster ${CLUSTER} NODES="$NODES"
clmgr modify cluster $CLUSTER REPOSITORY=$HBPVID HEARTBEAT_TYPE=unicast
cldare -rt


####################################
### Add the service address
####################################

/usr/es/sbin/cluster/utilities/claddnode -Tservice -Bmycluster -wnet_ether_01 # -zignore
cllsif
cldare -rt


####################################
### file collections
####################################

clfilecollection -o coll -c Configuration_Files -'' -'AIX and HACMP config files' yes yes
clfilecollection -o coll -c HACMP_Files -'' -'HACMP Resource Group Files' yes yes
clfilecollection -o time -c 10
clfilecollection -o coll -a User_Files 'System user config' yes yes
clfilecollection -o file -a User_Files /etc/passwd
clfilecollection -o file -a User_Files /etc/group
clfilecollection -o file -a User_Files /etc/security/passwd
clfilecollection -o file -a User_Files /etc/security/limits
clfilecollection -o file -a User_Files /.profile
clfilecollection -o file -a User_Files /etc/environment
clfilecollection -o file -a User_Files /etc/profile
clfilecollection -o file -a User_Files /etc/exports
clfilecollection -o file -a User_Files /etc/sudoers
clfilecollection -o file -a User_Files /etc/qconfig
clfilecollection -o file -l Configuration_Files
clfilecollection -o file -l HACMP_Files
clfilecollection -o file -l User_Files


####################################
### mail events
####################################

/usr/es/sbin/cluster/utilities/claddcustom -t event -n'mail_event' \
-I'mail out when event occurs' -v'/usr/local/cluster/mail_event'
for EVENT in `cat /usr/local/cluster/mail_event.list`; do
/usr/es/sbin/cluster/utilities/clchevent -O"$EVENT" \
-s /usr/es/sbin/cluster/events/$EVENT -b mail_event -c 0
done
/usr/es/sbin/cluster/utilities/clacdNM -MA -nLVM_IO_FAIL -p0 -lLVM_IO_FAIL -m/usr/local/cluster/LVM_IO_FAIL
/usr/es/sbin/cluster/utilities/claddserv -s'my_app' \
-b'/usr/local/cluster/APP_start.ksh' -e'/usr/local/cluster/APP_stop.ksh'
/usr/es/sbin/cluster/utilities/claddserv -s'my_dsmc' \
-b'/usr/local/cluster/DSMC_start.ksh' -e'/usr/local/cluster/DSMC_stop.ksh'
cllsserv


####################################
### Resource group
####################################

/usr/es/sbin/cluster/utilities/claddgrp -g 'myclster_rg' -n 'node2 node1' -S 'OFAN' -O 'FNPN' -B 'FBHPN'
cllsgrp


####################################
### Resources
####################################

/usr/es/sbin/cluster/utilities/claddres -g 'myclster_rg' SERVICE_LABEL='myclster' \
APPLICATIONS='my_app my_dsmc' VOLUME_GROUP='prdappvg prdvg prdjrnvg' \
FORCED_VARYON='false' VG_AUTO_IMPORT='false' FILESYSTEM= FSCHECK_TOOL='fsck' \
RECOVERY_METHOD='sequential' PPRC_REP_RESOURCE='' FS_BEFORE_IPADDR='false' \
EXPORT_FILESYSTEM='' ERCMF_REP_RESOURCE='' MOUNT_FILESYSTEM='' \
NFS_NETWORK='' SHARED_TAPE_RESOURCES='' DISK='' AIX_FAST_CONNECT_SERVICES='' \
COMMUNICATION_LINKS='' MISC_DATA='' WPAR_NAME='' GMD_REP_RESOURCE='' SVCPPRC_REP_RESOURCE=''
cllsres
cllsres -g myclster_rg


####################################
### Application monitor
####################################

/usr/es/sbin/cluster/utilities/claddappmon MONITOR_TYPE=process name=my_dsmc_mon \
RESOURCE_TO_MONITOR=my_dsmc INVOCATION='longrunning' PROCESSES='dsm.opt.cluster' \
PROCESS_OWNER=root STABILIZATION_INTERVAL='60' RESTART_COUNT='3' FAILURE_ACTION='notify' \
INSTANCE_COUNT=1 RESTART_INTERVAL=360 NOTIFY_METHOD='/usr/local/cluster/mail_event' \
CLEANUP_METHOD='/usr/local/cluster/DSMC_stop.ksh' \
RESTART_METHOD='/usr/local/cluster/DSMC_start.ksh'
/usr/es/sbin/cluster/utilities/claddappmon name=my_app_mon \
RESOURCE_TO_MONITOR=my_app INVOCATION='both' MONITOR_TYPE=user \
STABILIZATION_INTERVAL=120 MONITOR_INTERVAL=120 \
RESTART_COUNT=3 RESTART_INTERVAL=800 FAILURE_ACTION=fallover \
NOTIFY_METHOD=/usr/local/cluster/mail_event FAILURE_ACTION='notify' \
CLEANUP_METHOD='/usr/local/cluster/APP_stop.ksh' \
RESTART_METHOD='/usr/local/cluster/APP_start.ksh' \
MONITOR_METHOD=/usr/local/cluster/APP_check.ksh HUNG_MONITOR_SIGNAL=9
cllsappmon
cllsappmon my_app_mon
cllsappmon my_dsmc_mon


####################################
### Sync all the changes
####################################

cldare -rt -C interactive


####################################
### Verify both nodes see it fine
####################################

cllsclstr
lscluster -m


####################################
### Start the cluster
####################################

smitty clstart

This is where it complains that hags is not up.
Rebooting does not bring up hags.
Manually starting, and it wil die after 20 mins or so.
Very little logging.

EVERY time I try to mess with HA, it's broken. It's always something different. Such a pain. Truly, I don't know why people do not just use their own scripts.

REF: http://omnitech.net/xaminmo/2016/04/06/powerha-quickbuild/
Like:

?

Log in

No account? Create an account