HACMP cluster

HACMP cluster
SANDEEP TANTI


Table of Contents
1. definition
1.1. What is HACMP?
1.2. What is SNMP?
1.3. What are the ways of restarting?
2. daemons
2.1. What are the demons HACMP?
2.2. How to know the status of demons HACMP?
3. Logs
3.1. Where are the logs HACMP?
3.2. Recall orders mistake?
3.3. get the error notify?
3.4. Events know how the entrant to the cluster?
4. Identification
4.1. How to identify groups of resources?
4.2. Determine how the resources of a group?
4.3. How to check the location of resources?
4.4. How to display information on all the nodes?
4.5. How to test a disk?
4.6. Finding filesystems mounted with a group voclume?
4.7. How to tell if the volume sharedvg group is currently online?
4.8. How to know the processes that occupy a shared filesystem?
4.9. What are the controls configuration for HACMP know?
4.10. How about the link between physical disks and logical in SSA?
4.11. How to identify network?
5. outages
5.1. How to plant a service?
5.2. How to replace a disk?
5.3. How to change your network card?
6. configuration
6.1. Completing / etc / hosts?
6.2. How do I change the order of the resolutions of names?
6.3. How to create shared disks?
6.4. How to create an alias address?

1. definition
1.1. What is HACMP?
HACMP is the abbreviation of High Availability Cluster Multi-Processing.
1.2. What is SNMP?
Simple Network Management Protocol.
1.3. What are the ways of restarting?
1. HACMP forced stops but resources are conserved
2. HACMP graceful stops but no rocking
3. HACMP takeover stops and there is rocking


2. daemons
2.1. What are the demons HACMP?
1. cluster manager - clstrmgr
2. SMUX peer cluster daemon-clsmuxpd
3. Cluster information services clinfo
4. cluster manager-lock cllockd
5. cluster topology services daemon-topsvcs
6. cluster group services daemon-grpsvcs
7. Global cluster server daemon-grplsm
8. Cluster event management daemon-emsvcs

2.2. How to know the status of demons HACMP?
# lssrc-g cluster
lssrc-g # topsvrc or lssrc-s topsvcs
lssrc-g # grpsvrc or lssrc-s grpsvcs
lssrc-g # grpsvrc or lssrc-s grpgplsm


3. Logs
3.1. Where are the logs HACMP?
1. / var / adm / cluster.log
2. / tmp / hacmp.out
3. / tmp / emuhacmp.out
4. AIX system error log
5. / usr / sbin / cluster / history / clutser.mmdd
6. / tmp / cm.log
7. / tmp / csproc.log

3.2. Recall orders mistake?
# t-errpt
# errpt-a | grep msg

3.3. get the error notify?
# odmget errnotify | grep-i-p x25> foo
# omdadd toto
The modifciations located in the file foo
note: to remove odmdelete
3.4. Know how the events coming to the cluster?
EVENT # grep / tmp / hacmp.out
# cldiag


4. Identification
4.1. How to identify groups of resources?
# clfindres
# clfindres-s

4.2. Determine how the resources of a group?
# clshowres

4.3. How to check the location of resources?
# netstat-i
# df
lsvg #-o

4.4. How to display information on all the nodes?
1. clstat for the console
2. xclstat mode for X11

4.5. How to test a disk?
# dd if = / dev/pdisk0 of = / dev / null bs = 4096
Then test the code return
4.6. Finding filesystems mounted with a group voclume?
Use the script that is among Matilda script.
4.7. How to tell if the volume sharedvg group is currently online?
# if lsvg-o | grep-q-w sharedvg; then
echo sharedvg is online
else
echo sharedvg is offline
fi

4.8. How to know the processes that occupy a shared filesystem?
# fuser-k / dev / sharedlv

4.9. What are the controls configuration for HACMP know?
1. lsattr-El ent0
2. lsattr-El scsi0
3. odmget CnAt | grep-p en0
4. Cc-disk lsdev
5. lsdev-Cc SSAR
6. Cc lsdev-scsi
7. lspv
8. lsvg-o

4.10. How about the link between physical disks and logical in SSA?
# ssaxlate-l hdisk5
# ssaxlate-l pdisk7

4.11. How to identify network?
Determination of roads
# netstat-rn
Determination addresses hardware
# netstat-rn


5. outages
5.1. How to plant a service?
# Cat / etc / file> / dev / kmem

5.2. How to replace a disk?
On the first node:
# rmdev-dl hdisk3
cfgmgr-l scsi2
exportvg sharelvg
recreatevg it sharelvg hdisk3
Attention, the name of LV and FS has changed. We must rename then edit / etc / filesystem <. On the other node: # rmdev-dl hdisk3 # cfgmgr-l scsi2 # importvg it sharelvg hdisk3 c # chdev-l-a hdisk3 pv = clear The pvil passes NONE status. # chdev-l-a hdisk3 pv = yes # recreatevg it sharelvg hdisk3 5.3. How to change your network card? # ifconfig down en0 # rmdev-l en0 # rmdev-l ent0 # chdev-the ent0 '-a media_speed ='100_Full_Duplex' # mkdev-l ent0 # mkdev-l en0 # ifconfig up en0 Attention, never with Auto HACMP. 6. configuration 6.1. Completing / etc / hosts? / etc / hosts is ideal for configurations with more than 25 knots. It must contain all the IP addresses of nodes in the cluster. 6.2. How do I change the order of the resolutions of names? > Edit / etc / netsvc.conf so it looks like this:
= local hosts, nis, bind, dns
The order is of course one you have chosen. Here, the room is the first, second nis ...
You can also delete one way.
NSORDER attention variable premium on the file / etc / netsvc.conf.
For HACMP take into account NIS and DNS requires that the corresponding services are launched.
# Smit HACMP
then cluster configuration, cluster resources, configure run time parameters
6.3. How to create shared disks?
mklv # 's jfslog' sharelvg 1-y 'jfslog2'
# logform / dev/jfslog2
# mklv Does' jfs' sharelvg-y 'mylv' 1
CRFs # 's jfs'-g' sharelvg 'd' mylv '-m / mydata-a log = jfslog2

6.4. How to create an alias address?
ifconfig en0 alias 10.6.100.10
To avoid problems, we must be careful not to use the address to boot.