Veritas CLuster Server Summary of basic cluster operations:
- Cluster Start/Stop:
+ stop VCS on all systems:
# hastop -all
+ stop VCS on hosta and move all groups out:
# hastop -sys hosta -evacuate
+ start VCS on local system:
# hastart
- Users:
+ add gui root user:
# haconf -makerw
# hauser -add root
# haconf -dump -makero
- Group:
+ group start, stop:
# hagrp -offline groupx -sys hostb
# hagrp -online groupx -sys hostb
+ switch a group to other system:
# hagrp -switch groupx -to hosta
+ freeze a group:
# hagrp -freeze groupx
+ unfreeze a group:
# hagrp -unfreeze groupx
+ enable a group:
# hagrp -enable groupx
+ disable a group:
# hagrp -disable groupx
+ enable resources a group:
# hagrp -enableresources groupx
+ disable resources a group:
# hagrp -disableresources groupx
+ flush a group:
# hagrp -flush groupx -sys hosta
- Node:
+ feeze node:
# hasys -freeze hosta
+ thaw node:
# hasys -unfreeze hosta
- Resources:
+ online a resouce:
# hares -online resource_name -sys hosta
+ offline a resouce:
# hares -offline resource_name-sys hosta
+ offline a resouce and propagte to children:
# hares -offprop resource_name -sys hosta
+ probe a resouce:
# hares -probe resource_name -sys hosta
+ clear faulted resource:
# hares -clear resource_name -sys hosta
- Agents:
+ list agent:
# haagent -list
+ list status of an agent:
# haagent -display IP
+ start agent:
# haagent -start IP -sys hosta
+ stop agent:
# haagent -stop IP -sys hosta
- Reboot a node with evacuation of all service groups:
(groupy is running on hosta)
# hastop -sys hosta -evacuate
# init 6
# hagrp -switch groupy -to hosta
- Cluster Start/Stop:
+ stop VCS on all systems:
# hastop -all
+ stop VCS on hosta and move all groups out:
# hastop -sys hosta -evacuate
+ start VCS on local system:
# hastart
- Users:
+ add gui root user:
# haconf -makerw
# hauser -add root
# haconf -dump -makero
- Group:
+ group start, stop:
# hagrp -offline groupx -sys hostb
# hagrp -online groupx -sys hostb
+ switch a group to other system:
# hagrp -switch groupx -to hosta
+ freeze a group:
# hagrp -freeze groupx
+ unfreeze a group:
# hagrp -unfreeze groupx
+ enable a group:
# hagrp -enable groupx
+ disable a group:
# hagrp -disable groupx
+ enable resources a group:
# hagrp -enableresources groupx
+ disable resources a group:
# hagrp -disableresources groupx
+ flush a group:
# hagrp -flush groupx -sys hosta
- Node:
+ feeze node:
# hasys -freeze hosta
+ thaw node:
# hasys -unfreeze hosta
- Resources:
+ online a resouce:
# hares -online resource_name -sys hosta
+ offline a resouce:
# hares -offline resource_name-sys hosta
+ offline a resouce and propagte to children:
# hares -offprop resource_name -sys hosta
+ probe a resouce:
# hares -probe resource_name -sys hosta
+ clear faulted resource:
# hares -clear resource_name -sys hosta
- Agents:
+ list agent:
# haagent -list
+ list status of an agent:
# haagent -display IP
+ start agent:
# haagent -start IP -sys hosta
+ stop agent:
# haagent -stop IP -sys hosta
- Reboot a node with evacuation of all service groups:
(groupy is running on hosta)
# hastop -sys hosta -evacuate
# init 6
# hagrp -switch groupy -to hosta
Here's a procedure to upgrade VCS or shutdown VCS during
hardware maintenance.
1. Open, freeze each Service Group, and close the VCS config.
haconf -makerw
hagrp -freeze -persistent
haconf -dump makero
2. Shutdown VCS but keep services up.
hastop -all -force
3. Confirm VCS has shut down on each system.
gabconfig -a
4. Confirm GAB is not running on any disks.
gabdisk -l (use this if upgrading from VCS 1.1.x)
gabdiskhb -l
gabdiskx -l
If it is, remove it from the disks on each system.
gabdisk -d (use this if upgrading from VCS 1.1.x)
gabdiskhb -d
gabdiskx -d
5. Shutdown GAB and confirm it's down on each system.
gabconfig -U
gabconfig -a
6. Identify the GAB kernel module number and unload it
from each system.
modinfo | grep gab
modunload -i
7. Shutdown LLT. On each system, type:
lltconfig -U
Enter "y" if any questions are asked.
8. Identify the LLT kernel module number and unload it from
each system.
modinfo | grep llt
modunload -i
9. Rename VCS startup and stop scripts on each system.
cd /etc/rc2.d
mv S70llt s70llt
mv S92gab s92gab
cd /etc/rc3.d
mv S99vcs s99vcs
cd /etc/rc0.d
mv K10vcs k10vcs
10. Make a backup copy of /etc/VRTSvcs/conf/config/main.cf.
Make a backup copy of /etc/VRTSvcs/conf/config/types.cf.
Starting with VCS 1.3.0, preonline and other trigger scripts must
be in /opt/VRTSvcs/bin/triggers. Also, all preonline scripts in
previous versions (such as VCS 1.1.2) must now be combined in one
preonline script.
11. Remove old VCS packages.
pkgrm VRTScsga VRTSvcs VRTSgab VRTSllt VRTSperl VRTSvcswz
If you are upgrading from 1.0.1 or 1.0.2, you must also remove the package
VRTSsnmp, and any packages containing a .2 extension, such as VRTScsga.2,
VRTSvcs.2, etc.
Also remove any agent packages such as VRTSvcsix (Informix),
VRTSvcsnb (NetBackup), VRTSvcssor (Oracle), and VRTSvcssy (Sybase).
Install new VCS packages.
Restore your main.cf and types.cf files.
12. Start LLT, GAB and VCS.
cd /etc/rc2.d
mv s70llt S70llt
mv s92gab S92gab
cd /etc/rc3.d
mv s99vcs S99vcs
cd /etc/rc0.d
mv k10vcs K10vcs
/etc/rc2.d/S70llt start
/etc/rc2.d/S92gab
/etc/rc3.d/S99vcs start
13. Check on status of VCS.
hastatus
hastatus -sum
14. Unfreeze all Service Groups.
haconf -makerw
hagrp -unfreeze -persistent
haconf -dump -makero
hardware maintenance.
1. Open, freeze each Service Group, and close the VCS config.
haconf -makerw
hagrp -freeze
haconf -dump makero
2. Shutdown VCS but keep services up.
hastop -all -force
3. Confirm VCS has shut down on each system.
gabconfig -a
4. Confirm GAB is not running on any disks.
gabdisk -l (use this if upgrading from VCS 1.1.x)
gabdiskhb -l
gabdiskx -l
If it is, remove it from the disks on each system.
gabdisk -d (use this if upgrading from VCS 1.1.x)
gabdiskhb -d
gabdiskx -d
5. Shutdown GAB and confirm it's down on each system.
gabconfig -U
gabconfig -a
6. Identify the GAB kernel module number and unload it
from each system.
modinfo | grep gab
modunload -i
7. Shutdown LLT. On each system, type:
lltconfig -U
Enter "y" if any questions are asked.
8. Identify the LLT kernel module number and unload it from
each system.
modinfo | grep llt
modunload -i
9. Rename VCS startup and stop scripts on each system.
cd /etc/rc2.d
mv S70llt s70llt
mv S92gab s92gab
cd /etc/rc3.d
mv S99vcs s99vcs
cd /etc/rc0.d
mv K10vcs k10vcs
10. Make a backup copy of /etc/VRTSvcs/conf/config/main.cf.
Make a backup copy of /etc/VRTSvcs/conf/config/types.cf.
Starting with VCS 1.3.0, preonline and other trigger scripts must
be in /opt/VRTSvcs/bin/triggers. Also, all preonline scripts in
previous versions (such as VCS 1.1.2) must now be combined in one
preonline script.
11. Remove old VCS packages.
pkgrm VRTScsga VRTSvcs VRTSgab VRTSllt VRTSperl VRTSvcswz
If you are upgrading from 1.0.1 or 1.0.2, you must also remove the package
VRTSsnmp, and any packages containing a .2 extension, such as VRTScsga.2,
VRTSvcs.2, etc.
Also remove any agent packages such as VRTSvcsix (Informix),
VRTSvcsnb (NetBackup), VRTSvcssor (Oracle), and VRTSvcssy (Sybase).
Install new VCS packages.
Restore your main.cf and types.cf files.
12. Start LLT, GAB and VCS.
cd /etc/rc2.d
mv s70llt S70llt
mv s92gab S92gab
cd /etc/rc3.d
mv s99vcs S99vcs
cd /etc/rc0.d
mv k10vcs K10vcs
/etc/rc2.d/S70llt start
/etc/rc2.d/S92gab
/etc/rc3.d/S99vcs start
13. Check on status of VCS.
hastatus
hastatus -sum
14. Unfreeze all Service Groups.
haconf -makerw
hagrp -unfreeze
haconf -dump -makero