Clustering Data Collectors with VCS and Veritas NetBackup (RHEL 7)
These instructions cover configuring APTARE data collectors with Veritas Infoscale Availability (VCS) with NetBackup running on Red Hat Enterprise Linux 7.
Prerequisites
• Veritas NetBackup (NBU) is installed and configured on the Veritas Infoscale Availability (VCS) in a clustered mode.
• Veritas NetBackup (NBU) data volume resides on a volume shared across cluster nodes.
• Data Collector is installed on a volume shared across cluster nodes.
Getting with started with Data Collector clustering
1. Install the Veritas NetBackup Data Collector on shared volume attached to the active node.
2. On the same node, delete the start up scripts, by running the command:
# find /etc -name "*aptare*"
3. Manually delete the following files:
/etc/rc.d/rc0.d/K30aptare_agent
/etc/rc.d/rc3.d/S80aptare_agent
/etc/rc.d/rc3.d/K40aptare_agent
/etc/rc.d/rc5.d/S80aptare_agent
/etc/rc.d/rc5.d/K40aptare_agent
4. Update the /etc/hosts file with IP and hostname mapping of aptareagent portal on all the cluster nodes.
5. Using Veritas Infoscale Availability (VCS), create a separate service group for the APTARE agent. The following set of screen shots show the visual representation in VCS. There are multiple methods to set up the configuration. These instructions use the main.cf to setup. See
Main.cf for details.
Main.cf
The main.cf for the previous configuration is as follows. Please note, the following configuration uses example values as required:
include "OracleASMTypes.cf"
include "types.cf"
include "CRSResource.cf"
include "CSSD.cf"
include "Db2udbTypes.cf"
include "MultiPrivNIC.cf"
include "/etc/VRTSvcs/conf/config/NetBackupTypes.cf"
include "OracleTypes.cf"
include "PrivNIC.cf"
include "SybaseTypes.cf"
cluster vcs_cluster (
UserNames = { admin = XXXXXXXXXX }
Administrators = { admin }
)
system node1 (
)
system node2 (
)
group aptare_grp (
SystemList = { node1 = 0, node2 = 1 }
)
Application aptare_app (
StartProgram = "/aptare-vol/mbs/bin/aptare_agent start"
StopProgram = "/aptare-vol/mbs/bin/aptare_agent stop"
CleanProgram = "/aptare-vol/mbs/bin/aptare_agent stop"
PidFiles = { "/aptare-vol/mbs/logs/watchdog.pid",
"/aptare-vol/mbs/logs/metadatacollector.pid",
"/aptare-vol/mbs/logs/validatemetadatacollector.pid" }
)
DiskGroup aptare_dg (
DiskGroup = aptare-dg
)
Mount aptare_mnt (
MountPoint = "/aptare-vol"
BlockDevice = "/dev/vx/dsk/aptare-dg/aptare-vol"
FSType = vxfs
FsckOpt = "-y"
CreateMntPt = 1
)
Volume aptare_vol (
DiskGroup = aptare-dg
Volume = aptare-vol
)
requires group nbu_group online local hard
aptare_app requires aptare_mnt
aptare_mnt requires aptare_vol
aptare_vol requires aptare_dg
// resource dependency tree
//
// group aptare_grp
// {
// Application aptare_app
// {
// Mount aptare_mnt
// {
// Volume aptare_vol
// {
// DiskGroup aptare_dg
// }
// }
// }
// }
group nbu_group (
SystemList = { node1 = 0, node2 = 1 }
AutoStartList = { node1, node2 }
)
DiskGroup nbu_dg (
DiskGroup = nbu-dg
)
IP nbu_ip (
Device @node1 = abc123
Device @node2 = abc123
Address = "11.222.33.44"
NetMask = "222.233.244.0"
)
Mount nbu_mount (
MountPoint = "/opt/VRTSnbu"
BlockDevice = "/dev/vx/dsk/nbu-dg/nbu-vol"
FSType = vxfs
FsckOpt = "-y"
)
NIC nbu_nic (
Device @node1 = abc123
Device @node2 = abc123
NetworkHosts = { "11.222.33.444", "11.222.33.445" }
)
NetBackup nbu_server (
ResourceOwner = unknown
ServerName = NBU_Server
ServerType = NBUMaster
MonScript = NONE
RSPFile = "/usr/openv/netbackup/bin/cluster/NBU_RSP"
GroupName = nbu_group
)
Volume nbu_vol (
DiskGroup = nbu-dg
Volume = nbu-vol
)
nbu_ip requires nbu_nic
nbu_mount requires nbu_vol
nbu_server requires nbu_ip
nbu_server requires nbu_mount
nbu_vol requires nbu_dg
// resource dependency tree
//
// group nbu_group
// {
// NetBackup nbu_server
// {
// IP nbu_ip
// {
// NIC nbu_nic
// }
// Mount nbu_mount
// {
// Volume nbu_vol
// {
// DiskGroup nbu_dg
// }
// }
// }
// }
Upgrading a Clustered Data Collector
1. Using Veritas Infoscale Availability (VCS), verify the service group can be probed:
# /opt/VRTS/bin/hagroup -probe aptare_grp -sys <Node>
2. Freeze the "aptare_grp" cluster resource temporarily while the portal is being upgraded:
hagrp -freeze aptare_grp -sys <Node>
3. On the shared volume, delete the restore.txt if present at /aptare-vol/upgrade/restore.txt.
4. In the APTARE IT Analytics portal, perform the data collector upgrade using (Admin>Data Collection>Collector Upgrade); or wait for the automatic upgrade if the Data Collector is configured for it.
5. Verify the data collector has been upgraded to higher version by validating aptare-vol/aptare/upgrade/version.txt
6. In Veritas Infoscale Availability (VCS) unfreeze the "aptare_grp" cluster resource after the data collector has upgraded successfully:
#hagrp -unfreeze aptare_grp -sys <Node>