HP Serviceguard Extended Distance Cluster for Linux A.12.00.00 Deployment Guide, March 2014

NOTE: If you have added more than one filter in the /etc/lvm/lvm.conf file, the following
warning message will be displayed:
WARNING! Ignoring duplicate config node: filter (seeking filter)
The workaround is to edit the/etc/lvm/lvm.conf file to have only one filter as shown in the
example.
For example, let us assume/etc/lvm/lvm.conf file has two filters:
filter = [ "a|/dev/./by-id/.|", "a|/dev/md0|", "a|/dev/md1|",
"r|/dev/hpdev/md0_mirror0|" , "r|/dev/hpdev/md0_mirror1|" ,
"r|/dev/hpdev/md1_mirror0|" , "r|/dev/hpdev/md1_mirror1|" ] filter = [
"r|/dev/cdrom|" ]
Edit the filter in the/etc/lvm/lvm.conf file:
filter = [ "a|/dev/./by-id/.|", "a|/dev/md0|", "a|/dev/md1|",
"r|/dev/hpdev/md0_mirror0|" , "r|/dev/hpdev/md0_mirror1|" ,
"r|/dev/hpdev/md1_mirror0|" , "r|/dev/hpdev/md1_mirror1|" ,
"r|/dev/cdrom|" ]
Note: Edit /etc/lvm/lvm.conf file for every node that you add to the cluster.
To ensure the exclusive activation of a volume group on a node, you must create /etc/lvm/
lvm_<node_name>.conf file and add the following line:
activation { volume_list=["@[<Fully Qualified Domain Name>]"] }
For example, if the name of the node is node1 and the fully qualified domain name for the node
is node1.ind.hp.com, add the following line in the /etc/lvm/lvm_node1.conf file :
activation { volume_list=["@node1.ind.hp.com"] }
Repeat this procedure for every every node that you add in the cluster.
1. Start the package
Starting a package configured for Software RAID is the same as starting any other package
in Serviceguard for Linux.
You also need to keep in mind a few guidelines before you enable Software RAID for a particular
package. Following are some of these guidelines you need to follow:
Ensure that the Quorum Server link is close to the Ethernet links in your setup. In cases of
failures of all Ethernet and Fibre channel links, the nodes can easily access the Quorum Server
for arbitration.
The Quorum Server is configured in a third location only for arbitration. In scenarios where
the link between two nodes is lost, each node considers the other node to be dead. As a
result, both nodes will try to access the Quorum Server. The Quorum Server, as an arbitrator,
acknowledges the node that reaches it first and allows the package to start on that node.
You also need to configure Network Time Protocol (NTP) in your environment. This protocol
synchronizes the nodes irrespective of whether the nodes are in the same or different time
zone.
3.3.5 Configuring the VxVM Diskgroups
The serviceguard-xdc product supports disk group configuration in a remote mirror to be
site-consistent. When you create volumes in such a disk group, the volumes are mirrored across
all sites by default. For more information on administering sites and remote mirrors, see Veritas
Storage Foundation Cluster File System High Availability Administrator's Guide available at
https://sort.symantec.com/documents/doc_details/sfha/6.0.1/Linux/ProductGuides.
26 Configuring your Environment for Software RAID