OCFS2 Integration with HP Serviceguard for Linux Administrator's Guide, Second Edition, May 2009

O2CB_HEARTBEAT_THRESHOLD = (((60 seconds + 30 seconds)/2) + 1)
O2CB_HEARTBEAT_THRESHOLD = 46
For more information on configuring the O2CB_HEARTBEAT_THRESHOLD parameter, see the
Configuring Cluster Timeouts section in the OCFS2 – Frequently Asked Questions document. This
document is available at the following location:
http://oss.oracle.com/projects/ocfs2/dist/documentation/ocfs2_faq.html
Configuring the O2CB_KEEPALIVE_DELAY_MS Parameter
The Network Keep-alive parameter specifies the maximum delay in milliseconds before a
keep-alive packet is sent. This implies that a keep-alive packet is sent if a network connection
between two nodes is silent for the specified duration. If the other node is alive and connected,
it is expected to respond.
In an environment where HP Serviceguard for Linux and OCFS2 co-exist, it is recommended
that the value for the O2CB_KEEPALIVE_DELAY_MS parameter is left at its default value of 2000
milliseconds.
Configuring the O2CB_RECONNECT_DELAY_MS Parameter
The Network Reconnect parameter specifies the minimum delay, in milliseconds, between
connection attempts. The default value is 2000 ms.
In an environment where HP Serviceguard for Linux and OCFS2 co-exist, it is recommended
that the value for the O2CB_RECONNECT_DELAY_MS parameter is left at its default value of 2000
milliseconds.
Creating the OCFS2 File System
After installing OCFS2 in a Serviceguard environment and configuring the parameters, you can
create the OCFS2 file system on all shared devices that you need to include in the cluster. Run
the following command to create an OCFS2 file system:
mkfs.ocfs2 command
For example, to create an OCFS2 file system with a 4K block size and 8K cluster size, which can
be activated on a maximum of 4 nodes on the device /dev/sdf1, run the following command:
# mkfs.ocfs2 -b 4K -C 8K -N 4 /dev/sdf1
Following is the output that you will see:
mkfs.ocfs2 1.2.3
File system label=
Block size=4096 (bits=12)
Cluster size=8192 (bits=13)
Volume size=5358190592 (654076 clusters) (1308152 blocks)
21 cluster groups (tail covers 8956 clusters, rest cover 32256 clusters)
Journal size=67108864
Initial number of node slots: 4
Creating bitmaps: done
Initializing superblock: done
Writing system files: done
Writing superblock: done
Writing backup superblock: 2 block(s)
Formatting Journals: done
Writing lost+found: done
mkfs.ocfs2 successful
Repeat this procedure for all other devices that you plan to use in the cluster.
The value for the -N option in the above example specifies the maximum number of nodes that
can concurrently mount and use the OCFS2 file system configured on the device. This value
specified for the -N option during the creation of the file system describes the number of node
22 Integrating OCFS2 with HP Serviceguard for Linux