Database Reference
In-Depth Information
num=int(raw_input("How many nodes are in your
cluster? "))
for i in range(0, num):
print 'node %d: %d' % (i, (i*(2**127)/num))
7. Update the following property in the
conf/log4j-server.properties
file. Create the
temp
directory under
cassandra
:
Log4j.appender.R.File=/usr/local/cassandra/temp/
system.log
8. Increase the
rpc_timeout
property in
Cassandra.yaml
(if this timeout is
very small and the network latency is high, Cassandra might assume the nodes are
dead without waiting long enough for a response to propagate).
9. Run the Cassandra server at
/usr/local/Cassandra/apache-cas-
sandra-1.1.6 using bin/Cassandra -f
.
10. Run the Cassandra client at
/usr/local/Cassandra/apache-cas-
sandra-1.1.6 using bin/Cassandra-cli
with a host and port.
11. Use the
bin/nodetool
ring utility at
/usr/local/Cassandra/
apache-cassandra-1.1.6
to verify a properly connected cluster:
bin/nodetool -host <ip-adress> -p <port number> ring
192.168.1.30 datacenter1 rack1 Up Normal 755.25
MB 25.00% 0
192.168.1.31 datacenter1 rack1 Up Normal 400.62
MB 25.00% 42535295865117307932921825928970
192.168.1.51 datacenter1 rack1 Up Normal 400.62
MB 25.00% 42535295865117307932921825928971
192.168.1.32 datacenter1 rack1 Up Normal 793.06
MB 25.00% 85070591730234615865843651857941
The preceding output displays a connected cluster. This configuration shows that it's cor-
rectly configured and connected.
Here is a screenshot of the output:
Search WWH ::
Custom Search