natas_hw
/
cassandra
mirror of https://github.com/apache/cassandra
0
0
Fork 0
Code Issues Packages Projects Releases Wiki Activity

Fix word count 

Patch by  Ala' Alkhaldi, reviewed by brandonwilliams for CASSANDRA-7200
This commit is contained in:
Brandon Williams 2014-07-18 12:31:17 -05:00
parent 062addb4c5
commit 337e4a8292
6 changed files with 38 additions and 8 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

6
examples/hadoop_cql3_word_count/README.txt
View File

@ -36,8 +36,10 @@ Read the code in src/ for more details.
The word_count_counters example sums the counter columns for a row. The output
is written to a text file in /tmp/word_count_counters.
*If you want to point wordcount at a real cluster, modify the seed
and listenaddress settings accordingly.
*It is recommended to turn off vnodes when running Cassandra with hadoop.
This is done by setting "num_tokens: 1" in cassandra.yaml. If you want to
point wordcount at a real cluster, modify the seed and listenaddress
settings accordingly.
Troubleshooting

2
examples/hadoop_cql3_word_count/conf/log4j.properties
View File

@ -15,7 +15,7 @@
# specific language governing permissions and limitations
# under the License.
log4j.rootLogger=DEBUG,stdout,F
log4j.rootLogger=INFO,stdout,F
#stdout
log4j.appender.stdout=org.apache.log4j.ConsoleAppender

5
examples/hadoop_cql3_word_count/src/WordCountSetup.java
View File

@ -70,7 +70,10 @@ public class WordCountSetup
client.execute_cql3_query(ByteBufferUtil.bytes(query), Compression.NONE, ConsistencyLevel.ONE);
int magnitude = client.describe_ring(WordCount.KEYSPACE).size();
String verifyQuery = "select count(*) from system.peers";
CqlResult result = client.execute_cql3_query(ByteBufferUtil.bytes(verifyQuery), Compression.NONE, ConsistencyLevel.ONE);
long magnitude = ByteBufferUtil.toLong(result.rows.get(0).columns.get(0).value);
try
{
Thread.sleep(1000 * magnitude);

6
examples/hadoop_word_count/README.txt
View File

@ -37,8 +37,10 @@ Read the code in src/ for more details.
The word_count_counters example sums the counter columns for a row. The output
is written to a text file in /tmp/word_count_counters.
*If you want to point wordcount at a real cluster, modify the seed
and listenaddress settings accordingly.
*It is recommended to turn off vnodes when running Cassandra with hadoop.
This is done by setting "num_tokens: 1" in cassandra.yaml. If you want to
point wordcount at a real cluster, modify the seed and listenaddress
settings accordingly.
Troubleshooting

2
examples/hadoop_word_count/conf/log4j.properties
View File

@ -15,7 +15,7 @@
# specific language governing permissions and limitations
# under the License.
log4j.rootLogger=DEBUG,stdout,F
log4j.rootLogger=INFO,stdout,F
#stdout
log4j.appender.stdout=org.apache.log4j.ConsoleAppender

25
examples/hadoop_word_count/src/WordCountSetup.java
View File

@ -176,10 +176,33 @@ public class WordCountSetup
KsDef ksDef = new KsDef(WordCount.KEYSPACE, "org.apache.cassandra.locator.SimpleStrategy", cfDefList);
ksDef.putToStrategy_options("replication_factor", "1");
client.system_add_keyspace(ksDef);
int magnitude = client.describe_ring(WordCount.KEYSPACE).size();
int magnitude = getNumberOfHosts(client);
Uninterruptibles.sleepUninterruptibly(magnitude, TimeUnit.SECONDS);
}
private static int getNumberOfHosts(Cassandra.Iface client)
throws InvalidRequestException, UnavailableException, TimedOutException, TException
{
client.set_keyspace("system");
SlicePredicate predicate = new SlicePredicate();
SliceRange sliceRange = new SliceRange();
sliceRange.setStart(new byte[0]);
sliceRange.setFinish(new byte[0]);
predicate.setSlice_range(sliceRange);
KeyRange keyrRange = new KeyRange();
keyrRange.setStart_key(new byte[0]);
keyrRange.setEnd_key(new byte[0]);
//keyrRange.setCount(100);
ColumnParent parent = new ColumnParent("peers");
List<KeySlice> ls = client.get_range_slices(parent, predicate, keyrRange, ConsistencyLevel.ONE);
return ls.size();
}
private static Cassandra.Iface createConnection() throws TTransportException
{
if (System.getProperty("cassandra.host") == null || System.getProperty("cassandra.port") == null)