HDFS-7186. Document the "hadoop trace" command. (Masatake Iwasaki via Colin P. McCabe)

This commit is contained in:
Colin Patrick Mccabe 2014-10-07 11:55:49 -07:00
parent 2e789eb226
commit 9196db9a08
2 changed files with 66 additions and 12 deletions

View File

@ -350,6 +350,9 @@ Release 2.7.0 - UNRELEASED
HADOOP-11156. DelegateToFileSystem should implement HADOOP-11156. DelegateToFileSystem should implement
getFsStatus(final Path f). (Zhihai Xu via wang) getFsStatus(final Path f). (Zhihai Xu via wang)
HDFS-7186. Document the "hadoop trace" command. (Masatake Iwasaki via Colin
P. McCabe)
OPTIMIZATIONS OPTIMIZATIONS
BUG FIXES BUG FIXES

View File

@ -46,21 +46,32 @@ public void receiveSpan(Span span);
+---- +----
<property> <property>
<name>hadoop.trace.spanreceiver.classes</name> <name>hadoop.htrace.spanreceiver.classes</name>
<value>org.htrace.impl.LocalFileSpanReceiver</value> <value>org.htrace.impl.LocalFileSpanReceiver</value>
</property> </property>
<property> <property>
<name>hadoop.local-file-span-receiver.path</name> <name>hadoop.htrace.local-file-span-receiver.path</name>
<value>/var/log/hadoop/htrace.out</value> <value>/var/log/hadoop/htrace.out</value>
</property> </property>
+---- +----
You can omit package name prefix if you use span receiver bundled with HTrace.
+----
<property>
<name>hadoop.htrace.spanreceiver.classes</name>
<value>LocalFileSpanReceiver</value>
</property>
+----
** Setting up ZipkinSpanReceiver ** Setting up ZipkinSpanReceiver
Instead of implementing SpanReceiver by yourself, Instead of implementing SpanReceiver by yourself,
you can use <<<ZipkinSpanReceiver>>> which uses you can use <<<ZipkinSpanReceiver>>> which uses
{{{https://github.com/twitter/zipkin}Zipkin}} {{{https://github.com/twitter/zipkin}Zipkin}}
for collecting and dispalying tracing data. for collecting and displaying tracing data.
In order to use <<<ZipkinSpanReceiver>>>, In order to use <<<ZipkinSpanReceiver>>>,
you need to download and setup {{{https://github.com/twitter/zipkin}Zipkin}} first. you need to download and setup {{{https://github.com/twitter/zipkin}Zipkin}} first.
@ -82,22 +93,63 @@ public void receiveSpan(Span span);
+---- +----
<property> <property>
<name>hadoop.trace.spanreceiver.classes</name> <name>hadoop.htrace.spanreceiver.classes</name>
<value>org.htrace.impl.ZipkinSpanReceiver</value> <value>ZipkinSpanReceiver</value>
</property> </property>
<property> <property>
<name>hadoop.zipkin.collector-hostname</name> <name>hadoop.htrace.zipkin.collector-hostname</name>
<value>192.168.1.2</value> <value>192.168.1.2</value>
</property> </property>
<property> <property>
<name>hadoop.zipkin.collector-port</name> <name>hadoop.htrace.zipkin.collector-port</name>
<value>9410</value> <value>9410</value>
</property> </property>
+---- +----
** Turning on tracing by HTrace API
In order to turn on Dapper-like tracing, ** Dynamic update of tracing configuration
You can use <<<hadoop trace>>> command to see and update the tracing configuration of each servers.
You must specify IPC server address of namenode or datanode by <<<-host>>> option.
You need to run the command against all servers if you want to update the configuration of all servers.
<<<hadoop trace -list>>> shows list of loaded span receivers associated with the id.
+----
$ hadoop trace -list -host 192.168.56.2:9000
ID CLASS
1 org.htrace.impl.LocalFileSpanReceiver
$ hadoop trace -list -host 192.168.56.2:50020
ID CLASS
1 org.htrace.impl.LocalFileSpanReceiver
+----
<<<hadoop trace -remove>>> removes span receiver from server.
<<<-remove>>> options takes id of span receiver as argument.
+----
$ hadoop trace -remove 1 -host 192.168.56.2:9000
Removed trace span receiver 1
+----
<<<hadoop trace -add>>> adds span receiver to server.
You need to specify the class name of span receiver as argument of <<<-class>>> option.
You can specify the configuration associated with span receiver by <<<-Ckey=value>>> options.
+----
$ hadoop trace -add -class LocalFileSpanReceiver -Chadoop.htrace.local-file-span-receiver.path=/tmp/htrace.out -host 192.168.56.2:9000
Added trace span receiver 2 with configuration hadoop.htrace.local-file-span-receiver.path = /tmp/htrace.out
$ hadoop trace -list -host 192.168.56.2:9000
ID CLASS
2 org.htrace.impl.LocalFileSpanReceiver
+----
** Starting tracing spans by HTrace API
In order to trace,
you will need to wrap the traced logic with <<tracing span>> as shown below. you will need to wrap the traced logic with <<tracing span>> as shown below.
When there is running tracing spans, When there is running tracing spans,
the tracing information is propagated to servers along with RPC requests. the tracing information is propagated to servers along with RPC requests.
@ -133,7 +185,6 @@ import org.htrace.TraceScope;
+---- +----
import org.apache.hadoop.conf.Configuration; import org.apache.hadoop.conf.Configuration;
import org.apache.hadoop.fs.FsShell; import org.apache.hadoop.fs.FsShell;
import org.apache.hadoop.hdfs.HdfsConfiguration;
import org.apache.hadoop.tracing.SpanReceiverHost; import org.apache.hadoop.tracing.SpanReceiverHost;
import org.apache.hadoop.util.ToolRunner; import org.apache.hadoop.util.ToolRunner;
import org.htrace.Sampler; import org.htrace.Sampler;
@ -146,8 +197,8 @@ public class TracingFsShell {
FsShell shell = new FsShell(); FsShell shell = new FsShell();
conf.setQuietMode(false); conf.setQuietMode(false);
shell.setConf(conf); shell.setConf(conf);
SpanReceiverHost.getInstance(conf);
int res = 0; int res = 0;
SpanReceiverHost.init(new HdfsConfiguration());
TraceScope ts = null; TraceScope ts = null;
try { try {
ts = Trace.startSpan("FsShell", Sampler.ALWAYS); ts = Trace.startSpan("FsShell", Sampler.ALWAYS);
@ -165,5 +216,5 @@ public class TracingFsShell {
+---- +----
$ javac -cp `hadoop classpath` TracingFsShell.java $ javac -cp `hadoop classpath` TracingFsShell.java
$ HADOOP_CLASSPATH=. hdfs TracingFsShell -put sample.txt /tmp/ $ java -cp .:`hadoop classpath` TracingFsShell -ls /
+---- +----