This is an automated email from the ASF dual-hosted git repository.
kenn pushed a commit to branch asf-site
in repository https://gitbox.apache.org/repos/asf/beam-site.git
commit 0bc08e3eec87fc7ba936b63e73adca1c57500b88
Author: xiliu <xi...@linkedin.com>
AuthorDate: Wed Jun 6 16:16:51 2018 -0700
Update the option docs
---
src/documentation/runners/samza.md | 22 ++++++++++++----------
1 file changed, 12 insertions(+), 10 deletions(-)
diff --git a/src/documentation/runners/samza.md
b/src/documentation/runners/samza.md
index fe48576..2b63189 100644
--- a/src/documentation/runners/samza.md
+++ b/src/documentation/runners/samza.md
@@ -72,7 +72,7 @@ The Samza Runner is built on Samza version greater than
0.14.1, and uses Scala v
## Executing a pipeline with Samza Runner
-If you run your pipeline locally or deploy it to a standalone cluster bundled
with all the jars and resource files, no packaging is required. For example,
the following command runs the WordCount example:
+If you run your pipeline locally or deploy it to a standalone cluster with all
the jars and resource files, no packaging is required. For example, the
following command runs the WordCount example:
```
$ mvn exec:java -Dexec.mainClass=org.apache.beam.examples.WordCount \
@@ -82,7 +82,7 @@ $ mvn exec:java
-Dexec.mainClass=org.apache.beam.examples.WordCount \
--output=/path/to/counts"
```
-To deploy your pipeline to a YARN cluster, you need to package your
application jars and resource files into a `.tgz` archive file. In your config,
you need to specify the URI of the TGZ file for Samza Runner to download:
+To deploy your pipeline to a YARN cluster, here is the
[instructions](https://samza.apache.org/startup/hello-samza/latest/) of
deploying a sample Samza job. First you need to package your application jars
and resource files into a `.tgz` archive file, and make it available to
download for Yarn containers. In your config, you need to specify the URI of
this TGZ file location:
```
yarn.package.path=${your_job_tgz_URI}
@@ -93,7 +93,9 @@ job.coordinator.system=${job_coordinator_system}
job.default.system=${job_default_system}
```
-The config file can be passed to Samza Runner by setting the command line arg
`--configFilePath=/path/to/config.properties`. For more details on the Samza
configuration, see [Samza Configuration
Reference](https://samza.apache.org/learn/documentation/latest/jobs/configuration-table.html).
+For more details on the configuration, see [Samza Configuration
Reference](https://samza.apache.org/learn/documentation/latest/jobs/configuration-table.html).
+
+The config file will be passed in by setting the command line arg
`--configFilePath=/path/to/config.properties`. With that, you can run your main
class of Beam pipeline in a Yarn Resource Manager, and the Samza Runner will
submit a Yarn job under the hood.
## Pipeline options for the Samza Runner
@@ -111,14 +113,14 @@ When executing your pipeline with the Samza Runner, you
can use the following pi
<td>Set to <code>SamzaRunner</code> to run using Samza.</td>
</tr>
<tr>
- <td><code>samzaConfig</code></td>
- <td>The config for Samza runner.</td>
- <td>Config for running locally.</td>
+ <td><code>configFilePath</code></td>
+ <td>The config for Samza using a properties file.</td>
+ <td><code>empty</code>, i.e. use local execution.</td>
</tr>
<tr>
- <td><code>ConfigFilePath</code></td>
- <td>The config for Samza runner using a properties file.</td>
- <td><code>empty</code>, i.e. use default samzaConfig</td>
+ <td><code>configOverride</code></td>
+ <td>The config override to set programmatically.</td>
+ <td><code>empty</code>, i.e. use config file or local execution.</td>
</tr>
<tr>
<td><code>watermarkInterval</code></td>
@@ -146,4 +148,4 @@ When executing your pipeline with the Samza Runner, you can
use the following pi
You can monitor your pipeline job using metrics emitted from both Beam and
Samza, e.g. Beam source metrics such as `elements_read` and `backlog_elements`,
and Samza job metrics such as `job-healthy` and `process-envelopes`. A complete
list of Samza metrics is in [Samza Metrics
Reference](https://samza.apache.org/learn/documentation/latest/container/metrics-table.html).
You can view your job's metrics via JMX in development, and send the metrics
to graphing system such as [Graphite](http: [...]
-For a running Samza YARN job, you can use YARN web UI to monitor the job
status and check logs.
\ No newline at end of file
+For a running Samza YARN job, you can use YARN web UI to monitor the job
status and check logs.
