Monitoring and Instrumentation
There are several ways to monitor Spark applications: web UIs, metrics, and external instrumentation.
Every SparkContext launches a web UI, by default on port 4040, that displays useful information about the application. This includes:
- A list of scheduler stages and tasks
- A summary of RDD sizes and memory usage
- Environmental information.
- Information about the running executors
You can access this interface by simply opening
http://<driver-node>:4040 in a web browser.
If multiple SparkContexts are running on the same host, they will bind to successive ports
beginning with 4040 (4041, 4042, etc).
Note that this information is only available for the duration of the application by default.
To view the web UI after the fact, set
spark.eventLog.enabled to true before starting the
application. This configures Spark to log Spark events that encode the information displayed
in the UI to persisted storage.
Viewing After the Fact
If Spark is run on Mesos or YARN, it is still possible to construct the UI of an application through Spark’s history server, provided that the application’s event logs exist. You can start the history server by executing:
This creates a web interface at
http://<server-url>:18080 by default, listing incomplete
and completed applications and attempts.
When using the file-system provider class (see
spark.history.provider below), the base logging
directory must be supplied in the
spark.history.fs.logDirectory configuration option,
and should contain sub-directories that each represents an application’s event logs.
The spark jobs themselves must be configured to log events, and to log them to the same shared,
writeable directory. For example, if the server was configured with a log directory of
hdfs://namenode/shared/spark-logs, then the client-side options would be:
The history server can be configured as follows:
||Memory to allocate to the history server (default: 1g).|
||JVM options for the history server (default: none).|
||The public address for the history server. If this is not set, links to application history may use the internal address of the server, resulting in broken links (default: none).|
Spark configuration options
||Name of the class implementing the application history backend. Currently there is only one implementation, provided by Spark, which looks for application logs stored in the file system.|
For the filesystem history provider, the URL to the directory containing application event
logs to load. This can be a local
|spark.history.fs.update.interval||10s||The period at which the filesystem history provider checks for new or updated logs in the log directory. A shorter interval detects new applications faster, at the expense of more server load re-reading updated applications. As soon as an update has completed, listings of the completed and incomplete applications will reflect the changes.|
|spark.history.retainedApplications||50||The number of application UIs to retain. If this cap is exceeded, then the oldest applications will be removed.|
|spark.history.ui.port||18080||The port to which the web interface of the history server binds.|
Indicates whether the history server should use kerberos to login. This is required
if the history server is accessing HDFS files on a secure Hadoop cluster. If this is
true, it uses the configs
|spark.history.kerberos.principal||(none)||Kerberos principal name for the History Server.|
|spark.history.kerberos.keytab||(none)||Location of the kerberos keytab file for the History Server.|
Specifies whether acls should be checked to authorize users viewing the applications.
If enabled, access control checks are made regardless of what the individual application had
|spark.history.fs.cleaner.enabled||false||Specifies whether the History Server should periodically clean up event logs from storage.|
How often the filesystem job history cleaner checks for files to delete.
Files are only deleted if they are older than
|spark.history.fs.cleaner.maxAge||7d||Job history files older than this will be deleted when the filesystem history cleaner runs.|
|spark.history.fs.numReplayThreads||25% of available cores||Number of threads that will be used by history server to process event logs.|
Note that in all of these UIs, the tables are sortable by clicking their headers, making it easy to identify slow tasks, data skew, etc.
The history server displays both completed and incomplete Spark jobs. If an application makes multiple attempts after failures, the failed attempts will be displayed, as well as any ongoing incomplete attempt or the final successful attempt.
Incomplete applications are only updated intermittently. The time between updates is defined by the interval between checks for changed files (
spark.history.fs.update.interval). On larger clusters the update interval may be set to large values. The way to view a running application is actually to view its own web UI.
Applications which exited without registering themselves as completed will be listed as incomplete —even though they are no longer running. This can happen if an application crashes.
One way to signal the completion of a Spark job is to stop the Spark Context explicitly (
sc.stop()), or in Python using the
with SparkContext() as sc:construct to handle the Spark Context setup and tear down.
In addition to viewing the metrics in the UI, they are also available as JSON. This gives developers
an easy way to create new visualizations and monitoring tools for Spark. The JSON is available for
both running applications, and in the history server. The endpoints are mounted at
for the history server, they would typically be accessible at
for a running application, at
In the API, an application is referenced by its application ID,
When running on YARN, each application may have multiple attempts; each identified by their
In the API listed below,
[app-id] will actually be
[base-app-id] is the YARN application ID.
||A list of all applications.
A list of all jobs for a given application.
||Details for the given job.|
||A list of all stages for a given application.|
A list of all attempts for the given stage.
||Details for the given stage attempt|
Summary metrics of all tasks in the given stage attempt.
A list of all tasks for the given stage attempt.
||A list of all executors for the given application.|
||A list of stored RDDs for the given application.|
||Details for the storage status of a given RDD.|
||Download the event logs for all attempts of the given application as files within a zip file.|
||Download the event logs for a specific application attempt as a zip file.|
The number of jobs and stages which can retrieved is constrained by the same retention
mechanism of the standalone Spark UI;
"spark.ui.retainedJobs" defines the threshold
value triggering garbage collection on jobs, and
spark.ui.retainedStages that for stages.
Note that the garbage collection takes place on playback: it is possible to retrieve
more entries by increasing these values and restarting the history server.
API Versioning Policy
These endpoints have been strongly versioned to make it easier to develop applications on top. In particular, Spark guarantees:
- Endpoints will never be removed from one version
- Individual fields will never be removed for any given endpoint
- New endpoints may be added
- New fields may be added to existing endpoints
- New versions of the api may be added in the future at a separate endpoint (eg.,
api/v2). New versions are not required to be backwards compatible.
- Api versions may be dropped, but only after at least one minor release of co-existing with a new api version.
Note that even when examining the UI of a running applications, the
applications/[app-id] portion is
still required, though there is only one application available. Eg. to see the list of jobs for the
running app, you would go to
http://localhost:4040/api/v1/applications/[app-id]/jobs. This is to
keep the paths consistent in both modes.
Spark has a configurable metrics system based on the
Dropwizard Metrics Library.
This allows users to report Spark metrics to a variety of sinks including HTTP, JMX, and CSV
files. The metrics system is configured via a configuration file that Spark expects to be present
$SPARK_HOME/conf/metrics.properties. A custom file location can be specified via the
spark.metrics.conf configuration property.
Spark’s metrics are decoupled into different
instances corresponding to Spark components. Within each instance, you can configure a
set of sinks to which metrics are reported. The following instances are currently supported:
master: The Spark standalone master process.
applications: A component within the master which reports on various applications.
worker: A Spark standalone worker process.
executor: A Spark executor.
driver: The Spark driver process (the process in which your SparkContext is created).
Each instance can report to zero or more sinks. Sinks are contained in the
ConsoleSink: Logs metrics information to the console.
CSVSink: Exports metrics data to CSV files at regular intervals.
JmxSink: Registers metrics for viewing in a JMX console.
MetricsServlet: Adds a servlet within the existing Spark UI to serve metrics data as JSON data.
GraphiteSink: Sends metrics to a Graphite node.
Slf4jSink: Sends metrics to slf4j as log entries.
Spark also supports a Ganglia sink which is not included in the default build due to licensing restrictions:
GangliaSink: Sends metrics to a Ganglia node or multicast group.
To install the
GangliaSink you’ll need to perform a custom build of Spark. Note that
by embedding this library you will include LGPL-licensed
code in your Spark package. For sbt users, set the
SPARK_GANGLIA_LGPL environment variable before building. For Maven users, enable
-Pspark-ganglia-lgpl profile. In addition to modifying the cluster’s Spark build
user applications will need to link to the
The syntax of the metrics configuration file is defined in an example configuration file,
Several external tools can be used to help profile the performance of Spark jobs:
- Cluster-wide monitoring tools, such as Ganglia, can provide insight into overall cluster utilization and resource bottlenecks. For instance, a Ganglia dashboard can quickly reveal whether a particular workload is disk bound, network bound, or CPU bound.
- OS profiling tools such as dstat, iostat, and iotop can provide fine-grained profiling on individual nodes.
- JVM utilities such as
jstackfor providing stack traces,
jmapfor creating heap-dumps,
jstatfor reporting time-series statistics and
jconsolefor visually exploring various JVM properties are useful for those comfortable with JVM internals.