<!--

Licensed to the Apache Software Foundation (ASF) under one or more
contributor license agreements. See the NOTICE file distributed with
this work for additional information regarding copyright ownership.
The ASF licenses this file to You under the Apache License, Version 2.0
(the "License"); you may not use this file except in compliance with
the License. You may obtain a copy of the License at

 http://www.apache.org/licenses/LICENSE-2.0

Unless required by applicable law or agreed to in writing, software
distributed under the License is distributed on an "AS IS" BASIS,
WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
See the License for the specific language governing permissions and
limitations under the License.
-->

MiNiFi System Administrator's Guide

Apache NiFi Team

MiNiFi Homepage

This documentation is for MiNiFi 0.6.0.

Automatic Warm-Redeploy

When many MiNiFi agents running on the edge, it may not be possible to manually stop, edit the config.yml and then restart every one every time their configuration needs to change. The Config Change Coordinator and its Ingestors were designed to automatically redeploy in response to a configuration update.

The Config Change Ingestors are the means by which the agent is notified of a potential new configuration. Currently there are three:

After a new configuration has been pulled/received the Ingestors use a Differentiator in order to determine if the currently running config is different than the new config. Which Differentiator is used, is configurable for each Ingestor. Currently there is only one Differentiator:

After a new config is determined to be new, the MiNiFi agent will attempt to restart. The bootstrap first saves the old config into a swap file. The bootstrap monitors the agent as it restarts and if it fails it will roll back to the old config. If it succeeds then the swap file will be deleted and the agent will start processing using the new config.

Note: Data left in connections when the agent attempts to restart will either be mapped to a connection with the same ID in the new config, or orphaned and deleted.

The configuration for Warm-Redeploy is done in the bootstrap.conf and primarily revolve around the Config Change Ingestors. The configuration in the bootstrap.conf is done using the "nifi.minifi.notifier.ingestors" key followed by the full path name of the desired Ingestor implementation to run. Use a comma separated list to define more than one Ingestor implementation. For example:

nifi.minifi.notifier.ingestors=org.apache.nifi.minifi.bootstrap.configuration.ingestors.PullHttpChangeIngestor

Ingestor specific configuration is also necessary and done in the bootstrap.conf as well. Specifics for each are detailed below.

FileChangeIngestor

class name: org.apache.nifi.minifi.bootstrap.configuration.ingestors.FileChangeIngestor

This Config Change Ingestor watches a file and when the file is updated, the file is ingested as a new config.

Note: The config file path configured here and in "nifi.minifi.config" cannot be the same. This is due to the swapping mechanism and other implementation limitations.

Below are the configuration options. The file config path is the only required property.

OptionDescription
nifi.minifi.notifier.ingestors.file.config.pathPath of the file to monitor for changes. When these occur, the FileChangeNotifier, if configured, will begin the configuration reloading process
nifi.minifi.notifier.ingestors.file.polling.period.secondsHow frequently the file specified by 'nifi.minifi.notifier.file.config.path' should be evaluated for changes. If not set then a default polling period of 15 seconds will be used.
nifi.minifi.notifier.ingestors.file.differentiatorWhich differentiator to use. If not set then it uses the WholeConfigDifferentiator as a default.

RestChangeIngestor

class name: org.apache.nifi.minifi.bootstrap.configuration.ingestors.RestChangeIngestor

This Config Change Ingestor sets up a light-weight Jetty HTTP(S) REST service in order to listen to HTTP(S) requests. A potential new configuration is sent via a POST request with the BODY being the potential new config.

Note: The encoding is expected to be Unicode and the exact version specified by the BOM mark ('UTF-8','UTF-16BE' or 'UTF-16LE'). If there is no BOM mark, then UTF-8 is used.

Here is an example post request using 'curl' hitting the local machine on pot 8338 and it is executed with the config file "config.yml" in the directory the command is run from:

curl --request POST --data-binary "@config.yml" http://localhost:8338/

Below are the configuration options. There are no required options. If no properties are set then the server will bind to hostname "localhost" on a random open port, will only connect via HTTP and will use the WholeConfigDifferentiator.

OptionDescription
nifi.minifi.notifier.ingestors.receive.http.hostHostname on which the Jetty server will bind to. If not specified then it will bind to localhost.
nifi.minifi.notifier.ingestors.receive.http.portPort on which the Jetty server will bind to. If not specified then it will bind to a random open port.
nifi.minifi.notifier.ingestors.receive.http.truststore.locationIf using HTTPS, this specifies the location of the truststore.
nifi.minifi.notifier.ingestors.receive.http.truststore.passwordIf using HTTPS, this specifies the password of the truststore.
nifi.minifi.notifier.ingestors.receive.http.truststore.typeIf using HTTPS, this specifies the type of the truststore.
nifi.minifi.notifier.ingestors.receive.http.keystore.locationIf using HTTPS, this specifies the location of the keystore.
nifi.minifi.notifier.ingestors.receive.http.keystore.passwordIf using HTTPS, this specifies the password of the keystore.
nifi.minifi.notifier.ingestors.receive.http.keystore.typeIf using HTTPS, this specifies the type of the keystore.
nifi.minifi.notifier.ingestors.receive.http.need.client.authIf using HTTPS, this specifies whether or not to require client authentication.
nifi.minifi.notifier.ingestors.receive.http.differentiatorWhich differentiator to use. If not set then it uses the WholeConfigDifferentiator as a default.

PullHttpChangeIngestor

class name: org.apache.nifi.minifi.bootstrap.configuration.ingestors.PullHttpChangeIngestor

This Config Change Ingestor periodically sends a GET request to a REST endpoint using HTTP(S) to order to pull the potential new config.

Below are the configuration options. The hostname and port are the only required properties.

OptionDescription
nifi.minifi.notifier.ingestors.pull.http.hostnameHostname on which to pull configurations from
nifi.minifi.notifier.ingestors.pull.http.portPort on which to pull configurations from
nifi.minifi.notifier.ingestors.pull.http.proxy.hostnameProxy server hostname
nifi.minifi.notifier.ingestors.pull.http.proxy.portProxy server port
nifi.minifi.notifier.ingestors.pull.http.proxy.usernameProxy username
nifi.minifi.notifier.ingestors.pull.http.proxy.passwordProxy password
nifi.minifi.notifier.ingestors.pull.http.pathPath on which to pull configurations from
nifi.minifi.notifier.ingestors.pull.http.period.msPeriod on which to pull configurations from, defaults to 5 minutes if not set.
nifi.minifi.notifier.ingestors.pull.http.queryQuerystring value for the URL
nifi.minifi.notifier.ingestors.pull.http.use.etagIf the destination server is set up with cache control ability and utilizes an "ETag" header, then this should be set to true to utilize it. Very simply, the Ingestor remembers the "ETag" of the last successful pull (returned 200) then uses that "ETag" in a "If-None-Match" header on the next request.
nifi.minifi.notifier.ingestors.pull.http.connect.timeout.msSets the connect timeout for new connections. A value of 0 means no timeout, otherwise values must be a positive whole number in milliseconds.
nifi.minifi.notifier.ingestors.pull.http.read.timeout.msSets the read timeout for new connections. A value of 0 means no timeout, otherwise values must be a positive whole number in milliseconds.
nifi.minifi.notifier.ingestors.pull.http.truststore.locationIf using HTTPS, this specifies the location of the truststore.
nifi.minifi.notifier.ingestors.pull.http.truststore.passwordIf using HTTPS, this specifies the password of the truststore.
nifi.minifi.notifier.ingestors.pull.http.truststore.typeIf using HTTPS, this specifies the type of the truststore.
nifi.minifi.notifier.ingestors.pull.http.keystore.locationIf using HTTPS, this specifies the location of the keystore.
nifi.minifi.notifier.ingestors.pull.http.keystore.passwordIf using HTTPS, this specifies the password of the keystore.
nifi.minifi.notifier.ingestors.pull.http.keystore.typeIf using HTTPS, this specifies the type of the keystore.
nifi.minifi.notifier.ingestors.pull.http.differentiatorWhich differentiator to use. If not set then it uses the WholeConfigDifferentiator as a default.

Status Reporting and Querying

In NiFi there is a lot of information, such as stats and bulletins, that is only available to view through the UI. MiNiFi provides access to this information through a query mechanism. You can query FlowStatus either using the MiNiFi.sh script or by configuring one of the Periodic Status Reporters. The API for the query is the same for the reporters and the "flowStatus" script option. The API is outlined in the "FlowStatus Query Options" section below.

FlowStatus Script Query

From the minifi.sh script, you can manually query to get the current status of your dataflow. The following is an example of a minifi.sh query you might run to view health, stats, and bulletins for the TailFile processor. This query returns information to your command-line.

minifi.sh flowStatus processor:TailFile:health,stats,bulletins

Currently the script only accepts one high level option at a time. Also any names of connections, remote process groups, or processors that contain " " (a space), ":", ";" or "," cause parsing errors when querying.

Note: The examples in this documentation are provided for *nix based environments. Windows support is also provided by by the flowstatus-minifi.bat file. To perform one of the listed interactions, minifi.sh flowStatus would simply be replaced by flowstatus-minifi.bat. For example, the sample query above in *nix environments,

minifi.sh flowStatus processor:TailFile:health,stats,bulletins`

would translate to Windows environments as,

flowstatus-minifi.bat processor:TailFile:health,stats,bulletins

Periodic Status Reporters

You can set up Periodic Status Reporters to periodically report the status of your dataflow. The query executes at configurable intervals and the results are reported using the configured implementation. Configure the Reporters in the bootstrap.conf file, using the "nifi.minifi.status.reporter.components" key followed by the full path name of the desired Reporter implementation to run. Use a comma separated list to define more than one Reporter implementation. For example:

nifi.minifi.status.reporter.components=org.apache.nifi.minifi.bootstrap.status.reporters.StatusLogger

StatusLogger

class name: org.apache.nifi.minifi.bootstrap.status.reporters.StatusLogger

The Periodic Status Reporter logs the results of the query to the logs. By default it is logged to the minifi-bootstrap.log but you can modify logback.xml to log to an alternate file and location.

OptionDescription
nifi.minifi.status.reporter.log.queryThe FlowStatus query to run.
nifi.minifi.status.reporter.log.levelThe log level at which to log the status. Available options are "TRACE", "DEBUG", "INFO", "WARN" and "ERROR".
nifi.minifi.status.reporter.log.periodThe delay (in milliseconds) between each query.

Example bootstrap.conf configuration:

# The FlowStatus query to submit to the MiNiFi instance
nifi.minifi.status.reporter.log.query=instance:health,bulletins
# The log level at which the status will be logged
nifi.minifi.status.reporter.log.level=INFO
# The period (in milliseconds) at which to log the status
nifi.minifi.status.reporter.log.period=60000

Example logback.xml configuration to output the status to its own rolling log file:

<appender name="STATUS_LOG_FILE" class="ch.qos.logback.core.rolling.RollingFileAppender">
    <file>logs/minifi-status.log</file>
    <rollingPolicy class="ch.qos.logback.core.rolling.TimeBasedRollingPolicy">
        <!--
          For daily rollover, use 'user_%d.log'.
          For hourly rollover, use 'user_%d{yyyy-MM-dd_HH}.log'.
          To GZIP rolled files, replace '.log' with '.log.gz'.
          To ZIP rolled files, replace '.log' with '.log.zip'.
        -->
        <fileNamePattern>./logs/minifi-status_%d.log</fileNamePattern>
        <!-- keep 5 log files worth of history -->
        <maxHistory>5</maxHistory>
    </rollingPolicy>
    <encoder class="ch.qos.logback.classic.encoder.PatternLayoutEncoder">
        <pattern>%date %level [%thread] %logger{40} %msg%n</pattern>
    </encoder>
</appender>

<logger name="org.apache.nifi.minifi.bootstrap.status.reporters.StatusLogger" level="INFO" additivity="false">
    <appender-ref ref="STATUS_LOG_FILE" />
</logger>

FlowStatus Query Options

This section outlines each option to query the MiNiFi instance for the FlowStatus.

Processors

To query the processors use the "processor" flag followed by the processor ID or name, to get (or "all") followed by one of the processor options. The processor options are below. Note: In Windows environments, all minifi.sh flowStatus invocations should be replaced with flowstatus-minifi.bat. See FlowStatus Script Query for an illustration.

OptionDescription
healthThe processor's run status, whether or not it has bulletins and the validation errors (if there are any).
bulletinsA list of all the current bulletins (if there are any).
statsThe current stats of the processor. This includes but is not limited to active threads and FlowFiles sent/received.

An example query to get the health, bulletins and stats of the "TailFile" processor is below.

minifi.sh flowStatus processor:TailFile:health,stats,bulletins

Connections

To query the connections use the "connection" flag followed by the connection ID or name, to get (or "all") followed by one of the connection options. The connection options are below.

OptionDescription
healthThe connections's queued bytes and queued FlowFile count.
statsThe current stats of the connection. This includes input/output count and input/output bytes.

An example query to get the health and stats of the "TailToS2S" connection is below.

minifi.sh flowStatus connection:TailToS2S:health,stats

Remote Process Groups

To query the remote process groups (RPG) use the "remoteProcessGroup" flag followed by the RPG ID or name to get (or "all") followed by one of the remote process group options. The remote process group options are below.

OptionDescription
healthThe connections's queued bytes and queued FlowFile count.
bulletinsA list of all the current bulletins (if there are any).
inputPortsA list of every input port for this RPG and their status. Their status includes its name, whether the target exists and whether it's currently running.
outputPortsA list of every output port for this RPG and their status. Their status includes its name, whether the target exists and whether it's currently running.
statsThe current stats of the RPG. This includes the active threads, sent content size and count.

An example query to get the health, bulletins, input ports and stats of all the RPGS is below.

minifi.sh flowStatus remoteprocessgroup:all:health,bulletins,inputports,stats

Controller Services

To query the controller services use the "controllerServices" flag followed by one of the controller service options. The controller service options are below.

OptionDescription
healthThe controller service's state, whether or not it has bulletins and any validation errors.
bulletinsA list of all the current bulletins (if there are any).

An example query to get the health and bulletins of all the controller services is below.

minifi.sh flowStatus controllerservices:health,bulletins

Provenance Reporting

To query the status of the provenance reporting use the "provenancereporting" flag followed by one of the provenance reporting options. The provenance reporting options are below.

OptionDescription
healthThe provenance reporting state, active threads, whether or not it has bulletins and any validation errors.
bulletinsA list of all the current bulletins (if there are any).

An example query to get the health and bulletins of the provenance reporting is below.

minifi.sh flowStatus provenancereporting:health,bulletins

Instance

To query the status of the MiNiFi instance use the "instance" flag followed by one of the instance options. The instance options are below.

OptionDescription
healthThe instance reporting state, active threads, whether or not it has bulletins and any validation errors.
bulletinsA list of all the current bulletins (if there are any).
statsThe current stats of the instance. This including but not limited to bytes read/written and FlowFiles sent/transferred.

An example query to get the health, stats and bulletins of the instance is below.

minifi.sh flowStatus instance:health,stats,bulletins

System Diagnostics

To query the system diagnostics use the "systemdiagnostics" flag followed by one of the system diagnostics options. The system diagnostics options are below.

OptionDescription
heapInformation detailing the state of the JVM heap.
processorstatsThe system processor stats. This includes the available processors and load average.
contentrepositoryusageA list of each content repository and stats detailing its usage.
flowfilerepositoryusageStats about the current usage of the FlowFile repository.
garbagecollectionA list of the garbage collection events, detailing their name, collection count and time.

An example query to get the heap, processor stats, content repository usage, FlowFile repository usage and garbage collection from the system diagnostics is below.

minifi.sh flowStatus systemdiagnostics:heap,processorstats,contentrepositoryusage,flowfilerepositoryusage,garbagecollection

Example

This is an example of a simple query to get the health of all the processors and its results from a simple flow:

User:minifi-0.0.1-SNAPSHOT user ./bin/minifi.sh flowStatus processor:all:health

Java home: /Library/Java/JavaVirtualMachines/jdk1.8.0_74.jdk/Contents/Home
MiNiFi home: /Users/user/projects/nifi-minifi/minifi-assembly/target/minifi-0.0.1-SNAPSHOT-bin/minifi-0.0.1-SNAPSHOT

Bootstrap Config File: /Users/user/projects/nifi-minifi/minifi-assembly/target/minifi-0.0.1-SNAPSHOT-bin/minifi-0.0.1-SNAPSHOT/conf/bootstrap.conf

{"controllerServiceStatusList":null,"processorStatusList":[{"name":"Connection Diagnostics","processorHealth":{"runStatus":"Running","hasBulletins":false,"validationErrorList":[]},"processorStats":null,"bulletinList":null},{"name":"UpdateAttribute","processorHealth":{"runStatus":"Running","hasBulletins":false,"validationErrorList":[]},"processorStats":null,"bulletinList":null},{"name":"Processor Diagnostics","processorHealth":{"runStatus":"Running","hasBulletins":false,"validationErrorList":[]},"processorStats":null,"bulletinList":null},{"name":"System Diagnostics","processorHealth":{"runStatus":"Running","hasBulletins":false,"validationErrorList":[]},"processorStats":null,"bulletinList":null},{"name":"GenerateFlowFile","processorHealth":{"runStatus":"Running","hasBulletins":false,"validationErrorList":[]},"processorStats":null,"bulletinList":null}],"connectionStatusList":null,"remoteProcessGroupStatusList":null,"instanceStatus":null,"systemDiagnosticsStatus":null,"reportingTaskStatusList":null,"errorsGeneratingReport":[]}

Periodic Status Reporters

Config File

The config.yml in the conf directory is the main configuration file for controlling how MiNiFi runs. This section provides an overview of the properties in this file. The file is a YAML
and follows the YAML format laid out here.

Alternatively, the MiNiFi Toolkit Converter can aid in creating a config.yml from a generated template exported from a NiFi instance. This
tool can be downloaded from https://nifi.apache.org/minifi/download.html under the MiNiFi Toolkit Binaries section. Information on the toolkit's usage is
available at https://nifi.apache.org/minifi/minifi-toolkit.html.

Note: Values for periods of time and data sizes must include the unit of measure,
for example "10 sec" or "10 MB", not simply "10".

Versioning

The "MiNiFi Config Version" property is used to indicate to the configuration parser which version of the config file it is looking at. If the property is empty or missing, version 1 is assumed.

The MiNiFi Toolkit Converter is capable of parsing previous versions (possibly subject to a future deprecation policy) and writing out the current version. It can also validate that a given config file
parses and upconverts to the current version without issue.

Version 1 -> Version 2 changes

  1. Use ids instead of names for processors, connections.
  2. Allow multiple source relationships for connections.
  3. Added support for process groups, and internal input ports an output ports.
  4. Change Id Key for RPGs from "Remote Processing Groups" to the proper "Remote Process Groups" (not "ing").

Version 2 -> Version 3 changes

  1. Added support for Controller Services.
  2. Added support for Site-To-Site over proxy.
  3. Added support for overriding nifi.properties values
  4. Added support for Output Ports to Remote Process Groups

Flow Controller

The first section of config.yml is for naming and commenting on the file.

PropertyDescription
MiNiFi Config VersionThe version of the configuration file. The default value if this property is missing or empty is 1, the current value is 2.
nameThe name of the file.
commentA comment describing the usage of this config file.

Core Properties

The Core Properties section applies to the core framework as a whole.

PropertyDescription
flow controller graceful shutdown periodIndicates the shutdown period. The default value is 10 sec.
flow service write delay intervalWhen many changes are made to the flow.xml, this property specifies how long to wait before writing out the changes, so as to batch the changes into a single write. The default value is 500 ms.
administrative yield durationIf a component allows an unexpected exception to escape, it is considered a bug. As a result, the framework will pause (or administratively yield) the component for this amount of time. This is done so that the component does not use up massive amounts of system resources, since it is known to have problems in the existing state. The default value is 30 sec.
bored yield durationWhen a component has no work to do (i.e., is "bored"), this is the amount of time it will wait before checking to see if it has new data to work on. This way, it does not use up CPU resources by checking for new work too often. When setting this property, be aware that it could add extra latency for components that do not constantly have work to do, as once they go into this "bored" state, they will wait this amount of time before checking for more work. The default value is 10 millis.
max concurrent threadsThe maximum number of threads any processor can have running at one time.

FlowFile Repository

The FlowFile repository keeps track of the attributes and current state of each FlowFile in the system. By default,
this repository is installed in the same root installation directory as all the other repositories; however, it is advisable
to configure it on a separate drive if available.

PropertyDescription
partitionsThe number of partitions. The default value is 256.
checkpoint intervalThe FlowFile Repository checkpoint interval. The default value is 2 mins.
always syncIf set to true, any change to the repository will be synchronized to the disk, meaning that NiFi will ask the operating system not to cache the information. This is very expensive and can significantly reduce NiFi performance. However, if it is false, there could be the potential for data loss if either there is a sudden power loss or the operating system crashes. The default value is false.

Swap Subsection

A part of the FlowFile Repository section there is a Swap subsection.

NiFi keeps FlowFile information in memory (the JVM)
but during surges of incoming data, the FlowFile information can start to take up so much of the JVM that system performance
suffers. To counteract this effect, NiFi "swaps" the FlowFile information to disk temporarily until more JVM space becomes
available again. The "Swap" subsection of properties govern how that process occurs.

PropertyDescription
thresholdThe queue threshold at which NiFi starts to swap FlowFile information to disk. The default value is 20000.
in periodThe swap in period. The default value is 5 sec.
in threadsThe number of threads to use for swapping in. The default value is 1.
out periodThe swap out period. The default value is 5 sec.
out threadsThe number of threads to use for swapping out. The default value is 4.

Content Repository

The Content Repository holds the content for all the FlowFiles in the system. By default, it is installed in the same root
installation directory as all the other repositories; however, administrators will likely want to configure it on a separate
drive if available. If nothing else, it is best if the Content Repository is not on the same drive as the FlowFile Repository.
In dataflows that handle a large amount of data, the Content Repository could fill up a disk and the
FlowFile Repository, if also on that disk, could become corrupt. To avoid this situation, configure these repositories on different drives.

PropertyDescription
content claim max appendable sizeThe maximum size for a content claim. The default value is 10 MB.
content claim max flow filesThe maximum number of FlowFiles to assign to one content claim. The default value is 100.
always syncIf set to true, any change to the repository will be synchronized to the disk, meaning that NiFi will ask the operating system not to cache the information. This is very expensive and can significantly reduce NiFi performance. However, if it is false, there could be the potential for data loss if either there is a sudden power loss or the operating system crashes. The default value is false.

Provenance Repository

PropertyDescription
provenance rollover timeThe amount of time to wait before rolling over the latest data provenance information so that it is available to be accessed by components. The default value is 1 min.
implementationThe implementation of ProvenanceRepository to use. The default value is org.apache.nifi.provenance.MiNiFiPersistentProvenanceRepository.

Component Status Repository

The Component Status Repository contains the information for the Component Status History tool in the User Interface. These
properties govern how that tool works.

The buffer.size and snapshot.frequency work together to determine the amount of historical data to retain. As an example to
configure two days worth of historical data with a data point snapshot occurring every 5 minutes you would configure
snapshot.frequency to be "5 mins" and the buffer.size to be "576". To further explain this example for every 60 minutes there
are 12 (60 / 5) snapshot windows for that time period. To keep that data for 48 hours (12 * 48) you end up with a buffer size
of 576.

PropertyDescription
buffer sizeSpecifies the buffer size for the Component Status Repository. The default value is 1440.
snapshot frequencyThis value indicates how often to present a snapshot of the components' status history. The default value is 1 min.

Security Properties

These properties pertain to various security features in NiFi. Many of these properties are covered in more detail in the
Security Configuration section of this Administrator's Guide.

PropertyDescription
keystoreThe full path and name of the keystore. It is blank by default.
keystore typeThe keystore type. It is blank by default.
keystore passwordThe keystore password. It is blank by default.
key passwordThe key password. It is blank by default.
truststoreThe full path and name of the truststore. It is blank by default.
truststore typeThe truststore type. It is blank by default.
truststore passwordThe truststore password. It is blank by default.
ssl protocolThe protocol to use when communicating via https. Necessary to transfer provenance securely.

Note: A StandardSSLContextService will be made automatically with the ID "SSL-Context-Service" if "ssl protocol" is configured.

Sensitive Properties Subsection

Some properties for processors are marked as sensitive and should be encrypted. These following properties will be used to encrypt the properties while in use by MiNiFi. This will currently not be used to encrypt properties in the config file.

PropertyDescription
keyThis is the password used to encrypt any sensitive property values that are configured in processors. By default, it is blank, but the system administrator should provide a value for it. It can be a string of any length, although the recommended minimum length is 10 characters. Be aware that once this password is set and one or more sensitive processor properties have been configured, this password should not be changed.
algorithmThe algorithm used to encrypt sensitive properties. The default value is PBEWITHMD5AND256BITAES-CBC-OPENSSL.
providerThe sensitive property provider. The default value is BC.

Processors

The current implementation of MiNiFi supports multiple processors. The "Processors" subsection is a list of these processors. Each processor must specify these properties. They are the basic configuration general to all processor implementations. Make sure that all relationships for a processor are accounted for in the auto-terminated relationship list or are used in a connection.

PropertyDescription
nameThe name of what this processor will do. This is not used for any underlying implementation but solely for the users of this configuration and MiNiFi agent.
idThe id of this processor. This can be omitted but in processors without this field, there should not be any duplicate names and connections will need to specify source and destination name instead of id. If set it should be a filesystem-friendly value (regex: [A-Za-z0-9_-]+)
classThe fully qualified java class name of the processor to run. For example for the standard TailFile processor it would be: org.apache.nifi.processors.standard.TailFile
max concurrent tasksThe maximum number of tasks that the processor will use.
scheduling strategyThe strategy for executing the processor. Valid options are CRON_DRIVEN or TIMER_DRIVEN
scheduling periodThis property expects different input depending on the scheduling strategy selected. For the TIMER_DRIVEN scheduling strategy, this value is a time duration specified by a number followed by a time unit. For example, 1 second or 5 mins. The default value of 0 sec means that the Processor should run as often as possible as long as it has data to process. This is true for any time duration of 0, regardless of the time unit (i.e., 0 sec, 0 mins, 0 days). For an explanation of values that are applicable for the CRON driven scheduling strategy, see the description of the CRON driven scheduling strategy in the scheduling tab section of the NiFi User documentation.
penalization periodSpecifies how long FlowFiles will be penalized.
yield periodIn the event the processor cannot make progress it should yield which will prevent the processor from being scheduled to run for some period of time. That period of time is specific using this property.
run duration nanosIf the processor supports batching this property can be used to control how long the Processor should be scheduled to run each time that it is triggered. Smaller values will have lower latency but larger values will have higher throughput. This period should typically only be set between 0 and 2000000000 (2 seconds).
auto-terminated relationships listA YAML list of the relationships to auto-terminate for the processor.
annotation dataSome processors make use of "Annotation Data" in order to do more complex configuration, such as the Advanced portion of UpdateAttribute. This data will be unique to each implementing processor and more than likely will not be written out manually.

Processor Properties

Within the Processor Configuration section, there is the Properties subsection. The keys and values in this section are the property names and values for the processor. For example the TailFile processor would have a section like this:

Properties:
    File to Tail: logs/nifi-app.log
    Rolling Filename Pattern: nifi-app*
    State File: ./conf/state/tail-file
    Initial Start Position: Beginning of File

Controller Services

The current implementation of MiNiFi supports Controller Services. The "Controller Services" subsection is a list of these services. Each Controller Service must specify the following properties. They are the basic configuration general to all Controller Service implementations.

PropertyDescription
nameThe name of what this Controller Service will do. This is not used for any underlying implementation but solely for the users of this configuration and MiNiFi agent.
idThe id of this Controller Service. This must be a valid UUID. To reference this Controller Service in the properties of another component, this ID is used.
typeThe fully qualified java class name of the processor to run. For example for the standard StandardSSLContextService processor would be: org.apache.nifi.ssl.StandardSSLContextService

Note: If the "Security Properties" is configured with an "ssl protocol" then a StandardSSLContextService will be made automatically with the ID "SSL-Context-Service".

Controller Service Properties

Within the Controller Service Configuration section, there is the Properties subsection. The keys and values in this section are the property names and values for the service. For example the StandardSSLContextService would have a section like this:

Properties:
  Keystore Filename: /tmp/localhost/keystore.jks
  Keystore Password: keystorePassword
  Keystore Type: JKS
  SSL Protocol: TLS
  Truststore Filename: /tmp/truststore.jks
  Truststore Password: truststorePassword
  Truststore Type: JKS
  key-password: keyPassword

Process Groups

Process groups can be nested from the top level. They can contain other process groups as well and can be used to logically group related operations.

PropertyDescription
nameThe name of what this process group will do.
idThe id of this process group. This needs to be set to a unique filesystem-friendly value (regex: [A-Za-z0-9_-]+)
ProcessorsThe processors contained in this Process Group. (Defined above)
Remote Process GroupsThe remote process groups contained in this Process Group. (Defined below)
ConnectionsThe connections contained in this Process Group. (Defined below)
Input PortsThe input ports contained in this Process Group. (Defined below)
Output PortsThe output ports contained in this Process Group. (Defined below)
FunnelsThe funnels contained in this Process Group. (Defined below)
Process GroupsThe child Process Groups contained in this Process Group.

Input Ports

These ports provide input to the Process Group they reside on. (Currently only for internal Input ports.)

PropertyDescription
nameThe name of what this input port will do.
idThe id of this input port. This needs to be set to a unique filesystem-friendly value (regex: [A-Za-z0-9_-]+)

Output Ports

These ports provide output from the Process Group they reside on. (Currently only for internal Output ports.)

PropertyDescription
nameThe name of what this output port will do.
idThe id of this output port. This needs to be set to a unique filesystem-friendly value (regex: [A-Za-z0-9_-]+)

Funnels

Funnels can be used to combine outputs from multiple processors into a single connection for ease of design.

PropertyDescription
idThe id of this funnel. This needs to be set to a unique filesystem-friendly value (regex: [A-Za-z0-9_-]+)

Connections

There can be multiple connections in this version of MiNiFi. The "Connections" subsection is a list of connections. Each connection must specify these properties.

PropertyDescription
nameThe name of what this connection will do. This is used for the id of the connection so it must be unique.
idThe id of this connection. This needs to be left empty or set to a filesystem-friendly value (regex: [A-Za-z0-9_-]+)
source nameThe name of what of the processor that is the source for this connection.
source relationship nameThe name of the processors relationship to route to this connection
destination nameThe name of the component to receive this connection.
max work queue sizeThis property is the max number of FlowFiles that can be in the queue before back pressure is applied. When back pressure is applied the source processor will no longer be scheduled to run.
max work queue data sizeThis property specifies the maximum amount of data (in size) that should be queued up before applying back pressure. When back pressure is applied the source processor will no longer be scheduled to run.
flowfile expirationIndicates how long FlowFiles are allowed to exist in the connection before be expired (automatically removed from the flow).
queue prioritizer classThis configuration option specifies the fully qualified java class path of a queue prioritizer to use. If no special prioritizer is desired then it should be left blank. An example value of this property is: org.apache.nifi.prioritizer.NewestFlowFileFirstPrioritizer

Remote Process Groups

MiNiFi can be used to send data using the Site to Site protocol (via a Remote Process Group) or a Processor. These properties configure the Remote Process Groups that use Site-To-Site to send data to a core instance. The proxy settings are for HTTP Site-To-Site.

PropertyDescription
nameThe name of what this Remote Process Group points to. This is not used for any underlying implementation but solely for the users of this configuration and MiNiFi agent.
commentA comment about the Remote Process Group. This is not used for any underlying implementation but solely for the users of this configuration and MiNiFi agent.
urlThe URL of the core NiFi instance.
timeoutHow long MiNiFi should wait before timing out the connection.
yield periodWhen communication with this Remote Process Group fails, it will not be scheduled again for this amount of time.
transport protocolThe transport protocol to use for this Remote Process Group. Can be either "RAW" or "HTTP"
proxy hostThe hostname of the proxy server
proxy portThe port to connect to on the proxy server
proxy userThe user name on the proxy server
proxy passwordThe password for the proxy server
Input PortsThe Input Ports for this Remote Process Group (See below)
Output PortsThe Output Ports for this Remote Process Group (See below)

Input/Output Ports Subsection

When connecting via Site to Site, MiNiFi needs to know which input or output port to communicate to of the core NiFi instance. These properties designate and configure communication with that port.

PropertyDescription
idThe id of the port as it exists on the core NiFi instance. To get this information access the UI of the core instance, right click the port that is desired to be connect to and select "configure". The id of the port should under the "Id" section.
nameThe name of the port as it exists on the core NiFi instance. To get this information access the UI of the core instance, right click the port that is desired to be connect to and select "configure". The name of the port should under the "Port name" section.
comments:A comment about the Port. This is not used for any underlying implementation but solely for the users of this configuration and MiNiFi agent.
max concurrent tasksThe number of tasks that this port should be scheduled for at maximum.
use compressionWhether or not compression should be used when communicating with the port. This is a boolean value of either "true" or "false"

Provenance Reporting

MiNiFi is currently designed only to report provenance data using the Site to Site protocol. These properties configure the underlying reporting task that sends the provenance events.

PropertyDescription
commentA comment about the Provenance reporting. This is not used for any underlying implementation but solely for the users of this configuration and MiNiFi agent.
scheduling strategyThe strategy for executing the Reporting Task. Valid options are CRON_DRIVEN or TIMER_DRIVEN
scheduling periodThis property expects different input depending on the scheduling strategy selected. For the TIMER_DRIVEN scheduling strategy, this value is a time duration specified by a number followed by a time unit. For example, 1 second or 5 mins. The default value of 0 sec means that the Processor should run as often as possible as long as it has data to process. This is true for any time duration of 0, regardless of the time unit (i.e., 0 sec, 0 mins, 0 days). For an explanation of values that are applicable for the CRON driven scheduling strategy, see the description of the CRON driven scheduling strategy in the scheduling tab section of the NiFi User documentation.
destination urlThe URL to post the Provenance Events to.
port nameThe name of the input port as it exists on the receiving NiFi instance. To get this information access the UI of the core instance, right the input port that is desired to be connect to and select "configure". The id of the port should under the "Port name" section.
originating urlThe URL of this MiNiFi instance. This is used to include the Content URI to send to the destination.
use compressionIndicates whether or not to compress the events when being sent.
timeoutHow long MiNiFi should wait before timing out the connection.
batch sizeSpecifies how many records to send in a single batch, at most. This should be significantly above the expected amount of records generated between scheduling. If it is not, then there is the potential for the Provenance reporting to lag behind event generation and never catch up.

Note: In order to send via HTTPS, the "Security Properties" must be fully configured. A StandardSSLContextService will be made automatically with the ID "SSL-Context-Service" and used by the Provenance Reporting.

NiFi Properties Overrides

This is a yaml map that contains values to be put into nifi.properties. This will supercede any hardcoded or other schema values that are substituted into nifi.properties file.

Example NiFi Properties Overrides

NiFi Properties Overrides:
  nifi.flowfile.repository.directory: ./flowfile_repository_override
  nifi.content.repository.directory.default: ./content_repository_override
  nifi.database.directory: ./database_repository_override

Running as a Windows Service

MiNiFi can run as a Windows service. To do so, you must modify the conf/bootstrap.conf to set absolute paths for some properties. The properties are:

You can now install the MiNiFi service by running the install-service.bat script. To remove the service run the delete-service.bat file.

The minifi.exe in MiNiFi bin directory is used to run MiNiFi Windows service. The bundled one is for 64 bit architecture and requires 64 bit JRE. If you have to use 32 bit JRE for some reason, you need to replace the minifi.exe file with the one for 32 bit to make MiNiFi service runs successfully. To do so, go to Commons Daemon project download page, download the binary (e.g. commons-daemon-1.1.0-bin.zip), extract it and replace bin/minifi.exe by copying commons-daemon-x.x.x-bin/prunsrv.exe into MiNiFi bin directory as minifi.exe to overwrite the 64 bit exe with the 32 bit one.

Example Config File

Below are two example config YAML files. The first tails the minifi-app.log, send the tailed log and provenance data back to a secure instance of NiFi. The second uses a series of processors to tail the app log, routes off only lines that contain "WriteAheadFlowFileRepository" and puts it as a file in the "./" directory.

MiNiFi Config Version: 1
Flow Controller:
    name: MiNiFi Flow
    comment:

Core Properties:
    flow controller graceful shutdown period: 10 sec
    flow service write delay interval: 500 ms
    administrative yield duration: 30 sec
    bored yield duration: 10 millis

FlowFile Repository:
    partitions: 256
    checkpoint interval: 2 mins
    always sync: false
    Swap:
        threshold: 20000
        in period: 5 sec
        in threads: 1
        out period: 5 sec
        out threads: 4

Provenance Repository:
    provenance rollover time: 1 min

Content Repository:
    content claim max appendable size: 10 MB
    content claim max flow files: 100
    always sync: false

Component Status Repository:
    buffer size: 1440
    snapshot frequency: 1 min

Security Properties:
    keystore: /tmp/ssl/localhost-ks.jks
    keystore type: JKS
    keystore password: localtest
    key password: localtest
    truststore: /tmp/ssl/localhost-ts.jks
    truststore type: JKS
    truststore password: localtest
    ssl protocol: TLS
    Sensitive Props:
        key:
        algorithm: PBEWITHMD5AND256BITAES-CBC-OPENSSL
        provider: BC

Processors:
    - name: TailFile
      class: org.apache.nifi.processors.standard.TailFile
      max concurrent tasks: 1
      scheduling strategy: TIMER_DRIVEN
      scheduling period: 1 sec
      penalization period: 30 sec
      yield period: 1 sec
      run duration nanos: 0
      auto-terminated relationships list:
      Properties:
          File to Tail: logs/minifi-app.log
          Rolling Filename Pattern: minifi-app*
          Initial Start Position: Beginning of File

Connections:
    - name: TailToS2S
      source name: TailFile
      source relationship name: success
      destination name: 8644cbcc-a45c-40e0-964d-5e536e2ada61
      max work queue size: 0
      max work queue data size: 1 MB
      flowfile expiration: 60 sec
      queue prioritizer class: org.apache.nifi.prioritizer.NewestFlowFileFirstPrioritizer

Remote Processing Groups:
    - name: NiFi Flow
      comment:
      url: https://localhost:8090/nifi
      timeout: 30 secs
      yield period: 10 sec
      Input Ports:
          - id: 8644cbcc-a45c-40e0-964d-5e536e2ada61
            name: tailed log
            comments:
            max concurrent tasks: 1
            use compression: false

Provenance Reporting:
    comment:
    scheduling strategy: TIMER_DRIVEN
    scheduling period: 30 sec
    destination url: https://localhost:8090/
    port name: provenance
    originating url: http://${hostname(true)}:8081/nifi
    use compression: true
    timeout: 30 secs
    batch size: 1000
Flow Controller:
    name: MiNiFi Flow
    comment:

Core Properties:
    flow controller graceful shutdown period: 10 sec
    flow service write delay interval: 500 ms
    administrative yield duration: 30 sec
    bored yield duration: 10 millis
    max concurrent threads: 1

FlowFile Repository:
    partitions: 256
    checkpoint interval: 2 mins
    always sync: false
    Swap:
        threshold: 20000
        in period: 5 sec
        in threads: 1
        out period: 5 sec
        out threads: 4

Content Repository:
    content claim max appendable size: 10 MB
    content claim max flow files: 100
    always sync: false

Component Status Repository:
    buffer size: 1440
    snapshot frequency: 1 min

Security Properties:
    keystore: /tmp/ssl/localhost-ks.jks
    keystore type: JKS
    keystore password: localtest
    key password: localtest
    truststore: /tmp/ssl/localhost-ts.jks
    truststore type: JKS
    truststore password: localtest
    ssl protocol: TLS
    Sensitive Props:
        key:
        algorithm: PBEWITHMD5AND256BITAES-CBC-OPENSSL
        provider: BC

Processors:
    - name: TailAppLog
      class: org.apache.nifi.processors.standard.TailFile
      max concurrent tasks: 1
      scheduling strategy: TIMER_DRIVEN
      scheduling period: 10 sec
      penalization period: 30 sec
      yield period: 1 sec
      run duration nanos: 0
      auto-terminated relationships list:
      Properties:
          File to Tail: logs/minifi-app.log
          Rolling Filename Pattern: minifi-app*
          Initial Start Position: Beginning of File
    - name: SplitIntoSingleLines
      class: org.apache.nifi.processors.standard.SplitText
      max concurrent tasks: 1
      scheduling strategy: TIMER_DRIVEN
      scheduling period: 0 sec
      penalization period: 30 sec
      yield period: 1 sec
      run duration nanos: 0
      auto-terminated relationships list:
          - failure
          - original
      Properties:
          Line Split Count: 1
          Header Line Count: 0
          Remove Trailing Newlines: true
    - name: RouteErrors
      class: org.apache.nifi.processors.standard.RouteText
      max concurrent tasks: 1
      scheduling strategy: TIMER_DRIVEN
      scheduling period: 0 sec
      penalization period: 30 sec
      yield period: 1 sec
      run duration nanos: 0
      auto-terminated relationships list:
          - unmatched
          - original
      Properties:
          Routing Strategy: Route to 'matched' if line matches all conditions
          Matching Strategy: Contains
          Character Set: UTF-8
          Ignore Leading/Trailing Whitespace: true
          Ignore Case: true
          Grouping Regular Expression:
          WALFFR: WriteAheadFlowFileRepository
    - name: PutFile
      class: org.apache.nifi.processors.standard.PutFile
      max concurrent tasks: 1
      scheduling strategy: TIMER_DRIVEN
      scheduling period: 0 sec
      penalization period: 30 sec
      yield period: 1 sec
      run duration nanos: 0
      auto-terminated relationships list:
          - failure
          - success
      Properties:
          Directory: ./
          Conflict Resolution Strategy: replace
          Create Missing Directories: true
          Maximum File Count:
          Last Modified Time:
          Permissions:
          Owner:
          Group:

Connections:
    - name: TailToSplit
      source name: TailAppLog
      source relationship name: success
      destination name: SplitIntoSingleLines
      max work queue size: 0
      max work queue data size: 1 MB
      flowfile expiration: 60 sec
      queue prioritizer class: org.apache.nifi.prioritizer.NewestFlowFileFirstPrioritizer
    - name: SplitToRoute
      source name: SplitIntoSingleLines
      source relationship name: splits
      destination name: RouteErrors
      max work queue size: 0
      max work queue data size: 1 MB
      flowfile expiration: 60 sec
      queue prioritizer class: org.apache.nifi.prioritizer.NewestFlowFileFirstPrioritizer
    - name: RouteToS2S
      source name: RouteErrors
      source relationship name: matched
      destination name: PutFile
      max work queue size: 0
      max work queue data size: 1 MB
      flowfile expiration: 60 sec
      queue prioritizer class: org.apache.nifi.prioritizer.NewestFlowFileFirstPrioritizer

Provenance Reporting:
    comment:
    scheduling strategy: TIMER_DRIVEN
    scheduling period: 30 sec
    destination url: https://localhost:8080/
    port name: provenance
    originating url: http://${hostname(true)}:8081/nifi
    use compression: true
    timeout: 30 secs
    batch size: 1000