- Home
- Unravel 4.6.2.1 Documentation
- Configuring Unravel
- OnDemand reports
- OnDemand configurations
OnDemand configurations
These configurations control the features which use OnDemand:
File Reports
Forecasting
Small Files
Cluster Optimization
Cluster KPIS
Queue Analysis
Top X (including User Reports)
Cluster Discovery
Cloud Mapping Per Host
Services and Versions Compatibility
Required
Property/Description | Set by user | Unit | Default |
---|---|---|---|
unravel.server.ip FQDN or IP-Address of Unravel Server | Required | string | - |
com.unraveldata.python.enabled Enable/disable all ondemand reports and Sessions features in UI. (This property is configured during the OnDemand installation.) | boolean | true |
Property/Description | Set by user | Unit | Default |
---|---|---|---|
unravel.hive.server2.host FQDN or IP-Address of the HiveServer2 instance. | string | - | |
unravel.hive.server2.port Port for the HiveServer2 instance. | number | 10000 |
Property/Description | Set by user | Unit | Default |
---|---|---|---|
com.unraveldata.ngui.proxy.celery | string (URL) | http://localhost:5000 | |
unravel.celery.broker.url | Optional | string (URL) | - |
unravel.celery.result.backend | Optional | - |
Property/Description | Set by user | Unit | Default |
---|---|---|---|
javax.jdo.option.ConnectionDriverName JDBC Driver class name for the hive metastore containing the metadata Examples: MySQL: Oracle: Microsoft: | Required | string | - |
javax.jdo.option.ConnectionPassword Password used to access the metastore. | Required | string | - |
javax.jdo.option.ConnectionUserName Username used to access the metastore. | Required | string | - |
javax.jdo.option.ConnectionURL JDBC connection string for the metastore containing the metadata of the form: jdbc: Example: Oracle: Microsoft: | Required | string (URL) | - |
If KERBEROS
(unravel.hive.server2.authentication=KERBEROS)
set unravel.hive.server2.kerberos.service.name=hive
define Kerberos properties.
See Configuring Forecasting and Migration Planning Reports for required cluster configurations.
See Configuring FSImage for complete configuration requirements.Configuring FSImage
Configuration properties for reports
Migration planning
Note
Note: As of 29 July 2019, the default is false for new installations for Unravel 4.5.3.0 and later.
Property/Description | Set by user | Unit | Default |
---|---|---|---|
unravel.python.reporting.cloudreport.enable Controls the generation of Migration Reports.
| boolean | true (See note.) |
Queue analysis
Property/Description | Set by user | Unit | Default |
---|---|---|---|
com.unraveldata.report.queue.metrics.sensor.enabled Enables or disables the queue metric sensor. | boolean | true | |
com.unraveldata.report.queue.http.retries The number YARN Resource Manager HTTP connection retries. | count | 3 | |
com.unraveldata.report.queue.http.retry.period.msec Period to wait between attempts to connect to the YARN Resource Manager HTTP. | ms | 0 | |
com.unraveldata.report.queue.http.timeout.msec YARN Resource Manager HTTP connection timeout. | ms | 10000 | |
com.unraveldata.report.queue.poll.interval.msec How often queue metric sensor polls YARN Resource Manager. | ms | 60000 | |
unravel.python.queueanalysis.daterange.span UI report date picker range. | days | 30 | |
unravel.python.queueanalysis.metrics.scale UI rendered graph metrics scale factor. | integer | 1000 | |
com.unraveldata.report.excluded.queues The queues to exclude from metric collection and the queue analysis report. |
| CSL | - |
Sessions
Property/Description | Set by user | Unit | Default |
---|---|---|---|
com.unraveldata.session.enabled Enables Sessions features tab in the UI.
You must restart the | boolean | true | |
com.unraveldata.session.max.autotune.runs Maximum number of runs allowed in an auto-tune session. | count | 8 | |
com.unraveldata.session.dynamicAllocation.enabled If set to true Spark sessions recommends dynamic allocation. If users don't want that they need to explicitly set it to false. | boolean | true | |
com.unraveldata.session.spark1.submit.command The command to use for Spark1 applications while applying the session run. This path should be fully qualified. | string | spark-submit | |
com.unraveldata.session.spark2.submit.command The command to use for Spark2 applications while applying the session run. This path should be fully qualified. | string | spark2-submit |
Small files and file reports
Property/Description | Set by user | Unit | Default |
---|---|---|---|
com.unraveldata.ngui.sfhivetable.schedule.interval Frequency, in days, in which to trigger FSimage extraction, for example, every 3 days The scheduler schedules extraction relative to the 1st of the month and then sets each extraction such that it 1st, 1st + X days, 1st + 2X days until 1st + nX days crosses into the next month, at which point the schedule resets to the 1st. See below for an example. Format: | day | 1d | |
com.unraveldata.ngui.sfhivetable.schedule.time Specify the time to download in hours (using 24 hour time) the FSimage. Format: two digits between 00 and 23. | two digits (set member) | 00 |
Note
You must restart the unravel_ondemand
and unravel_ngui
daemons for any configuration changes to take effect. Execute the following commands:
/etc/init.d/unravel_ngui restart /etc/init.d/unravel_ondemand restart
Property/Description | Set by user | Unit | Default |
---|---|---|---|
unravel.python.reporting.files.disable Enables or disables Unravel ability to generate Small Files and File Reports. Note
| boolean | false | |
unravel.python.reporting.files.hive.database Hive Database where Ondemand creates five Hive tables (four temporary, one permanent) for Small Files/File Reports. When not set, tables are created in the default Hive Database. In addition, the Hive queries used for this feature run against the default MR queue. It must point to a valid Hive database. | string | default database | |
unravel.python.reporting.files.hive.mr.queue The Hive queries ran by the Ondemand process run against this MR queue. It must point to a valid MR queue. | string | default |
Note
These are global properties that apply to both Small Files Report and File Reports. They each have equivalent "local" properties. The "local" property takes precedence over the equivalent global property.
Should you unset/delete any of the below properties and their equivalent properties in Small Files Report or Files Report, Unravel has hard-coded values to ensure your reports are generated.
Property/Description | Set by user | Unit | Default |
---|---|---|---|
unravel.python.reporting.files.files.use.avg.file.size_flag
| Optional | boolean | - |
unravel.python.reporting.files.min.parent.dir.depth Directory depth to end the search at. For instance, if depth=2, a directory having depth more than or equal to 2 can be included in the report. Thus a directory such as | Optional | count | - |
unravel.python.reporting.files.max.parent.dir.depth Directory depth to end search at. The maximum is 50. For instance, if depth=5, a directory having a depth less than or equal to 5 can be included in the report. Thus a directory, for example, | Optional | count | - |
unravel.python.reporting.drill.down.subdirs.flag Determines to which ancestor directories a filesize is accounted for. If this is True then the file size is accounted for all ancestor directories. If this is False the file size is accounted for only the parent directory.
For example, given a file under a directory /one/two/three. when when /one - lists files in one. /one/two - lists files in /one/two. true: / - lists files in /, /one, and /one/two. /one - lists files in /one, and /one/two. /one/two - lists files in /one/two. | Optional | boolean | true |
Small Files
Note
You must restart the unravel_ondemand
daemon for any changes to take effect.
Property/Description | Set by user | Unit | Default |
---|---|---|---|
unravel.python.reporting.small.files.use.avg.file.size.flag
| boolean | true | |
unravel.python.reporting.files.small.files.min.parent.dir.depth Directory depth to end the search at. For instance, if depth=2, a directory having depth more than or equal to 2 can be included in the report. Thus a directory such as | count | 0 | |
unravel.python.reporting.files.small.files.max.parent.dir.depth Directory depth to end search at. The maximum is 50. For instance, if depth=5, a directory having a depth less than or equal to 5 can be included in the report. Thus a directory, for example, | count | 10 | |
unravel.python.reporting.files.small.files.drill.down.subdirs.flag Determines to which ancestor directories a filesize is accounted for. If this is True then the file size is accounted for all ancestor directories. If this is False the file size is accounted for only the parent directory.
For example, given a file under a directory /one/two/three. when when | boolean | true |
File Reports
The following properties are for Huge files report.
Property/Description | Set by user | Unit | Default |
---|---|---|---|
unravel.python.reporting.files.huge.files.drill.down.subdirs.flag Determines to which ancestor directories a filesize is accounted for. If this is True then the file size is accounted for all ancestor directories. If this is False the file size is accounted for only the parent directory. For example, given a file under a directory /one/two/three. when when | boolean | true | |
unravel.python.reporting.files.huge.files.max.parent.dir.depth Maximum depth of a directory to be included in the report. Directory depth to end search at. The maximum is 50. For instance, if depth=5, a directory having a depth less than or equal to 5 can be included in the report. Thus a directory, for example, | integer | 10 | |
unravel.python.reporting.files.huge.files.min.files The minimum number of files which must meet the threshold size for a directory to be included. | integer | 1 | |
unravel.python.reporting.files.huge.files.min.parent.dir.depth Minimum depth of a directory to be included in the report. Directory depth to end the search at. For instance, if depth=2, a directory having depth more than or equal to 2 can be included in the report. Thus a directory such as | integer | 0 | |
unravel.python.reporting.files.huge.files.threshold.size File size must be greater than or equal to the threshold_size for it to be counted. | bytes | 100 * 1024 * 1024 * 1024 | |
unravel.python.reporting.files.huge.files.top.n.dirs Number of directories to be included in the report. | integer | 200 | |
unravel.python.reporting.files.huge.files.use.avg.file.size.flag
| boolean | false |
The following properties are for Medium files report.
Property/Description | Set by user | Unit | Default |
---|---|---|---|
unravel.python.reporting.files.medium.files.drill.down.subdirs.flag Determines to which ancestor directories a filesize is accounted for. If this is True then the file size is accounted for all ancestor directories. If this is False the file size is accounted for only the parent directory.
For example, given a file under a directory /one/two/three. when when | boolean | true | |
unravel.python.reporting.files.medium.files.max.parent.dir.depth Maximum depth of a directory to be included in the report. For instance, if depth=5, a directory having a depth less than or equal to 5 can be included in the report. Thus a directory, for example, | integer | 10 | |
unravel.python.reporting.files.medium.files.max.threshold.size The file size must be less than or equal to the max_threshold_size for it to be counted. | bytes | 10 * 1024 * 1024 * 1024 | |
unravel.python.reporting.files.medium.files.min.files Minimum depth of a directory to be included in the report. | integer | 5 | |
unravel.python.reporting.files.medium.files.min.parent.dir.depth Directory depth to end the search at. For instance, if depth=2, a directory having depth more than or equal to 2 can be included in the report. Thus a directory such as | integer | 0 | |
unravel.python.reporting.files.medium.files.min.threshold.size The file size must be greater than or equal to the min_threshold_size for it to be counted. | integer | 5 * 1024 * 1024 * 1024 | |
unravel.python.reporting.files.medium.files.top.n.dirs Maximum number of directories to display, that is, only display the results for the top N directories. | integer | 200 | |
unravel.python.reporting.files.medium.files.use.avg.file.size.flag
| boolean | false |
The following properties are for Tiny files report.
Property/Description | Set by user | Unit | Default |
---|---|---|---|
unravel.python.reporting.files.tiny.files.drill.down.subdirs.flag Determines to which ancestor directories a filesize is accounted for. If this is True then the file size is accounted for all ancestor directories. If this is False the file size is accounted for only the parent directory.
For example, given a file under a directory /one/two/three. when when | boolean | true | |
unravel.python.reporting.files.tiny.files.max.parent.dir.depth Maximum depth of a directory to be included in the report. For instance, if depth=5, a directory having a depth less than or equal to 5 can be included in the report. Thus a directory, for example, | integer | 10 | |
unravel.python.reporting.files.tiny.files.min.files Minimum depth of a directory to be included in the report. | integer | 10 | |
unravel.python.reporting.files.tiny.files.min.parent.dir.depth Directory depth to end the search at. For instance, if depth=2, a directory having depth more than or equal to 2 can be included in the report. Thus a directory such as | integer | 0 | |
unravel.python.reporting.files.tiny.files.threshold.size The file size must be less than or equal to the max_threshold_size for it to be counted. | bytes | 100 * 1024 | |
unravel.python.reporting.files.tiny.files.top.n.dirs Maximum number of directories to display, in other words, only display the results for the top N directories. | integer | 200 | |
unravel.python.reporting.files.tiny.files.use.avg.file.size.flag
| boolean | false |
The following properties are for Empty files report.
Property/Description | Set by user | Unit | Default |
---|---|---|---|
unravel.python.reporting.files.empty.files.drill.down.subdirs.flag Determines to which ancestor directories a filesize is accounted for. If this is True then the file size is accounted for all ancestor directories. If this is False the file size is accounted for only the parent directory.
For example, given a file under a directory /one/two/three. when when | boolean | true | |
unravel.python.reporting.files.empty.files.max.parent.dir.depth Maximum depth of a directory to be included in the report. For instance, if depth=5, a directory having a depth less than or equal to 5 can be included in the report. Thus a directory, for example, | integer | 10 | |
unravel.python.reporting.files.empty.files.min.files Minimum depth of a directory to be included in the report. | integer | 10 | |
unravel.python.reporting.files.empty.files.min.parent.dir.depth Directory depth to end the search at. For instance, if depth=2, a directory having depth more than or equal to 2 can be included in the report. Thus a directory such as | integer | 0 | |
unravel.python.reporting.files.empty.files.top.n.dirs Maximum number of directories to display, in other words, only display the results for the top N directories. | integer | 200 | |
unravel.python.reporting.files.empty.files.use.avg.file.size.flag
| boolean | false |
Top X
Property/Description | Set by user | Unit | Default |
---|---|---|---|
com.unraveldata.ngui.topx.enabled Controls the generation of Top X reports.
| boolean | true |