Veritas Data Insight Administrator's Guide

Maximum scans to run in parallel on this server

The Collector can perform multiple full scans in parallel. This setting puts a limit on the total number of full scans that can run in parallel on a Collector. The default value is two threads. Configure more threads, if you want scans to finish faster.

The setting is disabled by default.

Maximum shares per filer to scan in parallel

If multiple shares of a filer can be scanned in parallel, this setting puts a limit on the total number of shares of a filer that you can scan in parallel.

Default scan schedule

Specifies how often full scans must be performed. By default, full scans are scheduled to repeat at 19:00 P.M. on the last Friday of each month.

Select the check box to override this setting at a filer or at a share level.

Pause scanner for specific times

You can configure the hours of the day when scanning should not be allowed. This setting ensures that Data Insight does not scan during peak loads on the filer.

The setting is enabled by default. Scans resume from the point they were at before they were paused.

Pause scanner schedule

Configures when scanning should not be allowed to run. By default, scanning is paused from 7:00 A.M. to 7.00 P.M., Monday to Friday.

You can specify multiple scanner pause schedules for different days of the week. For example, you can choose to pause scanning from 7:00 A.M. to 7:00 P.M. on weekdays and from 7:00 A.M. to 9:00 P.M. on Saturdays and Sundays.

To add a scanning schedule:

You can also edit or delete existing scanning schedules.

Maximum scans to run in parallel on this server

The Collector can perform multiple full scans in parallel. This setting puts a limit on the total number of full scans that can run in parallel on a Collector. The default value is two threads. Configure more threads, if you want scans to finish faster.

The setting is disabled by default.

Maximum shares per filer to scan in parallel

If multiple shares of a filer can be scanned in parallel, this setting puts a limit on total number of shares of a filer that can be scanned in parallel.

The default value is 2.

Default scan schedule

Specifies how often incremental scans must be performed. By default, incremental scans are scheduled to run at 7:00 P.M. each night.

Schedule incremental scans more or less frequently based on how up-to-date you need information in Data Insight to be.

Pause scanner for specific times

You can configure hours of the day when scanning should not be allowed. This setting ensures that Data Insight does not scan during peak loads on the filer.

This setting is enabled by default. Scans resume from the point they were at before they were paused.

Override default incremental scan schedule (Default: 19:00 on last Friday of each month)

Check the box if you want to override the default incremental scan schedule which runs at 19.00 hrs on last Friday of each month.

Pause scanner schedule

Configures when scanning should not be allowed to run. By default, scanning is paused from 7:00 A.M. to 7:00 P.M. Monday to Friday.

Use global settings

This option is selected by default. When this option is selected, Data Insight uses the throttling thresholds defined on the Settings > Global Settings > Scanning and Event Monitoring page.

See Configuring scanning and event monitoring .

Use custom settings

Select Use custom settings to disable scanning or override the global throttling thresholds and define custom values.

Throttle scanning based on latency of the filer

Clear the check box to disable scanning for the filers that the collector is monitoring.

Select to enable throttling of Data Insight scans for NetApp 7-mode and Cluster-Mode file servers. This option is not selected by default.

Data Insight collects latency information from NetApp file servers. It can use this information to throttle scanning, if latency of the file server increases above a certain level. This ensures scanner does not put additional load on the file server during peak load conditions.

You can configure the following parameters to enable throttling for NetApp file servers:

Scanner snapshot interval

Scanning a big share can take several hours. The scanner periodically saves information to a disk so that information is visible sooner without waiting for the entire scan to finish.

You can configure how often information is saved to the disk by the scanner. By default, the scanner creates a snapshot of new information every 300 seconds (5 minutes). The minimum value you can set for this parameter is 300.

Total indexer threads

The indexer processes incoming scan and access event information for various shares and updates the per-share database. This setting configures how many databases can be updated in parallel. By default 2 threads are configured.

Specify a larger value for bigger setups where indexer is not able to keep up with incoming rate of information. This is indicated when you observe too many files in the inbox of the Indexer worker node. However, you must ensure that the Indexer has adequate CPU and memory when configuring a higher number of indexer threads. You need approximately 1 GB of RAM per indexer thread.

Limit maximum events processed in memory

By default, the indexer processes all new incoming events in memory before saving the events to the disk. If your are falling short of RAM on your Indexer, you can limit the maximum number of events that the indexer processes in memory before it saves them to the disk.

Reconfirm deleted paths when reconciling full scan information

After Data Insight indexes full scan data, it computes the paths that no longer seem to be present on the file system. Similarly, if Data Insight discovers folders or subsites missing in a SharePoint site, then it computes those subsites or folders as deleted without performing a reconfirmation scan.

Select this check box to have Data Insight re-confirm if those paths are indeed deleted using an incremental scan before removing them from the index.

When the check box is clear, Data Insight readily removes the missing paths from the indexes without carrying out a re-confirmation.

Indexer schedule

Specify how often an index should be updated with new information. By default, all new data is consumed once every four hours.

Indexer gets better throughput if more information is given to it when indexing. However, if you configure a very high value, new information will not be visible in the Console for a much longer period.

Indexer integrity checking schedule

Data Insight checks the integrity of its databases once a week. If any errors are found in the database, an event is published. You can configure a different schedule if required.

Audit events preprocessor schedule

Incoming raw audit events from file servers must be pre-processed before sending them to the Indexer. At this stage, collector.exe applies various heuristics to the raw events and also removes transient events.

By default, raw events are processed every 2 hours.

Batch size (MB)

The maximum size of the raw audit event files that a single Collector thread can process.

The default batch size is 2 GB.

Total Collector threads

The Collector can run multiple pre-processors in parallel. This setting configures how many instances can run in parallel.

Use global settings

This option is selected by default. Use the FPolicy safeguard settings for NetApp 7-mode and Cluster-mode filers as defined on the Settings > Global Settings > Scanning and Event Monitoring page.

See Configuring scanning and event monitoring .

Data Insight collects latency information from NetApp file servers. It can use this information to initiate safeguard mode, if latency of the file server increases above or falls below a certain level. When the safeguard is in effect, Data Insight drops its FPolicy connection to the filer. This ensures event collection does not put additional load on the file server in peak load conditions.

If the latency on the physical file server increases above the configured threshold, Data Insight disconnects from the associated virtual file server. This information is also displayed on the Data Insight System Overview dashboard.

Use custom settings

Select Use custom settings to disable the safeguard settings or to override the global safeguard thresholds and define custom values.

Enable FPolicy safeguard settings

Select one of the following, as appropriate:

These FPolicy safeguard settings are not selected by default. When these check boxes are cleared, the safeguard settings are not in effect.

Configure the following values:

Ping timeout (in minutes)

If a worker node does not heartbeat in the specified interval, Management server will publish an event to that effect. This setting is only applicable for the Management Server.

Notify when CPU continuously over (percentage)

If CPU used on this server is consistently over the specified percentage, an event is published. (Default value: 90%)

Notify when memory continuously over (percentage)

If Memory used on this server is consistently over the specified percentage, an event is published. (Default value: 80%)

Notify when disk usage over (percentage)

If disk usage, either for the system drive or data drive, is over the specified threshold, an event is published. (Default value: 80%)

Notify when disk free size under (MB)

If the free disk space for the system drive or data drive is over the specified threshold in megabytes, an event is published. (Default value: 500 MB)

Notify when number of files in err folders over

If Data Insight is not able to process an incoming file for some reason, that file is moved to an err folder. Data Insight publishes an event if number of files in the err folder crosses the specified threshold. (Default value: 50)

Notify when number of files in inbox and outbox folder over

If Data Insight is not able to process incoming data fast enough, the number of files in the transient folders, inbox and outbox, goes on building up. Data Insight publishes an event if number of files crosses the configured threshold. (Default value: 5000)

Maximum content scans to run in parallel on this server

The Classification Server can perform multiple content scans in parallel. This setting puts a limit on the total number of scans that can run in parallel on a Classification Server. The default value is two threads. Configure more threads, if you want scans to finish faster.

Ensure that the server has enough resources to run the configured parallel threads.

Maximum classification threads to run in parallel on this server

When content is being classified in parallel, this setting puts a limit on the total number of classification threads that you can run in parallel.

Maximum shares per filer to content scan in parallel

When multiple shares of a filer are scanned in parallel, this setting puts a limit on the total number of shares of a filer that you can scan in parallel.

Pause content scanner for specific times

You can configure the hours of the day when content scanning should not be allowed. This setting ensures that Data Insight does not scan during peak loads on the filer. The setting is enabled by default. Scans resume from the point they were at before they were paused.

Content scanning is paused by default from Monday to Friday from 7:00 A.M. to 7:00 P.M.

To configure the pause schedule, click Edit on the scheduler to configure the days and time on which content scanning should be paused.

You can configure more than one pause schedules. Click Add to add new pause schedule.

Custom Property

Veritas Classification Server master ID

This property represents the master Classification Server ID assigned to slave Classification Server.

Veritas Classification lb. disabled

This property is applicable to Veritas Classification master server only if at least one server is in the pool used to turn load balancing feature on/off. By default, it is set to false. If you want to disable this feature set its value to true.

Matrix.classify.fetch.max_batches

This is the global property used to set maximum number of batches per priority to keep in content fetch queue on every slave. This is used only if load balancing is enabled. The default value is 10.

Maximum memory when generating report output

Specifies the maximum memory that can be used for generating a report output. By default, it is 1024 MB on a 32 bit machine and 2048 MB on a 64 bit machine

Total threads for generating report output

Configure the number of threads for generating report output (PDF/HTML/CSV) in parallel. Default value is 2.

This setting applies to the Management Server.

Number of threads for a single report run

Configure the number of threads responsible for generating the report output database for a single report. This configuration applies to the Indexer node.

This setting helps you speed up the process of report generation.

For a particular Data Insight server, the thread count applies to all types of reports.

Maximum reports that can run simultaneously

By default, Veritas Data Insight Software Compatibility List executes two reports in parallel. However, you can configure a higher value to run multiple reports in parallel.

Maximum kernel ring buffer size

The Windows File Server filter driver puts events in an in-memory buffer before the DataInsightWinnas service, consumes them. By default, it uses a 100MB buffer. You can use a bigger buffer. Data Insight publishes an event that indicates events are being dropped due to a high incoming rate.

Ignore accesses made by Local System account

The Windows File Server filter driver ignores accesses made by processes running with Local System account. This setting ensures that Data Insight can ignore most events originating from the operating system processes or other services like antivirus and backup.

Clear this check box to enable monitoring accesses made by LOCAL SYSTEM account. This is not recommended on a production file server.

Flush events on VxFS filer before audit

Set this option to true, if you want to force VxFS to flush its events to disk each time requests for information. This option is useful in Proof-of-Concept (POC) setups and enables you to see events faster in Data Insight

Maximum number of audit threads

This option determines how many filers to fetch audit information from in parallel.

Maximum kernel ring buffer size (Number of records)

The access event records are saved in a log file on the VxFS filer before Data Insight consumes them. By default, 50,000 records can be saved in the log file. You can also specify a larger number. Data Insight publishes an event that indicates that events are being dropped due to a high incoming rate.

Set default credentials for NFS scanner

Set this option to true if you want to allow Data Insight to use the specified User and Group ID to log in to scan NFS shares.

User ID

The ID of the NFS user that the Data Insight uses to scan the filer.

You can set the value to 0 to allow root access from the Data Insight scan hosts.

Group ID

The ID of the group that the Data Insight uses to scan the filer.

You can set the value to 0 to allow root access from the Data Insight scan hosts.

Preserve intermediate files

As new data comes into a Data Insight system, it moves between various modules. In this process the original files are deleted and a new processed file is generated for the next stage of processing.

To aid troubleshooting, select this check box to retain the intermediate data files. These files get stored in attic folder in the data directory.

Preserve raw audit event files

Events processed by the Audit Pre-processor stage are deleted once consumed. If this setting is enabled, raw audit event files will be preserved in the attic folder in the data directory.