Relativity Processing

Console Guide
May 2, 2025 | Version 24.0.375.2

For the most recent version of this document, visit our documentation website.
Table of Contents
1 Relativity Processing Console 5
1.1 RPC features 5
1.2 Running an RPC job 5
1.3 Special considerations 7
1.4 Logging in to the RPC 8
2 Installing the RPC 11
2.1 Licensing 11
2.2 Distribution 11
2.3 Pre-installation requirements 11
2.4 Installing the RPC 13
2.5 Validating the RPC installation 15
2.6 Repairing or uninstalling the RPC 15
2.7 Upgrading the RPC 16
3 Importing data 17
3.1 Creating a data store 17
3.1.1 Working with Relativity-generated data stores 18
3.2 Importing data 18
3.2.1 Import jobs 18
3.3 Troubleshooting import errors 23
3.4 Generating job error reports 23
4 Extracting text 25
4.1 Extracting text 25
4.2 Inspecting the extracted text 27
4.3 Manually re-running single documents with errors 30
4.4 Using filters to resolve errors 30
5 Indexing data 34
5.1 Running an indexing job 34
5.2 Language detection 35
5.3 Merging subindexes 36
6 Generating images 38

Relativity Processing Console Guide 2

 6.1 Generating images 38
7 Filtering data 48
7.1 Creating a filter 48
7.2 Displaying filter results 50
7.3 Filter types 51
7.3.1 Cross reference import filter 51
7.3.2 Date/Time filter 53
7.3.3 Deduplication filter 55
7.3.4 Expression filter 57
7.3.5 Full text search filter 60
7.3.6 Retrying imaging or text extraction errors through a filter 62
8 Exporting data 66
8.1 Creating an export job 66
8.2 Updating task parameters 73
8.2.1 Using the field editor 76
8.2.2 Editing dates 78
8.2.3 Editing multi-value fields 79
8.2.4 Endorsing PDFs 79
8.2.5 Switch statements 82
8.2.6 Setting metadata content and folder path 86
8.2.7 Defining a copy condition 87
9 Managing workers and jobs 88
9.1 Starting worker machines 88
9.1.1 Stopping versus taking worker machines offline 89
9.2 Viewing worker activities 90
9.3 Viewing worker status 91
9.4 Naming a worker 92
9.4.1 Viewing basic worker properties 94
9.5 Other control options 96
9.6 Running a job 97
10 Performing Quality Control tasks 99
10.1 Performing automatic QC 100

Relativity Processing Console Guide 3

 10.2 Non-image QC 105
11 Running standard reports 108
11.0.1 Exporting a report 111
12 Maintenance tasks 112
13 RPC user interface 113
13.1 Data Stores window 113
13.1.1 Right-click Data Store options 114
13.1.2 Hiding a data store 116
13.1.3 Deleting a data store 116
13.2 Filters window 116
13.3 Reports window 117
13.4 Worker Activity window 119
13.4.1 Worker Activity menu options 119
13.5 Matter Inspector window 120
13.5.1 Matter Browser menu options 120
13.5.2 Right-click Matter Browser options 122
13.6 Image Viewer window 123
13.6.1 Image viewer menu options 124
13.7 Extracted Text Viewer window 125
13.7.1 Extracted Text Viewer menu options 126
13.8 Page Text Viewer window 126
13.9 Job Activity window 127
13.9.1 Job Activity menu options 127
13.10 Configuring job settings 128
13.11 Properties window 131
14 RPC known issues 133
15 Field mappings 137

Relativity Processing Console Guide 4

1 Relativity Processing Console
The Relativity Processing Console is a desktop application that lets you control complex processing jobs on
a granular level. The RPC offers power processing users to extend the full capabilities of Processing.

1.1 RPC features

The RPC includes, but is not limited to the following features:

n Granular worker and job control – prioritize worker servers into work groups, which you can alloc-
ate to work on specific jobs or processing functions.
n Selective import of data – choose specific folders and files to import for a given custodian, import or
exclude specific files by extension, De-NIST documents.
n Extract text - able to extract text directly from documents, images, or files with embedded images
OCR’d automatically.
n Indexing options - uses the dtSearch as its full-text search engine, automatically indexes all
metadata along with extracted text to ensure nothing is overlooked.
n Filter options - multiple filtering options, including date/time, de-dupe with extremely flexible options,
cross reference, expression, and full text.
n Reporting – create several reports, including de-dupe filter, error, summary, search filter.
n Generate images – render documents as PDFs without the need for print drivers, significantly
improving performance. Imaging jobs are also multi-threaded, and you can monitor worker activities.
n QC Functionality – preview all images and extracted text - or a random sample - track progress,
auto-advance images, and flag them for further review or replacement.
n Full metadata access – all extracted metadata fields are identified and stored in individual fields for
potential use.
n Robust export options:
o Numerous filtering options, including date/time, cross reference, expression, and full text.
o Customize and brand/endorse the images for export.
o Select metadata fields.
o Format fields – for example, specify date or date/time for date fields.
o Variable substitution – for example, specifying the return of a variable in a specific field.
o Build switch statements – for example, similar to a v-lookup, "when x is found return y".
o Track status.

1.2 Running an RPC job

There are many ways to run a job using the RPC. Below is a typical workflow. Note that you're free to
deviate from it as necessary to meet your own needs.

Relativity Processing Console Guide 5

Depending on your needs, you can re-order or skip some of these steps completely. For example, if the
filtering you require doesn't include keyword searching, you have the option of skipping indexing. You can
also skip extracting text, indexing, and filtering and jump directly to imaging, and you'll still be able to export
text. Image generation also creates page-level text by default, which you can export either on a page or
document level, as needed.
It's important, however, to know the prerequisite steps for the tasks you intend to perform. The following
table breaks down those prerequisites.

Desired RPC task Prerequisite

Extracting text Imported data

n Can be done immediately after

import
Generating images Imported data

n Can be done immediately after

import
Indexing data Either text extraction or image generation
Full-text search filtering An indexed import

Relativity Processing Console Guide 6

 Desired RPC task Prerequisite
All other filtering Imported data

n Can be done immediately after

import
n No text extraction or imaging is
required
Exporting metadata Imported data

n Can be done immediately after

import
Exporting images or page level text Image generation
Exporting doc level text Either text extraction or image generation
Exporting native files Imported data

n Can be done immediately after

import
Exporting PDF files Image generation

1.3 Special considerations

You can use the RPC in conjunction with Processing, or as a standalone tool. For example, you might use
the RPC to monitor and troubleshoot jobs that you kick off from Processing, or run disparate jobs solely in
the RPC.
However, consider the following when using the RPC in conjunction or disparately from Processing.

n The Inventory feature only exists in Processing.

n The ingestion filtering feature only exists in Processing because Inventory only exists in Processing.
However, it is still possible to De-NIST an import or include/exclude specific file extensions from an
import job using the RPC.
n You can't direct export to Relativity for review at this time.
n Reports are different than Processing reports.
n You can't cancel a job in the RPC.
n You can use the RPC to track a processing job in Processing, however, the RPC doesn't capture the
following in Relativity:
o Re-leveling
o Retrying and/or clearing errors
o Altering document extracted text
o Changing the parent/child document relationship

Relativity Processing Console Guide 7

 n Changing the following RPC settings on a processing job may cause unexpected behavior in Pro-
cessing as there may be conflicting settings on the same data sets:
o Adding store-level settings
o Adding project-level settings

1.4 Logging in to the RPC

To provide visibility into who has accessed the RPC and for how long, Relativity requires you to provide user
authentication to log in to the RPC. Any active Relativity user can log in to the RPC with their Relativity
credentials.
This functionality was introduced in Relativity 9.4.398.62.
To log in, click on your instance of the RPC to bring up the login screen, enter your Relativity username, and
click Continue.

Enter your password and click Login. If you use RSA authentication to log in to the RPC, make sure that the
server on which the RPC is installed is configured for RSA, or you won't be able to log in.

Relativity Processing Console Guide 8

 Note: If you don't see the login screen, or your login attempt(s) fail, go to the machine on which the RPC
is installed, get the identity server URL, open a browser on the machine where the RPC is installed, enter
the identity server URL, and attempt to log in again. The identity server URL is case sensitive.

You now have access to all RPC functionality.

To log out of the RPC, navigate to the File tab and click Logout & Exit.

Relativity Processing Console Guide 9

Note: Relativity records RPC logins and logouts in the user's audit history; however, it doesn't designate
those actions as being RPC-specific. In this case, you could see duplicate records for "Login" and
"Logout" in your audit history, one for Relativity and another for the RPC.

Relativity Processing Console Guide 10

2 Installing the RPC
Before you begin installing the Relativity Processing Console (RPC), you must update your environment
with the required security and other configuration settings. See the System requirements guide.

2.1 Licensing
You must have a Processing license and be running Server 2024 or above in order to use the Relativity
Processing Console. See the Licensing guide.

2.2 Distribution
For RPC distribution, contact Customer Support.

2.3 Pre-installation requirements

The RPC is an application that you can install to your desktop on a local machine or to the same server on
which the Queue Manager is installed. The machine must be on the same domain as the processing worker
and file servers, and the SQL Server.
Before running the RPC installer, verify that the items in the following checklists have been completed. This
ensures that you won't encounter any disruptions during the RPC installation process:

Web UI Done
You’re running Relativity 8.2 or above.
Processing is installed to at least one workspace.
You have a valid Processing license.
The RPC version matches the Invariant version.
At least one Worker is running and is designated for processing work.
The version of Processing installed via the Servers tab is correct.
The version of the RPC Installation package corresponds to the Relativity version.
IE Enhanced Security Configuration is off.
Authentication process is now mandate client registration.
Use of a local admin user account, and running RPC with Run as administrator.

Processing SQL Server Done

In SQL Server Management Studio, the DataFiles location is set in the
Invariant.dbo.AppSettings table. This should be a network share accessible to the Relativity
Service Account:
SELECT Value1 from Invariant.dbo.AppSettings
WHERE Category = 'DataFiles'

Relativity Processing Console Guide 11

 Processing SQL Server Done
In SQL Server Management Studio, the dtSearchPath location is set in the
Invariant.dbo.AppSettings table. This should be a network share accessible to the Relativity
Service Account:
SELECT Value1 from Invariant.dbo.AppSettings
WHERE Category = 'dtSearchPath'
In SQL Server Management Studio, the IdentityServerURL is set correctly in the
Invariant.dbo.AppSettings table. This value is case sensitive. If it isn't entered in the proper
case, you can run the following statement to capitalize "relativity":
--begin tran
--update Invariant.dbo.AppSettings
--set Value1 = 'https://<WebServerName>/Relativity/Identity'
--where Category = 'IdentityServerURL'
--commit
The user (or group) account running the RPC from the desktop has access to SQL Server
Management Studio via Windows Authentication and has to have the following server roles
assigned:

n bulkadmin
n dbcreator
n public
All pre-existing and newly created processing server databases (Invariant and Store databases)
have RPC users mapped to the SQL Server logins with the db_owner permission set.

Local machine Done

The account you're logged in to has a Windows Authentication login on the Processing SQL
Server.
The account running the RPC is a local administrator on the worker servers. This is required to
stop a worker and take a worker offline.
The following four SQL CLR type packages are installed locally, as well as the Microsoft Report
Viewer 2012. These must be installed before you install the Microsoft Report Viewer 2012:

n CLR Types for SQL 2012 (x86 32 bit)

n CLR Types for SQL 2012 (64-bit)
The Microsoft Report Viewer 2012 is installed locally:

n Microsoft Report Viewer 2012

Before running the RPC installer, you are required to first whitelist and then register any processing-related
machines for the Secret Store so that those machines can read and write to and from the store. This
includes the Invariant queue manager, all workers, and all machines on which the RPC is installed. See The
Relativity Secret Store Guide for more information on configuring servers for the Secret Store.
Once you've verified or completed all of the above items, you can move on to running the RPC installer.

Relativity Processing Console Guide 12

2.4 Installing the RPC
The RPC is available in a 64-bit version and is only available on a local install, not on a network install.
To install the RPC:

1. Log in to the machine with the local admin account where you want to install the RPC.
2. Open the installer executable to launch the setup wizard.
3. Click Next on the Welcome screen.

Relativity Processing Console Guide 13

4. Enter the installation directory for the RPC. This path must be 200 characters or fewer.

5. Click Next and enter the following database information:

n Invariant instance database server name with the optional port number
n Relativity instance database server name with the optional port number
n Sql username must be EDDSDBO
n Sql password

Relativity Processing Console Guide 14

 6. Click Next and then click Install.

2.5 Validating the RPC installation

Before you can use the RPC for the first time, you must successfully submit one processing or inventory job
from Processing.

2.6 Repairing or uninstalling the RPC

Use the installation wizard to repair or uninstall the RPC and the Queue Manager Server.

1. Log in to the machine with the local admin account where the RPC and the Queue Manager are
installed.
2. Open the installer executable to launch the setup wizard.
3. Click Next on the Maintenance screen.

Relativity Processing Console Guide 15

 4. Click Repair or Remove to modify your installation.

n Repair - click this button to repair your RPC installation. The installer adds a fresh copy of any
deleted RPC files, but it doesn't make any configuration changes or replace other files that
already exist. On the Ready to Repair the RPC window, click Repair.
n Remove - click this button to uninstall all RPC components from your machine. On the Ready
to Remove the RPC window, click Remove.

2.7 Upgrading the RPC

To upgrade the RPC, run the latest version's installer. You don't need to uninstall your current RPC version.

Relativity Processing Console Guide 16

3 Importing data
You can use the RPC to import data into a store. The RPC extensively uses the job queue to distribute the
import workload across multiple machines in order to maximize processing resource efficiency.

3.1 Creating a data store

Before you can import data, you must create a data store in the RPC, which adds a new database to the
SQL Server. Data stores allow you to organize your data by a store, group, project, and custodian. This
organization allows you to optimally use the RPC when you perform text extraction, generate images, filter,
and export.

Note: You can't create a data store enabled for DataGrid.

Any databases that you create in Relativity also appear in the RPC. See Working with Relativity-generated
data stores on the next page.

Note: The RPC is designed to work in distributed environment across multiple SQL, file, and worker
servers. You may see multiple SQL Servers listed in the Data Store window.

To create a data store:

1. Open the RPC.

2. In the Data Stores window, right-click on an SQL Server.

3. Choose one of the following menu options:

n New Store - creates a new database on the server.

n New Group - creates a group that you can use to organize data stores. For example, you can add all
data stores for the same customer to a group.

Note: You can delete groups but not data stores. You can hide a data store by right-clicking the data store
and select Hidden.

Once you've created a data store, you can import data. See Importing data on the next page.

Relativity Processing Console Guide 17

3.1.1 Working with Relativity-generated data stores
You can use the RPC export feature to generate a load file from a Relativity-generated data store. For
example, you may want to create a load file for Relativity import. See Exporting data on page 66.

3.2 Importing data

Once you create a data store, you can add documents to the RPC by importing them to a data store. The
RPC extensively uses the job queue to distribute the import workload across multiple machines in order to
maximize processing resource efficiency.
The RPC generally uses the following workflow to import data (such as a PST file) from a hard drive:

1. Identifies the file type, and calculates the hash for the file so that it can be copied to the network.
These hashes are later used for deduplication filters. See Filter types on page 51.
2. Calls the specific handler for the file type, and passes in the required parameters. Each file type has
its own handler (or plugin) that requires its own set of parameters.
3. Adds job to the queue so worker can execute the method for opening the file (such as a PST), count
the items in it, and add 100 item groups to the job queue.
4. Extracts first group of, for example, 100 items from a container file and adds them to the queue as a
job. Continues with the second group of 100 items, and so on (Multiple workers may be running these
jobs.)
n You may see something different than a 100 items in your queue for a container file. This is
because for large container files, the RPC breaks the job into a number of smaller sub jobs,
and this number is not fixed. It varies, as it depends on the original size of the container file.
5. Adds jobs for attachments to the queue, jobs for metadata extraction, and so on. This process con-
tinues until there are no remaining sub tasks to add to the queue.
The RPC extracts only metadata from the files during an import job. See Extracting text on page 25 for more
information.

3.2.1 Import jobs

Create an import to organize your data in a single database (data store). For example, you create a project
and designate different custodians under that job. Use jobs to determine your export sequence.

Note: Prioritizing custodians for de duplication is not supported in the RPC. Custodian order is
determined during ingestion. These settings can be changed at any time including during the import
process.

To create an import job:

1. In the Data Stores window, right-click a data store.

2. Navigate to Tasks > Import.

Relativity Processing Console Guide 18

3. From the Create Data Import Job dialog, enter or select the information for the following fields:

n Data Store - the selected data store displays by default. Select another data store from the
drop-down list.
n (Optional) Project and Custodian - enter names for these fields. These names display in the
Data Store window.
o Project - the project that you want to specify for the data import job.
o Custodian - the custodian that you want specify in the project.

Note: You can import data without defining a project or custodian. You can change
these settings at any time, including during the import process.

n Workstation - select the machine with the data that you want to process. When you select a
workstation, the List box displays all the drives and CD-ROMS available on that computer, so
that you can make them visible to other worker machines. This option is useful if the data
resides on a CD-ROM, USB drive, or other physical media that is associated with a specific
worker machine.

Relativity Processing Console Guide 19

4. Use one of the following methods to select the folders or files that you want to import:
n Expand and drill-down in the Server List box to locate the required data. Select the check
boxes for the folders or files.
n Select the check box for Click to add path. Click this box to clear it, and paste or type the path
to the import folder. You can use this option if the data is already on the network.
n If you select a checkbox for a folder, the RPC imports all of its content even though you won't
see the individual check boxes for the documents selected when you expand the list.
n You have the option of adding a SQL setting if you have a fixed import location that you would
like to select in the folder window. To do this, go to the Invariant database, add an entry to the
AppSettings table with a category of MapVolume, and enter the UNC path from which you'd
like to import in the Value2 column. Keep the other columns as NULL. You can app multiple
MapVolume entries to this table.

Note: It is possible to add a SQL setting if you have a fixed import location that you would
like to see as a location to import data from. In the Invariant database add an entry to the
AppSettings table with a category of ‘MapVolume’ and enter the UNC path you would like to
import from in the Value2 column. The other columns should stay as NULL. Multiple
MapVolume entries can be added to the table.

5. Choose the option Include selected path in virtual path if you want the virtual path to start at the
server name rather than lower in the path hierarchy. If you don't select this option, the path starts with
the folder name.
6. Click the Additional Settings button.

Relativity Processing Console Guide 20

7. In the Job Settings dialog, select the Import tab.

8. Select any of the options listed in the following table.

Options Select this option if you want to...

File Handling
De-NIST Remove known system files from a set of data. The RPC uses the database provided
by the National Institute of Standards (NIST), which includes approximately 14 million
unique SHA-1 hashes for known system files, such as kernel32.dll, user32.dll, and
others. These files can automatically be eliminated from a forensic collection because
they are not custodian created files. (There is discoverable information in files on the
NIST list, but none of it is pertinent to the case/matter unless you're using NIST.)
Do not Excludes various image file types found inside Microsoft office files. For example,
import .JPG, .BMP, or.PNG in a Word file.
embedded
images in
Office docs
Do not Excludes various files found inside Microsoft Office files. For example, an Excel
import spreadsheet in a Word file.
embedded
objects in
Office docs

Relativity Processing Console Guide 21

 Options Select this option if you want to...
Do not Performs a one for one import of the files selected for import extracting no children or
import chil- embedded items.
dren
Do not Excludes various files found inside emails. For example, .JPG, .BMP, or .PNG in an
import inline email file.
images in e-
mails
Filter by Exclude or include file types added to this text box during import. You can't select
extensions both in the same import. The RPC will filter on detected file extension, rather than the
box original extension.
To include a file type, enter a plus sign, file extension, and a semicolon. Replace the
plus with a minus sign to exclude a file type as illustrated below:

n Include: +mpp;+msp
n Exclude: -mpp; -msp

Note: You can choose to either include or exclude items on an import. You cannot
have both in the same import.

Only filter Applies filter only to top-level (or parent) and loose documents. Since this filters only
parent top-level documents, all the associated attachments will be returned.
documents
E-Mail
Enable Debug or troubleshoot a job that crashed, or other minor performance issues. This
logging log file is created on the C: drive of worker machine processing the file, and log file
names use the storage ID of the worker. The file contains subject names, entry IDs,
and other information.
Ignore Ignores errors generated when a cyclic redundancy check is performed on PST files.
PST/OST
CRC errors

Note: Time zone is applicable to OCR/imaging, text extraction, exporting, and publishing, but not
importing. By default, the RPC stores date type metadata in the database in Coordinated Universal
Time (UTC). While you can optionally set the time zone on the Job Settings General tab when
importing, this information is required prior to running text extraction, imaging or exporting.

9. Click OK to redisplay the Create Data Import Job dialog.

10. Click Start. The import job moves into the queue, which displays in the Job Activity window.
11. In the Job Activity window, highlight the import job, and select a Group for it. After a worker starts the
job, you can monitor its progress in this window. See Managing workers and jobs on page 88.
Once you've imported data to a data store, you can extract text. See Extracting text on page 25.

Relativity Processing Console Guide 22

3.3 Troubleshooting import errors
The RPC imports all documents into the system, even those with unsupported file types. This
comprehensive import allows you to run exception reports that list any files that the RPC couldn't process.
While the RPC can identify most file types, it doesn't have corresponding handlers for all types. The RPC
imports unsupported file types but they aren't added to the job queue, and undergo no further processing.
You can generate a detailed error report that lists these files as well as any password-protected files.
The RPC also collects the following import information found in various reports:

n De-Nisted Files - when the RPC discovers a file, it queries the NIST database for a match. If the file
is found, the RPC doesn't import the file, but adds an entry for it to the De-NIST table associated with
the job. This table can later be referenced to obtain a list of de-NISTed files.
n Bad container files - these files include containers without children, as well as password protected,
incomplete, or corrupt containers that documents could not be extracted from. Common container file
types include ZIP, RAR, NSF, and PST. You can use this information to troubleshoot these container
files.
n File Details - the RPC captures detailed information about jobs in standard summary reports. See
Running standard reports on page 108.

3.4 Generating job error reports

1. In the Data Stores window, drill down to a completed import job. An import job is directly under a cus-
todian.

Relativity Processing Console Guide 23

 2. Right-click the import job > Discovery.

3. Click a report type. The RPC generates the report and displays it in a new tab.
The following reports are available for troubleshooting an import job:

n Error Report - lists each error that occurred with detailed exception information. You can also run
this report to troubleshoot extracted text errors and imaging errors.
n Summary - lists the frequency of each file type in the job, the total number of discovered documents
and counts of de-NISTed files. It also includes file sizes of all files, de-NISTed files and the size of
files remaining after de-NISTIng not counting containers. Additionally, there is a brief entry describing
each error encountered during import.
n Bad Containers - lists all container files that Invariant was unable to pull any documents from. If a
container throws an error on import but Invariant was able to get a single file from it, the container will
not appear on this report however it will appear on the error report.

Note: Each error listed in the summary report has a corresponding detailed version in the error report. If
you reprocess error files and they don't encounter new errors, they will still appear on the report when you
run that report again. If the reprocessed files encounter errors on the retry, these new errors are listed
along with the original errors. The bad container report is dynamic, which means if a container was
successfully re-imported, it doesn't appear on the bad container report when you run that report again.

See Running standard reports on page 108 for information about other reports.

Relativity Processing Console Guide 24

4 Extracting text
Once you've imported data to a data store, you can extract electronic text from the documents to build
indexes used for keyword searches. Depending on the document type, Invariant extracts the text directly, or
it may physically layout the document before it can extract the text.
Invariant can directly extract text from file formats that it can open, such as Word, Excel, PowerPoint, and
others. For example, Invariant instantiates Word, opens the document, and requests the electronic text from
Word. Any comments or markup added with the Track Changes feature are also extracted as electronic
text. In addition, Invariant extracts any embedded graphics that the document contains during the import
process. Invariant treats the graphics as children of the document, in which they separately undergo OCR.
When you run a keyword search on the entire document family, any text matches that Invariant pulls from
the images are indirectly linked to the document, since the images are children of the document.
For other file types, Invariant must physically layout the document by converting it to PDF. Invariant has a
PDF engine and other functionality to layout the document in memory as text only. It then extracts the
electronic text from each page of the PDF, as well as detects any graphics. Invariant handles the graphics
by sending them to the OCR engine for electronic text extraction. Some files requiring this type of
processing include JPEG, TIFF, PDF, Visio, and JungUm Global (GUL).

4.1 Extracting text

To extract text from an import job:

1. In the Data Stores window, drill down to a completed import job. An import job is directly under a cus-
todian.
2. Right-click the import job >Extract Text.

Relativity Processing Console Guide 25

3. From the Data Extract Job dialog box, on General tab, select options for the following fields:

n Time Zone - set time zone required by the data set. This information is especially important for
processing emails, as date and time are listed in messages and appointments. Obtain the cor-
rect time zone from the client.

Note: You should specify the time zone setting prior to running text extraction. If you change
the time zone after running text extraction, then you should run text extraction again. Also,
make sure to change the following settings prior to re-running: on the OCR/Image tab in the
General Options section, check Overwrite intermediate files and un-check Preserve
existing pages.

n Workgroup - select a group. If you want to start the job manually, leave this field set to Group
0.

Relativity Processing Console Guide 26

 n Filter to apply - use created filters to control the data that you want included in or excluded
from the extraction job. Leave this field blank if you want to extract all the data. See Filtering
data on page 48.
n Add - If you're using a filter, click Add. Select an option to include or exclude the data returned
by this filter in the text extraction job.

Note: We recommend not using filters on first-time extracted text. Instead, use filters to retry
errors on subsequent text extraction jobs. See Using filters to resolve errors on page 30. If
you intend to create a full text search filter on your data, the text will need to be indexed prior
to filter creation. Any items filtered out of text extraction will throw an error during the
indexing stage.

4. (Optional) Select the OCR/Image tab if you want to set options for documents that must undergo
OCR before the text can be extracted. See OCR/Image tab settings in Generating images on
page 38.
5. Click Start. The job is moved into the queue, which displays in the Job Activity window.
6. Monitor the progress of the job in the Job Activity window. See Managing workers and jobs on
page 88.

Note: If your workgroup is set to Group 0, highlight the job and select another group to begin processing.
If you have workers assigned to Group 0, this is not necessary, and the first available worker will begin the
job.

Once the text extraction job completes, you can index the data. See Indexing data on page 34.

4.2 Inspecting the extracted text

After the job completes, verify that the text is successfully extracted from the files.
To inspect the extracted data:

Relativity Processing Console Guide 27

1. In the Data Stores window, drill down to the import job that you used for the extract text job.
2. Right-click on the import job, and click Inspect.

3. From the Matter Inspector dialog box, click Refresh Needed to display the documents in the text
extraction job. If you did not apply a filter, all the documents display.

4. In the Matter Inspector window, scroll to the right to display Message, Extracted Text File Name, and
Intermediate file columns. You can sort each column in ascending or descending order by clicking on
the column header. Review the following columns and their content for troubleshooting information:

Relativity Processing Console Guide 28

 n Message - lists any errors that occurred during import, text extraction or image generation.
You may want to re-run text extraction on documents with errors, depending on the type of
error. See Manually re-running single documents with errors on the next page.
n Extracted Text File Name - lists the path where the extracted text is stored. If this column is
empty, the file may be an image type file, in which case the Intermediate File column will be
populated or the file may be of an unprocessable type (audio/video files, system files, etc). If
the document fits neither of these categories, you should re-run text extraction on the file. See
Manually re-running single documents with errors on the next page.
n Intermediate File - lists the path to the intermediate file (PDF) created when text can’t be dir-
ectly extracted from the native file. This intermediate file undergoes OCR to provide text for the
document. If there is no extracted text, this column is empty, and the file is not unprocessable,
you should re-run text extraction on the file. See Manually re-running single documents with
errors on the next page.
To examine the extracted text for a document, click Extracted Text Viewer. If the document only has an
intermediate file, click Page Text Viewer.

Relativity Processing Console Guide 29

4.3 Manually re-running single documents with errors
Use these steps to manually resolve an error in a single document:

1. Right-click an error in the Message column, and select Extract Text. A new extraction job creates.

2. In the Job Activity window, highlight the job and select a Group to begin processing. See Managing
workers and jobs on page 88.
3. After the job processes, repeat the steps in Inspecting the extracted text on page 27.

4.4 Using filters to resolve errors

You can resolve multiple errors simultaneously by using filters. You can filter by file ID and rerun only those
files which do not have extracted data. For more information, see Filtering data on page 48.

1. Highlight a data store that you want to filter for errors.

2. Navigate to the Filters tab in the Data Store window (You can also right-click on a data store, and
select Filters from the menu.)

Relativity Processing Console Guide 30

3. Click New Filters and select Expression.
4. Enter a descriptive name in the New Filter box and press TAB. (If you don't press TAB after entering
a filter name, the filter name might not get updated in the database.)

Relativity Processing Console Guide 31

5. Select the checkbox next to the Custodian that you want to run the filter on.
6. In the Expression box, enter the following text: ((matter.Message != "") && (matter.Unprocessable ==
false)).

Note: This expression returns any records where the Message column is not blank and the file is
not flagged as unprocessable. This means that any files from which the text was extracted
successfully are not returned. You must capitalize Message and Unprocessable, and use two sets
of double quotation marks in the expression.

Relativity Processing Console Guide 32

 7. Click Execute at bottom of Filter tab. When the filter is finished click Done in the pop-up dialog box.
8. In the Data Stores window, drill down to the import job that you used for the extract text job.
9. Right-click the import job > Extract Text.
10. On the General tab, select your filter in the Filter to apply box filter.
11. Click Add and select As an include filter. The RPC will only rerun documents that have an error
message and are not flagged as unprocessable.

12. In the Job Activity window, highlight the job and select a Group to begin processing. See Managing
workers and jobs on page 88.
13. After the job processes, repeat the steps in Inspecting the extracted text on page 27. All messages
should be cleared.

Relativity Processing Console Guide 33

5 Indexing data
Before you can run any full text search filters on the imported documents, you must first build an index from
your import job’s extracted text. You can then use operators and other features provided by a dtSearch
index to construct full text search filters.
The RPC breaks indexing jobs into 1000 document chunks, which are then run in parallel using multiple
workers and threads to improve performance. The RPC merges the dtSearch indexes after these jobs
complete.
During indexing, errors may occur from documents that don’t have any extractable text. The RPC continues
building the index when an error occurs, but those documents without extractable text aren’t represented in
the index. Using SQL Reporting Services, the RPC generates an error report that lists the documents and
error details. For example, a document without extracted text or intermediate file is listed in this report.

Note: You can index data at any point after text extraction or image generation. If you choose to skip text
extraction and proceed directly to image generation the indexing process will combine the page level text
into a doc level text file that will be used in the index.

5.1 Running an indexing job

Before running an indexing job, you must first run text extraction or generate images on the job. See
Extracting text on page 25 or Generating images.

1. From the Data Stores window, drill down to the import job that you used for the extract text job.
2. Right-click the import job > Indexing > Build Search Index.

3. In the Job Activity window, highlight the indexing job and select a Group to begin processing. See
Managing workers and jobs on page 88.
4. After the job processes, right-click the indexing job in the Data Store window > Discovery > Error
Report.
5. Review the report to determine if the index needs to be rebuilt.

Relativity Processing Console Guide 34

 Note: You can also use the Matter Inspector tab to investigate an indexing error. To do this, obtain the
storage ID from the error report. In the Data Store window, highlight the job, and select Inspect to open
the Matter Inspector. Use the storage ID to locate the document. Determine if the document is an image,
needs OCR, is password-protected, or has another issue.

5.2 Language detection

The RPC supports language detection by adding metadata to each document in an import job. Using the
metadata, the RPC creates a language map of the document. It then maps out a range in the document for
each language that it contains, and provides a score that identifies the dominant language.
Before you can perform language detections, you must first index the import job.

1. In the Data Stores window, drill down to an import job.

2. Right-click on the import job. Point to Indexing, and click Detect Languages.

3. Select the View menu, and click Properties window.

4. Right-click on the import job, and click Inspect to open the Matter Inspector.
5. To view the language map, highlight the document in the Matter Inspector, and click Properties.
6. In the Properties window, expand Storage Metadata, and then expand Language Map to display a
list of language ranges.

Relativity Processing Console Guide 35

 7. The range for the language lists the starting character position, the character length, the language,
and the percentage of the text in that language. For entry [3], the starting character position is 89,
character length is 93, the language is zh (that is Chinese), and the text is 100% Chinese: [3]
89,93,zh,100. When the Language Dominance is below 50, this value usually indicates unreliable
detection.

Note: For a complete list the two-digit codes that accompany each full language name, visit the ISO
standards site.

5.3 Merging subindexes

The RPC performs parallel indexing by distributing jobs across multiple machines, which create
subindexes. It then attempts to merge all the subindexes after the indexing completes. When a very large
import job indexes, this process can over run dtSearch capabilities, which has a limit of about 4 billion
words.
Errors can occur when the RPC automatically attempts to merge multiple subindexes into a single index. In
this case, you can manually attempt to merge a few subindexes at a time, gradually building a large index.

Note: The RPC can search across multiple indexes. It automatically detects them and spans the search
across them, but this process is slower than using a single index.

1. In the Data Stores window, drill down to an import job.

Note: The option Verify and Merge Subindexes verifies that all expected documents are
indexed, and then attempts to merge the subindexes.

Relativity Processing Console Guide 36

2. Right-click on the import job. Point to Indexing, and click Merge Subindexes

.
3. Enter a value and click OK.

Relativity Processing Console Guide 37

6 Generating images
The RPC references the options set on the Document and OCR/Image tab to determine how the images are
generated, and OCR runs. During OCR, the RPC converts the native files to intermediate PDFs, and the
text is then extracted from these files. For intermediates derived from natives with text, the text layer of the
created PDF is extracted. For intermediates derived from images the text is created by the OCR engine.
Natives with both text and images will have both of these methods performed on the intermediate file.

Note: You can index data at any point after text extraction, and you can image documents immediately
after import. You can also index after image generation. The system automatically combines page-level
extracted text into a document level extracted text and index that text.

The intermediate files are stored in the Intermediate Folder created for a job. This folder may contain
intermediate files using the following naming conventions:

n storageID_PH.PDF: Indicates a placeholder file.

n storageID_R.PDF: Indicates a file containing redactions.
n storageID.PDF: Indicates a PDF of a page.
When exporting, the RPC looks in the Intermediate Folder first for placeholders, files with redactions, and
then PDFs of pages. It uses the appropriate PDF to generate TIFs for export as determined by settings
selected on OCR/Image tab for a specific job.

Note: To locate an Intermediate folder for a job, right-click a job in the Data Stores window > Inspect.
Right-click a file in the Matter Inspector > Open > Explore Intermediate Folder. Use the same
procedure to select Explore Native Folder.

6.1 Generating images

1. In the Data Stores window, drill down to a completed import job. An import job is directly under a cus-
todian.
2. Right-click the import job, and select Generate Images.

Relativity Processing Console Guide 38

3. From the Generate Images Job window on the Documents tab, select options for the following fields.

Note: On the Document tab, the checkboxes have on, off, and indeterminate states. When the
checkbox is in an indeterminate state, the document’s internal setting for an option is used by
default. When the option is turned on or off in the RPC, this new value overrides the default setting
in the document.

Relativity Processing Console Guide 39

The following table lists the options available on the Document tab.

Relativity Processing Console Guide 40

 Option Description
Excel - Workbook Processing Options
Force Use this option to generate a placeholder for each Excel file instead of an image
placeholder
Ignore print Use this option to override any pre-defined print areas existing in an Excel file.
areas When this option is selected, the entire document is imaged.
Limit to X Use this option to image only the number of pages entered in this box.
pages box
Show change Use this option to display any modifications made to the document through the
tracking Track Changes option in Excel. By default, this option is selected.
Excel - Worksheet Layout
Paper size Choose an available paper size. The default value is the size internally stored in the
file.
Print Order Use this option to determine the order used to render the worksheets. (Print down
and then over, or vice versa.)
Page Fit - Reset the zoom level for a spreadsheet. The value in this field will override the
Zoom Level default setting internally stored in the file.
Fit to X pages Reset the page fit for a spreadsheet. The values in these fields will override the
wide by X tall default settings internally stored in the file.
Excel - Manipulation Options
Row and Select this option to render row and column headings (that is the number assigned
Column to rows, and the letters assigned to columns).
Headings
Headers and Select this option to render headers and footers in a spreadsheet.
footers
Note: Leave this option disabled. The printing of these items interferes with the
ability of the RPC to detect blank pages.

Gridlines Select this option to render the gridlines between columns and rows in a
spreadsheet.
Auto-fit rows Use these options to expand the dimensions of rows and columns to accommodate
and Auto-fit their content. When these options are selected, any hidden columns or rows are
columns displayed. They are selected by default.

Note: You can prevent the RPC from displaying hidden columns and rows by
setting the DontUnhide option on the Advanced tab to True.

Remove all Select this option to remove background color. It ensures that any hidden text or
background rows formatted to match the background color are displayed.
fill colors
Set text color Select this option to display the font color of text as black. It ensures that any hidden
to black text is displayed, such as text with a white font on a white background.
Clear empty Use these options to remove empty rows and columns from a spreadsheet, and
rows and render as few pages as possible. By default, these options are selected.

Relativity Processing Console Guide 41

 Option Description
Clear empty
columns
Use printer Use a printer driver to render the PDF rather than internal processing provided by
the RPC. By default, this option is not selected.

Note: Select this option to speed up processing when you are imaging
spreadsheets created in Excel 2007, especially if they contain charts or graphs.

Show cell This determines whether Relativity renders comments.

comments
Hide and Set a maximum value for consecutive blank rows or columns. By default, the RPC
page break checks for a run of 10 blank rows or columns, and compresses them to limit the
after X number of pages rendered for a spreadsheet.
consecutive
blank rows
box
E-Mail
Resize tables Modifies the table sizes to make them compatible with the page size. By default, this
to fit page option is selected.
Clear Removes the tabs from e-mail threads to improve their display. (Select this option to
indentations prevent the entries in the thread from aligning too far to the right.)
Resize Modifies the image sizes to make them compatible with the page size. By default,
images to fit this option is selected.
page
Render Changes the orientation of the message to landscape.
landscape
Do not Select this option to prevent the downloading of images from the internet.
download
images
Show Adds text in the header identifying the image as a message, appointment,
message type distribution list, and so on. By default, this option is selected.
in header
Display Select this option to render SMTP addresses for To, Cc, and Bcc fields instead of
SMTP Microsoft Exchange information.
address
Word
Show change Select this option to render any markup added to the document through the Track
tracking Changes feature in Word.
Show Select this option to render any comments added to the document through the
comments Review feature in Word.
Show field Select this option to render the field code rather than the value in the document.
codes

Relativity Processing Console Guide 42

 Option Description

Note: In general, this option is not enabled, because the RPC can preserve the
actual value for a field code by preventing modifications to it when the document
is opened. For example, field codes for file path and date will not be automatically
updated with the current user settings, but will retain the original settings.

Show hidden Select this option to render any text added to the document with Hidden feature in
text Word.
PowerPoint
Show Renders the slide at the top of the page, and speaker's notes at the bottom. By
speaker default, this option is selected, and the image orientation is portrait even when the
notes document does not contain speaker notes.
When this checkbox is not selected, the image orientation for the PowerPoint slides
is landscape.
HTML
Render with Select this option to render HTML in Word. By default, the RPC will render HTML
Word documents in Internet Explorer, and then generate a PDF. For some problematic
HTML documents, a better image will generated if the file is rendered in Word.
Remove nbsp Select this option to remove long rows of non-breaking spaces (nbsp) codes, which
codes prevent the text from wrapping properly. When it renders HTML as a PDF, the RPC
will automatically format page breaks without cutting text or margins.

Relativity Processing Console Guide 43

4. On the OCR tab, select options for the following fields.

Option Description
General Options
Overwrite Use this option to generate a new intermediate file. The RPC does not store
intermediate multiple copies of an intermediate file for a document. When an intermediate file
files already exists, the RPC will use it to generate images unless this option is selected.
By default, this option is not selected.
Preserve Use this option to prevent existing PDFs from being overwritten. The RPC will skip
existing existing pages when they do not need to be regenerated. By default, this option is
pages selected.
Discard Use this option to ignore redacted pages during processing, and select intermediate
redacted PDFs (named as storageID.PDF). By default, this option is not selected.
pages
You can use this option when redacted PDFs (storageID_R.PDF) have been added
to the Intermediate Folder for a job. Instead of preferring these files, the RPC will
process unredacted files (storageID.PDF).
Image Generation
Format Use this option to change the image format. For most processing, use Default,
which is CCITT v4 for generating black and white TIFs. This setting is the setting at
installation but it can be modified at the project or data store level as well.
Select another format if you want to force a specific format for images. Options
include Color JPG, and for TIFs, CCITT v3 and CCITT v4.

Relativity Processing Console Guide 44

 Option Description
Alt format See Format options above.
Alt format file Use this option to force a file type to be imaged in an alternate format.
extensions
For example, you could set the Alt Format as Color JPG, and then enter +XLSin
the file extensions box. The RPC will then render Excel files as Color JPGs. You
can also exclude file types from rendering as an Alt Format by prefixing the file
extension with a minus sign. This is a regular expression filter, which means that it
will return partial matches to the extension you enter. For example, if you enter
+XLS here, the RPC will also image Excel files with .xlsx and .xlt extensions.
Dithering Select one of the following algorithms for converting images from color to black and
Algorithm white:

n Default (Uses Clustered 6x6)

n Floyd & Steinberg
n Dispersed 4x4
n Dispersed 8x8
n Clustered 6x6
n Clustered 8x8
n Clustered 16x16
n Dispersed 16x16
n No Dithering (Uses setting for Threshold Weight option listed below.)
Threshold Enter a value for this option when you set the Dithering Algorithm to No Dithering.
Weight The default threshold value is 128, while the threshold range is between 0 - 255.
When the pixel RBG in an image is greater than the threshold, it is converted to
black, and when less, it is converted to white.
DPI Use this option to control the density of image (dots per inch). The default value is
300. A typical range used for setting the DPI is 200 - 600.
Custom Use this option to enter a path to a PDF file that you want used as a custom
placeholder placeholder.
PDF
n The file chosen will be used to create placeholders when imaging. It does not
set what placeholder to use when exporting. The distinction is that the
exporter looks for the imaged placeholder, not what it’s in this field. If you
have done imaging with the default placeholder and later enter a value for a
custom placeholder you will need to re-image the placeholder documents oth-
erwise the originally imaged default placeholders will be exported.
n Use UNC pathing to the custom placeholder and be sure to save the place-
holder in a location the Invariant workers can access. Also, make sure the
name of your custom placeholder ends with ‘_PH’ (Sample_PH.pdf).
n You're unable to enter a dynamic file path in the middle of a placeholder
name that changes for each file.

Relativity Processing Console Guide 45

 Option Description

n The RPC doesn’t support adding slipsheets in front of each exported file with
dynamic metadata fields on them. However, it is possible to use blank place-
holders and apply an endorsement on them of that file’s metadata through
the use of switch statements. For more information, see Using a switch state-
ment for custom logic on page 84.
Max pages Use this option to specify the maximum number of pages imaged for a document.
per doc (Set this value if you do not want the entire document imaged.)
Generate Select this option to not create and page level text when imaging.
PDF only
Searchable Select this option to perform in-place OCR on the PDF page image elements.
PDF
Do not show Select this option to generate blank placeholders. By default, placeholders display
error an error message.
messages on
placeholders
Render color Select this option to substitute JPGs for TIFFs when the pages are in color. (TIFFS
pages to JPG are generated only in black and white.)
OCR
Engine Use this option to choose the Nuance or no OCR engine. Select NoOCR when you
want extracted text but don’t want to OCR pages with images. The default engine is
Nuance.
OCR Type Use this option to control the performance of OCR job. Select Accurate for more
precise OCR, Fast for improved performance, and Balanced to equalize precision
and performance.
Preserve text Use this option to maintain the current layout of the text when extracting text from
layout PDF for OCR. By default, this option is selected.
Layout text in Use this option to maintain the order of text as in the PDF layout when extracting
stream order text for OCR.
Show OCR Use this option to display a separator between extracted text at the top of a page
text separator and text derived from OCR at the bottom of the page. The separator reads as, “---
OCR From Images ---“. With the separator set to off, the OCR will still be on the
page beneath the extracted text but there will be nothing to indicate where one
begins and the other ends. By default, this option is selected.
Allow OCR Clear this option if you do not want documents that do not have extractable
during data electronic text to undergo OCR. By default, this option is selected.
extraction
Exclude line Select this option to remove line art and annotations added to the original page.
art during (For example, it will remove any text boxes drawn over the original content of the
OCR page.)
OCR images Use this option to set the DPI level the Nuance engine will use when performing
DPI box OCR. Changing this can often allow the OCR engine to successfully OCR if it failed
using the default setting.

Relativity Processing Console Guide 46

 Option Description
Languages Use this option to make multiple language selections.

5. Click Start. The job is moved into the queue, which displays in the Job Activity window.
6. Monitor the progress of the job in the Job Activity window. See Managing workers and jobs on
page 88.

Note: If your workgroup is set to Group 0, highlight the job and select another group to begin
processing. If you have workers assigned to Group 0, this is not necessary, and the first available
worker will begin the job.

Once you've generated images, you can QC images. See Performing Quality Control tasks on page 99.

Relativity Processing Console Guide 47

7 Filtering data
After you’ve imported data you can use filters to isolate a set of data. Use filtering to run quality control tasks
to check or fix errors, extract text, as well as produce or export documents. Filters are similar to creating
saved searches in Relativity and are especially useful when you’re performing tasks that require you to
include or exclude a specific set of documents.

Note: For more information on how to execute filters in the RPC, see the Filtering scenarios in the RPC
video webinar on the Relativity Training site.

7.1 Creating a filter

This section includes generic steps for creating a filter. For information about specific filter type options, see
Filter types on page 51.

1. In the Data Stores window, highlight a data store that you want to filter, and click the Filters tab at the
bottom of the store list. If the Filters tab isn't visible here, right-click on the data store name and select
Filter(s)... The Filters tab won't be visible if you previously had it open but then closed out of it.

Relativity Processing Console Guide 48

2. Click New Filter to display a list of filter types.

3. Select a filter type, and enter a name in the New Filter box and press Tab. We recommend clicking
Tab after entering the filter name, as this will ensure that the name is saved. Under some cir-
cumstances, not clicking Tab lists the filter merely by its default name of "New Filter (<Type>)."

4. Enter the required information for the filter type that you selected. You will see specific fields for each
filter type displayed in the center of the tab. See Filter types on page 51.
5. Select one or both of the following options to determine additional actions when the filter flags an
item:
n Flag the parent if it is a child item - if the filter flags a child item, then the parent will also be
flagged.
n Flag all children of flagged parent - if the filter flags the parent, then the children will be
flagged. Select this option if you only want to search for parent documents, but you also want to
include the children of the documents that had a hit.

Relativity Processing Console Guide 49

 o If you simply click Execute without flagging anything, the filter you've set up will only flag
documents that actually meet the search criteria. This could be problematic if you want
to then export the filtered data, especially in situations in which you've hit on a keyword
that’s a child document but you didn’t hit on the parent document. This is because you
can’t export a child document without also exporting its parent. Thus, it's recommended
that you check either the flag the parent if it is a child item or the flag all children of
a flagged parent option. Checking both of these brings in the entire family, which you
can then export.
o Flagging all children of a flagged parent ensures that you capture all children of a parent
that contained a keyword hit.
o If you have a parent item and it has, for example, ten children and one of the children
hits on the keywords you entered and you check the flag for parent and then execute the
filter and then export the data, it will display as one parent and one child.
6. Click one of these buttons:
n Rediscover - queues the files in the filter for re-import. The filter must be executed first for this
button to function.
n Clone Filter - clones the filter.
o You can use the clone option to create a filter using identical date settings or keyword
terms and then adjust the job list it's filtering. Alternatively, you can keep the jobs being
filtered on and just change, for example, the end date on a date filter.
o The newly created filter clone won't have any items in the filter until you click Execute.
To clone a filter, click Clone Filter, enter a name for the new filter, press the Tab key,
modify the filter according to your needs, and click Execute.
o You can't modify filters used in an export that uses a setting of New, Replacement or
Supplemental. If you need to alter a locked filter, you must first clone the filter and then
edit the clone.
n Execute - runs the filter and populates the data in the RPC. Until you click Execute in the Fil-
ters window, you haven't yet set the filters you've created or added.

Note: When you see a list of import jobs in the filter window after selecting a filter type, those jobs are
sorted by Job ID by default, but you can sort on any column (Custodian, Project, Job ID, CreateOn, etc.)
you want to.

Once you've filtered data, you can create export files for review or production. See Exporting data on
page 66.

7.2 Displaying filter results

To display the filter results, complete the following steps:

Relativity Processing Console Guide 50

 1. In the Data Stores window, right-click on an import job, and select Inspect.

2. From the Matter Inspector, click , and select a filter in the Filter to apply box.

3. Click the Add drop-down arrow to select the filter type.

n As an Include filter - includes any document that matches the criteria set in this filter. For
example, you could add a keyword search as an include filter, so that all documents with this
keyword are exported.
n As an Exclude filter - excludes any document that matches the criteria set in this filter.
n As a Privilege filter - flags documents that are potentially privileged based on their inclusion
of privilege-relevant terms found in the full text search.
o In the matter inspector, adding a privilege filter has no impact on the documents dis-
played.
o Privilege filters are limited to exporting; however, a Boolean search is most often used
for a privilege filter.
4. Close the window after you have selected all the required filters.
5. Click Refresh Needed to display a list of documents match your filter settings. For more information
about the Matter Inspector, see Performing Quality Control Tasks.

7.3 Filter types

This section describes how to use the various filter types supported by the RPC.

7.3.1 Cross reference import filter

When you are importing data, this filter uses the file ID as a cross reference. You can have multiple cross
reference import filters for a single data store. You can define five metadata fields in the cross reference text

Relativity Processing Console Guide 51

file used by this filter. See Formatting the cross reference file below.

1. Select Cross Reference Import in the New Filter menu.

2. Enter a name in the New Filter box and press Tab.
3. To configure the filter, browse or enter the path to your file.
4. Select Strip duplicate lines if you want any duplicate file IDs in the cross reference filter removed.

5. To run the filter, continue with Step 5 in Creating a filter on page 48.

7.3.1.1 Formatting the cross reference file

When you import data using the Cross Reference Import filter, you need to supply a text file containing
additional metadata fields. These user-defined fields can be used for external data, such as branding,
tagging, reference, and export information, as well as external control numbers.

Note: The FileID is the only required field. Other fields are optional.

You must follow the required format for the cross reference file. The first row in the file must be a header
row, which is not imported when you run the filter. You can also define up to five metadata fields in the cross
reference text file:

n EXTRA1
n EXTRA2
n USERDEFINED1
n USERDEFINED2
n USERDEFINED3
Use the format in the sample file illustrated below:

Relativity Processing Console Guide 52

 Note: Use two sets of double quotation marks to represent a blank.

7.3.2 Date/Time filter

For this filter, you can set specific date ranges and metadata after ingestion. This filter uses a metadata
search so it does required extracted text for the selected custodian.

1. Select Date/Time Filter in the New Filter menu.

2. Enter a name in the New Filter box and press Tab.

Relativity Processing Console Guide 53

3. Click the Custodian checkbox for one or more jobs that you want to filter for date and time.

4. Select a Date Match Option:

n Any - filters on any date available for a document. Each document is associated with multiple
dates, such as document sent date, delivered date, date last printed, and so on. When you

Relativity Processing Console Guide 54

 select this option, the document will be returned if any of the dates match.
n Earliest - filters on the earliest date in the metadata list. Usually filters on the create date.
n Latest - filters on the latest date in the metadata list. Usually filters on the last modified date.
5. To specify a data range, select Before, Between, or After. Use the date drop-down option to select a
date. (The beginning date starts at midnight, while the end date terminates at 11:59.)
6. To control the metadata used by the filter, select Restrict to specific metadata, and then choose
the metadata fields. These fields are based on the metadata discovered in the document collection
during import. They are pulled dynamically from the database.

Note: When running a date filter, make sure to restrict that filter to specific metadata or else it will
return hits on any date that meets the parameters you entered. This may include parameters that
you don't care about when executing a date filter, such as Last Printed On. In this way, not
restricting to specific metadata often results in an over-inclusive document list and costs you time
during your QC process.

7. To run the filter, click Execute.

7.3.3 Deduplication filter

Use the Deduplication filter to remove multiple copies of a document from a data set by selecting two groups
of custodians for this process. When you run a deduplication filter for the first time, the RPC generates the
hash tables used for it, and adds them to the database. The performance time is improved when you re-run
the filter because these tables are already built.
You can use this filter in the following ways:

n Dedupe the custodian against self - you can de-dupe the data set of a custodian against itself. For
example, you could de-dupe the laptop against the desktop of a custodian.
n Dedupe one custodian against another - you can select one custodian whose document set may
contain duplicates, and then another custodian whose data set determines the documents to be
removed. For example, you may want to dedupe the messages in John’s mailbox against those in
Jane’s. See Targeted deduplication filter scenario on page 57.
n Dedupe against multiple custodians - you can dedupe one custodian against multiple custodians.
The RPC uses a weighting algorithm to determine which documents to dedupe, when the same cus-
todian is in both groups.

Note: The Deduplication filter does not delete any documents. It simply does not return them in the list of
documents.

1. Select Deduplication in the New Filter menu.

2. Enter a name in the New Filter box and press Tab.
3. Click the Custodian checkbox for one or more jobs in the top window that you want de-duplicated
against the jobs selected in the bottom window.

Relativity Processing Console Guide 55

4. Ensure that the Use content hash option under the Loose Files section is checked. This option
should always be enabled. Note that this option will be removed from the GUI in an upcoming
release.

5. Select filtering options for E-mails. You can select a combination of different hashes that the RPC will
use to identify duplicate files.
n Processing generates four different hashes for emails and keeps each hash value separate,
which allows users to de-duplicate in processing based on individual hashes and not an all-
inclusive hash string. For example, if you’re using processing, you have the ability to de-duplic-
ate one custodian’s files against those of another custodian based only on the body hash and
not the attachment or recipient hashes.
o Body hash - takes the text of the body of the e-mail and generates a hash.
o Header hash - takes the message time, subject, author’s name and e-mail, and gen-
erates a hash.
o Recipient hash - takes the recipient's name and emails and generates a hash.
o Attachment hash - takes each SHA256 hash of each attachment and hashes the
SHA256 hashes together.

Relativity Processing Console Guide 56

 6. Select Filtered Dedupe if you want to de-duplicate the data returned from another filter. Next, select
the filter in the activated list box. You will be performing the deduplication on a smaller data set, which
speeds up the processing. For example, you may want to use a keyword filter on a document.
7. To run the filter, click Execute.

7.3.3.1 Targeted deduplication filter scenario

Let's say that you want to dedupe John's documents against Jane's documents. Before deduplicating, you
need to ensure that any documents in John's data set containing the term "Fruit Loops" are relevant and
therefore should be included in the results. Likewise, any documents in Jane's data set containing the term
"Cheerios" should also be included in the results.
To perform this targeted deduplication filter:

1. Create Full text search filters on:

a. John's data set for the word "Fruit Loops". See Full text search filter on page 60.
b. Jane's data set for the word "Cheerios". See Full text search filter on page 60.
2. Create a new Deduplication filter and ensure the following are set:
n Source filter - Jane's Cheerios filter. Add as an Exclude filter.
n Target filter - John's Fruit Loops filter. Add as an Exclude filter.
3. Execute the filter to dedupe against both custodians. Documents returned will still contain documents
with terms "Fruit Loops" and "Cheerios" specified in John's and Jane's Exclude filters.

7.3.4 Expression filter

With an Expression Filter, you can use C# syntax to write custom code to manipulate the data set. The code
runs on the desktop, and it is dynamically compiled and executed on the database.
This filter type is frequently used for checking errors that occurred during text extraction.

1. Select Expression in the New Filter menu.

2. Enter a name in the New Filter box and press Tab.

Relativity Processing Console Guide 57

 3. Click the Custodian checkbox for one or more jobs that you want to filter with your custom code.

4. Enter a code for a simple filter in the Expression box. If you want to use C# syntax, click Advanced
Expression to auto-generate an outline of a method, and then enter your code.
5. To run the expression on a limited set of data, select a filter in the Sub-Filter box.
6. To run the filter, continue with Step 5 in Creating a filter on page 48.
The following sections provide examples of expression filters. Note that none of these examples requires
advanced mode.

7.3.4.1 Example: Find documents with no error message and aren't flagged as unprocessable
Use the following expression to isolate documents that have an error message but are not flagged as
unprocessable. This filter is useful for situations in which you've run text extraction and want to rerun
anything that threw an error.

public override bool IsMatch(Job job, Matter matter)

{
return (
// begin expression
IsIn((matter.Message != "") && (matter.Unprocessable == false))
// end expression

Relativity Processing Console Guide 58

 );
}

7.3.4.2 Example: Find password-protected documents

Use the following expression to isolate password-protected documents.

public override bool IsMatch(Job job, Matter matter)

{
return (
// begin expression
IsIn(matter.Message.Contains("Password"))||(matter.Message.Contains("password"))
// end expression
);
}

7.3.4.3 Example: Find text within the VirtualPath

Use the following expression to isolate a string of text within the VirtualPath.

public override bool IsMatch(Job job, Matter matter)

{
return (
// begin expression
IsIn(matter.VirtualPath.Contains("Sent Items")
// end expression
);
}

7.3.4.4 Example: Find specific file extensions

Use the following expression to isolate specific file extensions. You must enter all file extensions in CAPS,
not in lower case; if you enter the extensions in lower case, the filter will provide no results.

public override bool IsMatch(Job job, Matter matter)

{
return (
// begin expression
IsIn(matter.FileExtension, "XLS", "XLSX", "XLSM", "XLSB", "XLA", "XLB")
// end expression

Relativity Processing Console Guide 59

 );
}

7.3.4.5 Example: Find documents whose names start with a particular string
Use the following expression to isolate documents with names that start with a particular string.

public override bool IsMatch(Job job, Matter matter)

{
return (
// begin expression
IsIn(matter.FileName.StartsWith("image")
// end expression
);
}

7.3.4.6 Example: Find documents whose names end with a particular string
Use the following expression to isolate documents with names that end with a particular string. This is useful
for finding files with a specific literal extension instead of the identified extension.

public override bool IsMatch(Job job, Matter matter)

{
return (
// begin expression
IsIn(matter.FileName.EndsWith("emf")
// end expression
);
}

7.3.5 Full text search filter

You can use the Full Text Search filter to run keyword searches on data indexes built for specific import
jobs. The RPC uses the dtSearch engine to run these filters. However, it executes the search for each term
in parallel using a separate thread, and then combines the results of these searches.
A common use for a full text search filter is the removal of extraneous documents from a data set. For
example, you could run the filter on a group of documents before importing them into Relativity for review.
You could also use this filter to tag documents that are potentially privileged. See Exporting data on
page 66.

Note: You can run the Comprehensive Hits Report to display a list of statistics about the number of
documents that match a search term and other data. See Running standard reports on page 108.

Relativity Processing Console Guide 60

1. Select Full Text Search in the New Filter menu.
2. Enter a name in the New Filter box and press Tab.
3. Click the Custodian checkbox for one or more jobs that you want included in the full text search.

4. Either click next to the Search For window to browse for a file containing your search terms, or
enter then in the following format: "relativity","processing"

Relativity Processing Console Guide 61

 n When you import a list of terms through the browse option, the RPC automatically sets the
search to use the Boolean option.
n If you manually enter your own terms, the search defaults to Any words, and you must remem-
ber to switch to Boolean if this is the option you want.
n You can also add operators used by the dtSearch engine, such as Boolean operators AND,
OR, NOT.
n You can manually resize the Search For window.
5. Select any of the searching options that you want to use. These options follow the searching con-
ventions used by the dtSearch engine.
6. To run the filter, continue with Step 5 in Creating a filter on page 48.

7.3.5.1 Boolean filter

To create and execute a boolean filter:

1. Click New Filter and select Full Text Filter.

2. Name the filter at the top of the custodian window.
3. Select the data sets to which you want to apply the filter. Note that each data set has its own index
and each index records those noise words that were used when that index was built. Note also that
noise words aren’t configurable in the RPC.
4. Enter the terms you’d like to include in the filter. Each line is its own search, and each line will be
executed in its own thread. These will be joined together by an OR statement, in that the entire search
string is an implicit OR since you've selected the Boolean. The fact that they’re on separate lines
allows you to search for them in parallel across multiple threads.
5. Select Boolean search.
6. Click Execute. Note that the RPC now displays which threads it’s currently searching, as well as the
fact that it's currently populating the search results.

7.3.6 Retrying imaging or text extraction errors through a filter

There are many ways to isolate documents that you may want the system to retry, depending on your
needs. For example, in the matter inspector you may want to sort on the Message column to group all items
with an error message together. You can also de-select the Exclude the following checkbox and check the
Placeholder checkbox and then refresh the screen to display only items that received a placeholder image.
Alternatively, you could sort on the Page Count column to sort anything with a page count of -1 to the top.
To retry a group of errors through a filter in the RPC, perform the following steps:

1. Open the matter inspector for the job that contains the errors.
2. Select the File Ids of all the files that you want to retry.
3. Right-click and select Copy Selected Cells.

Relativity Processing Console Guide 62

4. Use the File IDs that you copied to the clipboard to create a cross reference import filter. See Cross
reference import filter on page 51 for more information.
5. Right-click on the import job and select Generate Images or Extract Text, as appropriate.

6. In the Generate Images or Data Extract Job window, select the filter you just created from the Filter to
apply drop-down list and select As an Include filter from the Add drop-down list.

Relativity Processing Console Guide 63

7. Select the OCR tab, check the box next to the Overwrite intermediate files option, and un-check
the Preserve existing pages box. Then click Start to retry all of the errors you filtered for.

Relativity Processing Console Guide 64

Note: Sometimes documents that error on a data extract job are handled without error on an image
generation job. You can try generating images for data extract errors prior to indexing. When you index
the job, the software uses page-level text derived from the image generation job if the doc level extracted
text doesn't exist. Additionally, during the indexing process the system copies the page-level text files into
a single document-level text file and saves it where the document-level text file would have been created
had the file not thrown an error during data extraction.

Relativity Processing Console Guide 65

8 Exporting data
When you are ready, you can export files for review, production or any other purpose you may have. Use
the steps provided in the following sections to run your exports.

Note: For more information on how to export through the RPC, see the Using the RPC Exporter video
webinar on the Relativity Training site.

8.1 Creating an export job

1. In the Data Stores window either right click on a data store or drill down to a specific import job, right
click on either, and click Export to open the Export Wizard.

2. When the Export Wizard is opened either no import jobs are checked (if opened by clicking on the
data store) or a single import job is checked (if opened by clicking on a specific import. Regardless of
which option was chosen, you can check or un-check import jobs as needed for your export. If you
want to use a saved export file (.EXF) from a previous job, click Load Export to select it before you
pick which jobs to export. There are no adverse effects of selecting the jobs first, but loading an
export un-checks all import jobs automatically and you’ll need to reselect the jobs again.

Relativity Processing Console Guide 66

 n When choosing which import jobs to export you can right click on any of them and make use of
the options displayed in the image above. They are all self-explanatory except for ‘Select All’
which highlights all the import jobs but doesn’t check or un-check any of them. You can also
left-click and drag to highlight a number of jobs then press the space bar to select or deselect
them.
n The default export sequence is by Project. If one or more import jobs have no value entered for
the Project, as soon as an import is checked a pop-up window will inform you of that the export
sequence is automatically being changed to JobId.
3. If you are selecting multiple imports you can choose which export sequence to be used. The
sequence setting dictates the order in which the exporter exports the selected items.
n The Export Sequence is relevant when you export multiple custodians in a single export job.
The sequence setting dictates the order in which the exporter exports those sources.
o Project - exports alphabetically by project name.
o Custodian - exports alphabetically by custodian name.
o JobId - exports by the JobId value in ascending order.

Note: The Comments window serves no purpose at this time.

4. If needed, click Options to display the Export Options dialog box. None of these are required to run
an export

Relativity Processing Console Guide 67

n Ignore Relationships - select this option to treat all documents like parents. For example, you
might want to export only the documents with keyword hits, so you may need to export a child
without a parent. You can use this option to flatten all family relationships.
n Sort By - performs primary sort based on FileId, FileName, ImportSource, ParentFileId,
RootFileId, or Virtual Path.
n Then By - performs secondary sort using selected metadata field.
n Descending - select to perform a descending sort. For example, you may want to use a des-
cending sort on the creation date of e-mail messages, so that you list the messages from new-
est to oldest.
n Do not persist export information - enabling this prevents any export information such as
completion time and assigned Bates numbers from being tracked in an export table in the
Store database. This is useful for cases in which you want to run a test export and you don't
want that job's data permanently tracked.

Relativity Processing Console Guide 68

5. Click OK to redisplay the Choose Sources window of the Export Wizard.
6. Click Next to display the Select Filters window.

7. (Optional) Select a filter from the Filter to apply drop-down list.

8. Click Add, and select one of following options to control the filter behavior.

Option Description
As an Includes any document that matches the criteria set in this filter. For example, you
Include filter could add a keyword search as an include filter, so that all documents with this
keyword are exported.

Note: Select the checkbox under Use Metadata to export keyword, and its
frequency in a document. This functionality is available because metadata is
actually stored in filters. New columns with the metadata are added.

As an Excludes any document that matches the criteria set in this filter.
Exclude filter
As a Dedupe Same as an Exclude filter, except that the RPC uses a different logging code for
filter deduplication. You can run a report on this code to identify files excluded due to
(Exclude) deduplication.
As a References the criteria in the filter and may perform an action based on a match. For
Reference example, tagging information obtained from Relativity could be used to apply tags to
Only documents that match the metadata in the filter criteria. All documents are exported
and tagged as required.
As a Similar to a Reference Only filter. The Potentially Privileged column is updated to Y
Privilege for any document that matches the filter criteria. All documents are exported.
Filter
(Reference)

Relativity Processing Console Guide 69

 Option Description
As a Inserts a placeholder for any document that meets the criteria set in this filter. For
Placeholder example, you could use this filter to substitute a placeholder for all Excel/Power
filter Point or large text files in a production. When you select this filter, the RPC runs the
export job and substitutes the placeholder for these files without modifying the PDFs
already generated. This eliminates the need for you to image these files, since you
can leave them as-is.

n When multiple include/exclude filters are used, the logic used to combine them is an AND state-
ment only. Reference, Privilege and Placeholder filters do nothing to limit any documents from
being exported.
n If any filters were added incorrectly, highlight them and click Remove. The Mask non-relevant
filters checkbox will hide any filters that do not apply to the import jobs selected for export. This
button also can be used to refresh the filters in the drop-down list if you created new ones while
the export wizard was open.

9. Select Next to display the Export Type window.

10. Select an Export Type:

n New - this is typically used for a first-time export. The tracking status for the export is updated
and tracked in the database. The RPC tracks bates numbers, page counts, and other inform-
ation so that you know where to begin numbering for your next export with the same custodian.

Relativity Processing Console Guide 70

 This information is displayed in the Previous Exports and Included Jobs panes of the export
wizard on future exports.
n Supplemental - is intended for partial exports for the same data set in sequence. In other
words, use this option if you want to add documents to an export job that already exists. Note
that you still need to specify a starting number for the supplemental documents, as the RPC
won't automatically number them.
n Replacement - replace an entire export job with a completely different export job. When you
select this option, you must specify a valid export ID value in the Replace Export ID field to the
right, which is only enabled when you select Replacement as the export type. Simply highlight
the job you want to replace in the Previous Exports pane to populate the field.

Note: You can't modify filters used in an export that uses a setting of New, Replacement or
Supplemental. If you need to alter a locked filter, you must first clone the filter and then edit
the clone.

n OtherCustodians - export the DeDuped Custodian and Deduped Path information via Relativ-
ity (the front end). You can't export either of these fields if you don't choose this option. As a res-
ult, standard practice is to export all other metadata using one of the other dropdown settings
and to export these two fields to a separate overlay metadata file using this option.
n Untracked - no tracking information for the export is added to the database. You may want to
select this option if you are doing some experimental export jobs that you do not want tracked
in the database. If you are building a new export file definition, you may run several test exports
to see how the data is displayed with the current settings.
11. Click Next to display Select Tasks window.
12. Click New. Point to Page Tasks or Document Tasks, and click on one or more export tasks. Each of
these options will have a variety of task parameters that can be set. For details, see Updating task
parameters on page 73.
n Page Tasks - available options are Copy Images and Copy Text.
o Copy Images - writes TIF and/or JPG image files to a desired location. An image gen-
eration job must have been performed first. The exporter will throw an error for each doc-
ument exported that was not previously imaged.
o Copy Text - writes page level text files to a desired location. An image generation job
must have been performed first. The exporter will throw an error for each document
exported that was not previously imaged.
n Document Tasks - available options are Copy Native files, Copy PDF Files, Write cross-ref-
erence file, Write extracted text, Write metadata, and Write Summation metadata.

Relativity Processing Console Guide 71

 o Copy native files - copies the native file for an exported document to a desired location.
o Copy PDF files - copies a PDF file for an exported document to a desired location. An
image generation job must have been performed first. The exporter will throw an error
for each document exported that was not previously imaged.
o Write cross-reference file - writes a stock-formatted cross-reference file. It is recom-
mended to include this task on ALL exports. It frequently is a very handy file to have.
o Write extracted text - writes document level extracted text to a desired location. A text
extraction or image generation job must have been performed first. If an image gen-
eration job was run but not a text extraction job, the page level text files will be copied
together to form a doc level text file automatically during the export. The exporter will
throw an error for each document exported that was not previously imaged or text extrac-
ted.
o Write metadata - writes custom metadata to a desired location. Alternatively, this task
can be used to write a file that contains no metadata. For example, you could write a
readme.txt file to include reminders about a customer’s preferences for their deliv-
erables.
o Write Summation metadata - currently inoperative.

Note: You can add the same task multiple times to an export job. Depending on the
task, you can assign a different file name or folder location to each copy of the task.
For example, you could export a set of metadata for a client and another for opposing
counsel by adding the Write metadata task twice. You would then modify the name of
the task added under Document Tasks, and the name of file output for each task. If
the export job needs to be repeated, you do not need to redo each of the document or
page tasks. You can clear the checkboxes for the tasks that do not need to be redone,
and then perform the export. For example, you may want to add new fields to the
metadata file, so you can select only Write metadata under Document Tasks.

13. Click Next and then Finish to run the export job. The Finished box contains any errors that occurred
along with the File ID for troubleshooting purposes. The window outlined in red below will contain any

Relativity Processing Console Guide 72

 errors that occurred along with the File ID for troubleshooting purposes.

14. (Optional) Click Save Export to save your current settings as a reusable export file (.EXF). You can
select this load file for use with another export job by clicking Load Export. You can save your export
at any point during the export process, you do not have to actually export anything before saving. If
you are creating a complicated export, you may want to consider saving your export periodically as
you add tasks.

8.2 Updating task parameters

In the export wizard, each document and page task has multiple parameters that you can update to meet
the requirements of a specific export job. Some of these parameters are universal to all export tasks while
others are specific to that task. The table below outlines the various parameters, what tasks utilize them and
if they are universal or specific.

Associated Universal
Parameter Notes
Tasks or Specific
Bates Number All Universal Also serves as control number for document level
exports. Should be left as {BatesBeginDoc},
expand this section to format the bates/control
number.
DocLevel All Universal If True, numbers are incremented at the doc level.
Ignore Number All Universal If True, the value in the StartAt field will be
disregarded and only the prefix and suffix will be

Relativity Processing Console Guide 73

 Associated Universal
Parameter Notes
Tasks or Specific
used.
Prefix All Universal The prefix to prepend each bates/control number.
Reset Condition All Universal When the value entered here changes, the
control/bates number will reset to the StartAt
number. Typically used with {Custodian}
Reset Counters All Universal When set to true, any generic or file counters will
reset when the reset condition is met.
StartAt All Universal The starting number to assign.
Suffix All Universal The suffix to append to each bates/control
number.
CopyMode Copy Images Specific Normal = export images and create image load
files, BuildNoCopy = create image load files but
don't actually export the images. This setting is
rarely changed from Normal.
Custom Foldering Copy Images & Specific When set to False the system will automatically
Copy Text (page the standard 00\00 folder structure after the Image
level) Sub-Folder. When True the system will use the
value in the Image Sub-Folder field only, allowing
the user to enter custom counters if desired.
Distributed Copy Images, Universal True = all copying and branding is performed by
Copy Text, Copy the workers. False = these tasks are performed by
Natives & Copy the machine running the export.
PDFs
Endorsement Copy Images, Semi- Settings for what to brand on images, page level
Copy Text & Universal* text and PDFs.
Copy PDFs
Bottom Left Copy Images, Semi- What to endorse in the bottom left corner.
Copy Text & Universal*
Copy PDFs
Bottom Right Copy Images, Semi- What to endorse in the bottom right corner.
Copy Text & Universal*
Copy PDFs
Gap Copy Images, Semi- Space between endorsement and image in inches.
Copy Text & Universal*
Copy PDFs
Margin Copy Images, Semi- The distance between the edge of the page and
Copy Text & Universal* the endorsement.
Copy PDFs
Top Left Copy Images, Semi- What to endorse in the top left corner.
Copy Text & Universal*
Copy PDFs
Top Right Copy Images, Semi- What to endorse in the top right corner.
Copy Text & Universal*

Relativity Processing Console Guide 74

 Associated Universal
Parameter Notes
Tasks or Specific
Copy PDFs
Image / Text / PDF / Copy Images, Specific The subfolder to write the files to. Will be a
Extracted Sub- Copy Text, Write subfolder of the value entered in the Working
Folder extracted text & Folder field.
Copy PDFs
Jpeg Natives Copy Images Universal True = use the native file if the native is a jpeg.
False = use the imaged version of the native.
Setting is ignored if endorsing is enabled.
LFP Filename Copy Images, Specific LFP file name to use. An OPT is also automatically
Copy PDFs created and will have the same name.
Overwrite Images / All Specific Overwrite existing files in the destination folder.
Files
Production Name Copy Images, Universal The export volume name. This is the default value
Copy PDFs used for the LFP name. This value is used to
populate the volume column in the LFP & OPT.
Working Folder All Univeral The base folder to write all output to. If the export
work is to be distributed to the workers it needs to
be a location they have write access to. UNC
pathing is strongly recommended.
Compress Text Copy Text, Write Specific When set to true, all blank lines and leading or
extracted text trailing spaces are removed from the text.
Default Text Copy Text, Write Specific If a document has no extractable text, this text is
extracted text used instead of nothing.
Multipage Copy Text Specific Specifies whether to combine single page text files
into a doc level text file or not.
Page Separator Copy Text Specific Specifies what separator to use between each
page of text if Multipage is set to True.
Strip Non-ASCII Copy Text, Write Specific Strips non-ASCII characters from the data or text
extracted text, before writing.
Write cross-
reference file,
Write metadata
Text Encoding Copy Text, Write Universal Specifies the type of encoding to use when writing
extracted text, to the file.
Write cross-
reference file,
Write metadata
Wrap Lines Copy Text, Write Specific Word-wrap lines so they do not exceed the
extracted text specified length. Use 0 to disable.
Copy Condition Copy native file, Specific The condition to evaluate to determine whether or
Copy PDFs not to copy the file. 0 or False means do not copy
the file, 1 or True means copy the file.
Default Extension Copy native file, Specific The extension to use if the file doesn't have an

Relativity Processing Console Guide 75

 Associated Universal
Parameter Notes
Tasks or Specific
Copy PDFs extension.
Filename All document Specific The file name to use.
level tasks
MaxExtLength Copy native file, Specific The maximum number of characters in the file
Copy PDFs extension, 0 for no limit.
MaxNameLength Copy native file, Specific The maximum number of characters in the file
Copy PDFs name, 0 for no limit.
Placeholders Only Copy native file, Specific True = File will only be copied if the image for the
Copy PDFs document is a placeholder.
Source File Copy native file, Specific You won't need to change this. Ever. We promise.
Copy PDFs
UnicodeFilenames Copy native file, Specific If set to False, filenames are converted to ANSI
Copy PDFs and unconvertable characters are replaced with an
underscore. If True, the original Unicode file name
is used.
Highlight Write extracted Specific Highlight search hits from the filters applied to the
text export. Requires Text Format to be set to Rich
Text
Text Format Write extracted Specific Specifies to use Text or Rich Text as the format.
text
Metadata Content Write Metadata Specific The metadata to write to the file. Can include any
number of metadata fields in just about any format.
Can also have no metadata if desired.

* Semi-universal items only apply to other tasks of the same type. For example, if you made endorsement
settings to one image export task, those settings would also apply to any other image export task but they
won’t apply to endorsements on a Copy PDF task.

Note: You can rename any of the tasks if desired, just left click once on the name to highlight and a
second time to edit it. For instance, if you are exporting two different metadata files, you may want to
name one ‘Opposing Metadata’ and the other ‘Our Metadata.'

8.2.1 Using the field editor

Any parameters that aren't accessible via a drop-down option, with the exception of Working Folder, have
an ellipsis button providing access to a Field Editor for that setting.

Relativity Processing Console Guide 76

 Note: You aren’t required to use the field editor; you can enter simple data directly in the window.

The field editor is a powerful tool that provides access to all metadata captured so far in the instance, as well
as numerous ways to manipulate that data.

The upper pane displays what will be returned for a given parameter. In the view above you see the default
fields for Metadata Content. The top line represents the header and the second line the fields of metadata.
The Show Fields button opens the lower pane which provides access to the various fields of metadata. The
presets option lets you quickly change between three common delimiter settings but you aren’t limited to
these.
You can enter any values you want in the Quote and Separator fields to the left. To add a field of metadata,
find what metadata you want to include in your export in the lower pane and do one of three things:

n Ctrl-Left Click will add both the field of metadata to the second line as well as adding the field name to
the header.
n Left Click only will only add the metadata without adding anything to the header.
n Shift-Click will add the name of the field but not the actual metadata.

Relativity Processing Console Guide 77

Once a field is added, if it’s an editable field, you can edit it by double clicking on the field in the upper pane.
Fields that are editable include dates and multi-value fields.
There are a multitude of advanced functions available under the Edit dropdown. Some are only available if
you have a field of metadata highlighted. The more common features are detailed below but all of them
provide a brief description if you mouse-over them. If you have questions about any of the ones not listed (or
listed for that matter), contact support.

n Save this field - used to save a customized field of metadata or switch statement for use in other
exports. Saving the field also allows the field to be mapped through the Relativity front end.
n Apply Formatting > Child Values - the exporter will return the corresponding metadata values of a
given document’s children instead of its own. For child documents it will return its own value only.
n Apply Formatting > Parent Values - the exporter will return the corresponding metadata value of a
given document’s parent instead of its own. For parent documents it will return its own value only.
n Remove Formatting - removes any applied formatting.
n File Counter - enter a customized counter that increments with every document. Double clicking on
this allows you to customize the format of the number returned. This is typically used for creating sub-
folder names for native file and extracted text exports.
n Generic Counter - similar to the File Counter but can used to increment on any field of metadata
rather than just with every document.
n Control Flow > Switch Statement - see below for information on creating a Switch Statement.

8.2.2 Editing dates

Use the following settings to edit dates in the field editor:

n Allow Empty Dates - True = Invalid or missing dates will return null (empty). False = Invalid or miss-
ing dates will return 1/1/1900.
n Date Time Format - you have the full Microsoft custom date and time formatting options available.

Relativity Processing Console Guide 78

 n Time Zone Index - 0 uses the time zone entered in the Job Settings. To use a different time zone on
the export, enter the appropriate index number (GMT = 831169467). It may be easier to keep this at 0
and change the Job Settings instead.

8.2.3 Editing multi-value fields

Use the following settings to edit multi-value fields in the field editor:

n Replacement - if the designated separator character exists in the metadata, it will be replaced with
this character.
n Separator - the character to be used between each value being exported.

Note: Unlike the preset field separator values, there is no function to change all multi-value separators. If
you are changing them from the default | character, be sure to change them on all multi-value fields.

8.2.4 Endorsing PDFs

This example illustrates how to endorse a PDF by adding a Bates number to it.

1. Under Document Tasks, highlight Copy PDF files to display a list of parameters that you can set for
this task.

Relativity Processing Console Guide 79

2. Expand Bates Number and Endorsement in the Parameters box.

Note: The Distributed parameter is set to True by default in the previous illustration. This export job
will queue up a task to the workers. The workers can then do the branding, imaging, generating
PDFs and other tasks. You do need to ensure that the destination location is visible to these worker
machines. When this parameter is set to False, the job will run only on the machine that you are
using, but it will be multithreaded. You might disable this parameter if you were exporting to a USB
drive that other machines cannot see, or if you were debugging.

3. Under Bates Number, highlight Prefix. Enter text or a metadata field from the document, job set-
tings, custodian information, or other source. In the following illustration, a variable substitution
(requiring curly braces) for custodian metadata has been added to the Prefix field.

You can also edit this field by clicking the Browse button. In the Field Editor dialog box, click Show
Fields to display a list of available metadata. Expand a metadata group, and double-click on a field to
add it to list box.

Relativity Processing Console Guide 80

 Some fields also include statistics about the document, such as the ClientSubmitTime under Matter
(Metadata). The occurrence value for this field lists how frequently it appears in the collection. You
can use this occurrence information to determine if a metadata field is relevant to the current data set.
4. To apply formatting, select the metadata (such as Custodian), and click Edit to display a list of format-
ting options.

If you want to truncate the string, click String Truncate to display the Field Properties pop-up where
you can define the maximum length.

Change the default value and close the pop-up to display the Field Editor with the updated formatting
information.

Relativity Processing Console Guide 81

 5. Close the Field Properties and Field Editor pop-ups.
6. In the Parameters box, locate Endorsement. Use the Browse button to select the metadata
BatesPageNumber in Bottom Left field.

8.2.5 Switch statements

A switch statement is an IF/THEN/ELSE statement. A simple IF/THEN/ELSE statement might look like the
following:
if (X = 1)
print “Hello”
else
print “Good Bye”
The colored text above and throughout this section corresponds to these parts of the switch statement:

Relativity Processing Console Guide 82

A switch statement can make multiple checks against the source and return different values depending on
the condition met.
The creation of a switch statement involves using two windows, the field properties window above and the
Switch Comparer Collection Editor.

n If the Custodian contains Maude return ML_ and close the switch statement.
n If the Custodian contains The Dude return TD_ and close the switch statement.
n Else return XX_ and close the switch statement.
n You can have as many conditions as you want in your switch statement but remember that they are
checked in order and the switch statement terminates on the first true condition. Because of this you
must make certain that your logic for the conditions accurately covers all possibilities. Let’s say we
have two custodians named John Smith and Sara Smith. Our first condition checks if the custodian

Relativity Processing Console Guide 83

 name contained Smith then returns JS_ and the second condition looks for Sara and returns SS_. In
this case, none of Sara’s documents would get a return value of SS_ because they all met condition 1
first, returned JS_, and the switch statement ended looking at no further logic.
n You can use switch statements inside of switch statements. It is perfectly acceptable to use another
switch statement for the Key value, Return value, Default value or even the Switch value itself – or
any combination of those – or use more switch statements inside those switches. This of course adds
a great deal of complexity but in return the user can perform very in-depth logic calculations on the
data to return exactly what is needed for the situation.
The field properties contains the following settings:

n Name - the system will automatically give a new switch statement a name consisting of a random
string of letters and numbers. You can change this to fit your needs however the name of the switch
must not match the name of an existing field of metadata. A simple way to avoid accidentally doing
this is to precede the name of your switch with an underscore (_).
n Cases - the various checks that will be made against the entry in the switch field.
n Switch - the field of metadata or fixed value that the collection of cases is compared to.
n Default - the value to return if none of the collection of cases is true.
The Switch Comparer Collection Editor contains the following settings:

n Members pane - list of cases that will be checked

n Properties Pane - displays the definition of the highlighted case in the Members pane
n Compare Method - how the Switch value will be compared the Key. The available options are:
o Equal To
o Less Than
o Less or Equal
o Not Equal
o Contains
o Not Contains
o Greater Than
o Greater or Equal
n Key -the value the Switch is compared against.
n Return Value - the value returned if the comparison is true.

8.2.5.1 Using a switch statement for custom logic

This example illustrates how to define a switch statement used to determine the prefix applied to a Bates
number.

1. Under Document Tasks, highlight Copy PDF files to display a list of parameters that you can set for
this task.
2. Expand Bates Number, and highlight Prefix. Click the Browse button to display the Field Editor.

Relativity Processing Console Guide 84

 3. Click Edit to display a list of options.

4. Point to Control-Flow, and click Switch Statement.

5. Enter a name for the switch and a metadata field used in the Switch box. In this example, the name is
Example_Switch, and a variable {Custodian} is used for the switch.
6. Highlight Cases and click the Browse button.
7. In the SwitchComparer Collection Editor, click Add.
8. In Compare Method, select an option. This example uses Contains.
9. Enter a value in the Key and Return Value fields. Click Add. As illustrated below, the first key-value
pair are Doe and DOE_, while the second is Smith and SMITH_. When the custodian contains one of
these names, the associated prefix will be added to the Bates number.
10. Click OK to display the Field Properties dialog box. For this example, the default switch is UNK_.
11. Close the dialog box to display the Field Editor with the new switch.

12. Highlight the switch, and click Edit. Select Save this Field from the menu.
13. In the Save Field As dialog box, enter a name for the switch. This field is now added to the Recently_
Used list under Show Fields, and be reused as necessary.

14. Close the dialog box.

Relativity Processing Console Guide 85

8.2.6 Setting metadata content and folder path
This example illustrates how you can set the metadata fields written to a file, and how to set a Working
Folder path.

1. Under Document Tasks, highlight Write metadata to display a list of parameters that you can set for
this task.

2. Highlight Working Folder. Enter a folder path or click the Browse button to select one.
Consider the following when defining a Working Folder path:

n This is a shared field among the Document and Page tasks. For example, when you use this
field for one task, it populates for all tasks with the Working Folder field specified.
n You can't specify a separate folder for separate tasks.
n You can also use variables in the folder path.
n This example uses metadata as a variable, so that each custodian has an individual folder:
c:\Exports\{Custodian}
3. Highlight Metadata Content, and click the Browse button to select the metadata fields to include in
the file as described in the previous section.
4. Click Presets to select a delimiter for use in the file.

For example, select Concordance to separate the metadata fields as illustrated below:

Relativity Processing Console Guide 86

 5. Close the dialog boxes.

8.2.7 Defining a copy condition

You can set up a copy condition that permits the file to be copied only when the given expression returns
true. In this example, a copy condition is defined for native files.

1. Under Document Tasks, highlight Copy native files to display a list of parameters that you can set
for this task.

2. Click Copy Condition to display the Field Editor.

3. Enter your own custom fields or click Show Fields to select them.
4. Close the dialog boxes.

Relativity Processing Console Guide 87

9 Managing workers and jobs
The RPC uses a queue manager to control the activities of worker machines. All jobs executed by the RPC
are designed to be distributed across these machines to improve performance. To centrally manage these
workers, the RPC includes the Worker Activity window, which displays the workers and their current
statuses. This central management console also provides you with the ability to manage the activities of
these machines.
Read more about worker machines in the RPC
In the RPC, you can coordinate the activities of worker machines and the processing of jobs by assigning
them both to workgroups. A worker machine processes only the jobs that are assigned to its workgroup.
You can assign multiple workers to a workgroup, and they will all work on the same job. The RPC breaks
ups large jobs into smaller subtasks to leverage parallelism and multithreading. For example, PST and NSF
files are divided into jobs of a 100 documents, while ZIP and RAR files are chunked into jobs of 25
documents. The size of these jobs is determined by document type, and it is hard-coded in the system.
For example, you might create an Import job and add it to Workgroup 1. Any worker machines assigned to
that workgroup will immediately begin processing the job. If you have a large job, you may want to add more
workers to the workgroup. The use of multiple workers splits a job across numerous machines leveraging
parallel processing, which increases the performance of the RPC.

Note: If you're starting an import job of a large PST or similar container you may note that only a single
worker in the assigned group is doing any work for an extended period of time. This is because the file
must be copied to the repository and have hash values calculated for it. This is not a distributable task and
can take considerable time on very large files which is normal. Once these tasks are complete the other
workers will be able to work on importing messages from the PST simultaneously.

Since you can constantly monitor the queue, you can also dynamically reallocate workers from one
workgroup to another as the job load requires. You must stop a worker in order to reallocate it to another
workgroup. The worker will complete its current job, and then display its status as Stopped.
You also have the option to bring the worker offline immediately. In this case, the worker doesn't complete
the current job. When you restart the worker, it automatically performs a series of cleanup tasks based on
stored procedures in the database. It then cleans up any entries added to the Matter table and other places
in the system, as well as removes any sub-jobs that it added to the queue. The restarted worker resets and
re-executes any of its open jobs from the beginning.
While you can queue multiple jobs to the same workgroup, we don't recommend this method because of the
first-in-first-out design of the queue. This method can create contention for shared resources that may
degrade performance if small and large jobs intermix. We instead recommend utilizing multiple workgroups
with their own workers processing a single job.

9.1 Starting worker machines

Use these instructions to start a worker machine:

1. In the View menu, click Workers Window. The Workers Activity window displays a list of available
workers and their current statuses.

Relativity Processing Console Guide 88

 2. From the Workers Activity window, highlight a worker machine with the Status of Stopped.

3. In the Workgroup box, select a group. The worker is now assigned to this group, and will only pro-
cess the jobs that are added to it.

4. Click Run to start the worker. Its status will be updated to Running.

9.1.1 Stopping versus taking worker machines offline

Note the following difference between stopping a worker versus taking it offline:

n
Stopping a worker - the worker stops receiving work from the queue, and it finishes only the tasks
it’s working on, not the entire root job. The remainder of that job is then available to be picked up by
another worker.

n
Taking a worker offline - the worker immediately stops and is brought offline, regardless of
whether or not it’s working on any tasks. After a short period of time the worker should bring itself
back online automatically. If you are concerned that a worker is hung up on a task it is recommended
to try stopping the worker first to allow as many of the tasks the worker has picked up to end grace-
fully. If the worker remains in a “stopping” state, then it may be necessary to bring the worker offline.

Note: In order to stop a worker and take a worker offline, the account running the RPC must be a local
administrator on the worker servers.

Relativity Processing Console Guide 89

9.2 Viewing worker activities
You can monitor the activities of a worker machine or the queue on the side panel in the Worker Activity
window. As illustrated below, you can highlight a worker and expand this panel to display statistics about the
number of microseconds (ms) taken to perform various tasks.

The panel to the right of the Worker Activity window provides the following information:

n Item - a list of tasks that the worker has performed since it was brought online.
n T (Threads) - a count of how many threads are currently working on the listed Item. It is recom-
mended to sort on this column so that items that are currently being worked on are always displayed
at the top of the list.
n Average - the average amount of time the worker has spent executing the items in that row since it
was last brought online (displayed in milliseconds).
n Max - the maximum amount of time the worker has spent executing an item in that row since it was
last brought online (displayed in milliseconds).
n Hit - the total number of times the worker has executed the item in that row since the worker was last
brought online.
Note the following details about the behavior of the Worker Activity window:

n If no worker is highlighted, the Tasks window displays the last worker selected.
n You aren't able to display tasks for all workers simultaneously. You can only display tasks for a single
worker.

Relativity Processing Console Guide 90

9.3 Viewing worker status
The following columns/fields are available in the Worker Activity window for you to monitor the current state
of your workers. Note that these also correspond to the data displayed in the Worker status tab in Relativity.
You can sort on any of the columns in this window.

n Status - reflects the current status of the worker. This displays one of the following values:
o Running - the worker thread is performing one of its designated jobs.
o Stopping - the worker has been instructed to stop and is attempting to complete and tasks that
it had already picked up. Once the worker has completed all tasks that it had picked up it will
change to a status of stopped. If the worker is unable to complete the tasks it has picked up the
worker will remain in a stopping state indefinitely and will need to be brought offline.
o Offline - the worker has been brought offline by the user or the InvariantWorker.exe program
was somehow forced to close. As long as the queue manager is running, the worker will auto-
matically try to bring itself back online even if it was deliberately taken offline by the user. While
the worker is offline only the Status, Name, ID, and Group columns will be populated.
o Stopped - the worker has been manually stopped. If the worker is either stopped of logged off,
the remaining columns will contain no data. A worker being brought online will briefly go into a
stopped state before going online.
o Logged Off - the worker is either powered off or the relativity service account is not currently
logged in as the user. While the worker is logged off only the Status, Name, ID, and Group
columns will be populated.

Note: The network on the Utility Server isn't set up to view the status of your workers;
therefore, you’ll see all workers logged off in the Worker Activity window in the RPC, and
you'll need to refer to the All Workers view of the Processing Administration tab tab in
Relativity to see the status of your workers. For details on the Processing Administration tab,
see the Processing User Guide.

n Name- the name of the worker.

n ID - the identifier of the worker.
n Tasks/min - how many singular units of designated work the worker is performing per minute.
o Examples of tasks are discovering a document and converting a native.
n Group - the workgroup the worker belongs to.

Relativity Processing Console Guide 91

 n CPU % - the amount of CPU resources the worker is using to perform its designated work across all
CPU's on the Invariant worker machine. If the actual CPU reading is between 0 and 1, Relativity
rounds this value up to 1.
o In most cases, a high CPU Activity reading with a low Memory in Use is indicative of smoothly
running worker that has no stuck jobs.
o This value can't necessarily tell you whether or not a worker is stuck on a job, but it can tell you
whether or not the worker is making a higher-than-normal effort to complete a job relative to
other jobs in the environment.
n Avail Mem (mb) - how many MB of RAM are still available after counting what the worker is currently
using.
n Temp (gb) - the amount of space in GB that is free on the disk drive assigned to the TEMP envir-
onment variable on the worker machine.
o The value will vary based on the disk's capacity.
o Only the disk associated with the TEMP environment variable is reflected here.
n Disk Busy % - the percentage of disk activity on the worker. If the actual disk activity reading is
between 0 and 1, Relativity rounds this value up to 1.
o If the disk drive is being hit excessively relative to other jobs, it can be an indicator that the
worker is either low on memory and burdening the paging file, or it can mean that it is working
on a large number of container files.
o If the disk drive activity is very high relative to other jobs for a sustained period of time, it's prob-
ably a sign that you're having an issue with your worker.
o Low CPU usage coupled with high disk activity can be indicative of low memory or a high con-
tainer count. You should always address low memory by adding more memory. With a high
container count, there's nothing to address, but note that performance can be impacted if the
disks are slow.

9.4 Naming a worker

To provide or edit a worker name, perform the following steps:

1. If the Worker Activity window isn't already open, navigate to the View menu and select Workers Win-
dow.

Relativity Processing Console Guide 92

2. In the Worker Activity pane, select the worker you want to name.
3. If the Properties window isn't already open on the right side, navigate to the View menu and select
Properties Window.

4. In the Properties Window on the right, expand the Workstation row, and scroll down the Description
row. This is the name of your worker. Edit the value directly to the right in the Description row by click-
ing once inside the current name and modifying it.

Relativity Processing Console Guide 93

 5. Click the refresh button in the Worker Activity pane to display the changed values.

9.4.1 Viewing basic worker properties

The Properties window provides information about the worker that you've selected in the Worker Activity
pane. To access this information, expand the Workstation column under the ActiveJobs column in the
Properties window. Note that the Workstation column represents the EDDS.Workstation table in the
database and that each value below it corresponds one-to-one with a column in the Workstation table.

Relativity Processing Console Guide 94

n Address - the address of the worker server, as specified in the Address column of the workstation
table.
n Available Categories - the categories available for selection when designating a worker for different
kinds of work in Relativity.
n Description - a description of the worker, as specified in the Description column of the workstation
table. Note that this value can be identical to the address and the workstation Name below.
n Enabled Categories - the categories of work for which the worker has been designated in Relativity.
For example, if the worker has been designated to perform native imaging and processing, those will
appear here.
n Flags - any database flags that have been applied to the worker.
n Location - the location of the worker server, as specified in the Location column of the workstation
table.
n Max Threads - the maximum number of threads the worker is allowed to make use of. If this is set to
0, the worker is permitted to make use of all available threads.

Relativity Processing Console Guide 95

 o The main reason to set this to something other than 0 is for troubleshooting purposes.
o On rare occasions, if you are aware that the worker machine needs to process difficult files
(large CAD drawings, blueprints, complicated excels, etc.), lowering the max threads can keep
the worker from opening too many of these simultaneously, which may result in out-of-memory
errors or similar issues.
o If you do set this to something other than 0 for one of these purposes make sure to reset it back
to 0 when done.
n Name - the name of the worker server, as specified in the Name column of the workstation table. This
can be identical to the Description and the Address columns.
n Resource Pools - the artifact IDs of the resource pools to which the worker belongs in Relativity.
n Workgroup - the number identifier of the workgroup to which the worker belongs.
n Workstation ID - the ID of the workstation as specified in the WorkstationID column of the work-
station table.

9.5 Other control options

The following table breaks down the other control options available in the worker activity window.

Menu
Description
Options
Refresh Displays newly added workers.

Brings a worker back online after being taken offline if you don’t want to wait for the worker to
Online come back online on its own in a few seconds.

Remote Opens a remote desktop connection to the worker, in which you can bring a worker back
Desktop online and add to your worker machines in order to troubleshoot issues.

Remote Logs in to the worker as the RCA if the worker has been logged off.

Logon
Remote Logs the worker off, which closes any tasks that were opened while logged in. This is
sometimes a useful last resort before rebooting a worker if it has been working on many
Logoff problem files that have been causing the worker to hang.
Reboot Reboots the worker if it needs it. If a worker is unresponsive this may work to reboot it but
usually if the worker is in such state it won’t respond to an instruction to reboot either. If the
Worker worker is up and running fine though and you need to cycle the worker, this button will work.
Remote Brings up a text box in which you can enter a command as though you had clicked Start then
command Run on the worker machine. This feature should not be needed for typical day to day work.

Relativity Processing Console Guide 96

Selected workers appear in a screen shot in the user interface bottom right corner, and updates every five
seconds.

9.6 Running a job

This section includes the steps for manually starting a job. You need to complete these steps when a job
has been added to workgroup that is not assigned to a started worker machine. You may need to complete
these steps for Group 0, the default workgroup in the RPC.
However, you can create jobs that will start automatically if you select workgroup that also has workers with
the status of Running. (For most job types, you can select the Workgroup on the General tab of the
associated job settings window.)

Note: Making changes to the workgroup or priority, or starting, stopping or deleting a job can have
serious consequences, especially to jobs that originated from Processing.

1. In the View menu, click Jobs Window. The Jobs window displays a list of available workers and their
current statuses.

Relativity Processing Console Guide 97

2. In the Job Activity window, highlight your job. It will have the status of NotStarted. See the Job Activ-
ity section in the RPC user interface on page 113 topic.

3. Click to start the job. The job's status is updated to Running. See Viewing worker activities on
page 90

Relativity Processing Console Guide 98

10 Performing Quality Control tasks
You can view extracted text and imaged documents, as well as perform other quality control tasks in the
Matter Inspector.

Note: For more information on how to use the matter inspector, see the Using the RPC Matter Inspector
video webinar on the Relativity Training site.

In the Matter Inspector you can:

n View all metadata for a selected file.

n Inspect multiple jobs.
n Apply filters to inspect specific files.
n View native files.

Note: There is no option for batching out documents for QC.

To QC documents:

1. In the Data Stores window, drill down to the import job that you want to review.
2. Right-click the import job, and click Inspect.

3. From the Matter Inspector window, click Refresh Needed to display a list of documents in the grid
box.

Relativity Processing Console Guide 99

 Note: In the Matter Browser, click the column heading to sort the list of documents. An indented file
is child of the file above it. Only one level of the parent-child hierarchy is displayed in the grid, but
the database records the full hierarchy.

Note: You're unable to add more columns to the Matter Browser.

10.1 Performing automatic QC

The RPC's QC process uses a check-in/check-out system, multiple users can work with the same document
at the same time. Automatic QC automatically checks out a document when navigating to it, and checks it
back in when it moves on the next one. Automatic QC is especially useful when you have to perform QC on
a large import job, as it presents you with a steady, controllable stream of images to inspect.
To perform automatic QC on a data set's documents:

Relativity Processing Console Guide 100

1. In the Matter Browser, click the Automatic QC icon.

2. If needed, in the Image Viewer, increase or decrease the length of the delay before the RPC moves to
the next image. Clicking the plus sign increases the delay and offers you more time to view the image

Relativity Processing Console Guide 101

and clicking the minus sign shortens the delay, thus making the QC process move faster.

Relativity Processing Console Guide 102

3. In the Image Viewer, click the play button to start the automatic advancement of images.

4. Perform QC on the images as they automatically appear, one at a time, in the Image Viewer.
n The RPC only flips through each image and presents them to you for your manual QC, it
doesn't perform the QC for you.
n A Placeholder is a document that the RPC couldn't image and thus automatically made a place-
holder for.

Note: You can't actually configure a placeholder in the RPC; you can only generate a blank
placeholder through the Generate Placeholder right-click option in the Matter Browser, or
you can upload a PDF file created outside of the RPC for the Custom placeholder PDF
option in the OCR/Image tab of the Job Settings window.

Relativity Processing Console Guide 103

 n
If you need more time on a single image, click to pause on that image.
5. When you're finished QC-ing, select the documents you want to mark as finalized and click the Fin-
alize QC button.

6. Note that the documents you marked as QC'd are green in the Matter Browser.

See Matter Inspector window on page 120 and Image Viewer window on page 123 for more QC options.
Alternatively, you can right-click on a file in the Matter Browser to display a pop-up menu. This menu
includes options for generating placeholders, displaying intermediate folders, and others.

Relativity Processing Console Guide 104

10.2 Non-image QC
The RPC gives you the option of performing QC on non-image files. You can do this in either the Extracted
Text or the Page Text viewers in the Image Viewer pane.
To perform non-image QC, follow these steps:

1. In the Matter Browser, select Exclude the following and select Has Images. Doing this will include
only non-image files from the import in the viewer.

Relativity Processing Console Guide 105

2. Click the Refresh Needed text when it appears. This will update the file list in the Matter Browser to
meet the criteria you've just set.

3. Open the Extracted Text or Page Text Viewer to perform QC on each file you select.

4. Once you've finished QCing a document, click the Finalize QC icon. Files for which you've finished
QC appear green in the Matter Browser.

Relativity Processing Console Guide 106

Once you've QC'd images, you can filter the data. See Filtering data on page 48.

Relativity Processing Console Guide 107

11 Running standard reports
This section includes generic steps for generating a standard reports listed on the Reports window. In the
Data Store window, you can also generate reports for a specific job by using the right-click menu.

1. In the Data Store window, highlight a data store that you want to report on, and click the Reports tab.
(You can also right-click on a data store, and select Reports from the menu.)

Relativity Processing Console Guide 108

2. Highlight one of the following reports in the list box. See Reports window on page 117.

3. Select the jobs that you want to include in the report.

4. To run the report on a subset of documents, select filters in the Filter to apply drop-down. Click
Add, and select the option to include or exclude the data returned by the filter.
n As an Include filter - Includes any document that matches the criteria set in this filter. For
example, you could add a keyword search as an include filter, so that all documents with this
keyword are exported.
n As an Exclude filter - Excludes any document that matches the criteria set in this filter.
n As a Privilege filter - This has no function in running reports.
5. (Optional) Select Mask non-relevant filters to hide any non-relevant filters from the list. If you only
want to see the filters that apply to one particular project out of several that exist in the RPC, you can
select only that project in the Reports tab and check this option. Doing this blocks the filters that apply
to the other projects from displaying in the Filters to apply drop-down.

Relativity Processing Console Guide 109

 6. (Optional) Click Use Export Filters to use the Export Filters on the report.

7. (Optional) Click Render Individually to view the report for each import job in a separate tab. When
you save the report with this option checked, individual files generate for each import job. The report
data for the selected import jobs is combined into one tab when you don't select this option. To render
reports individually, perform the following steps:
a. Select the report you'd like to run.
b. Select each import job you'd like to run an individual report for.
c. Check the box next to Render Individually.

d. Click View Report. Notice that a tab opens up for each import job you selected.
8. To create a report, perform one of the following tasks:
n Click Save Report to generate an Excel or Adobe PDF version of the report, and save it to dir-
ectory of your choice.
n Click View to display the report in the RPC.
There are additional reports available that you can run from the Data Stores window. See Right-click Data
Store options on page 114 for more information about these types of reports.

Relativity Processing Console Guide 110

11.0.1 Exporting a report
When you run a report on the data in your stores, you have the option of exporting it. To do this, click the
export icon in the tool bar at the top of the report window. You have the option of exporting the report as an
Excel, PDF, or Word file.

Relativity Processing Console Guide 111

12 Maintenance tasks
In the RPC, the Tools menu provides options for performing maintenance tasks such as updates to
databases updates, and XML file definitions.

1. Click the Tools menu to display a list of administrative options.

2. Select one of the following menu options:

n Load NIST Database - download the NIST database. Clicking this takes you to a folder
browser, from which you can select the folder where you've stored the NIST database. You
can procure the most recent version of the NIST database by contacting Support.
n Run Updates - updates the Invariant database schema to the most current version. This
includes all stores. Running updates on these databases isn't necessary if you've already
upgraded to the most recent version of Invariant. When you click this option, you bring up a win-
dow containing all databases available for upgrading.

Relativity Processing Console Guide 112

13 RPC user interface
This referential page describes the main RPC user interface windows and buttons. Find any specific RPC
task instructions in their respective topics.
The RPC user interface is comprised of a number of movable windows and panes. Click any of the windows
to un-dock, move, and re-dock panes across your desktop.

By default, only the Worker Activity, Job Activity, and Data Stores panes are open when you first access the
RPC.

Note: Open the various windows and position them to where you will always want them. Then close the
program without closing the individual windows first. If you do this, the desired window positions will be
stored to a configuration file so that each time you open the RPC, the windows will be where you want
them.

13.1 Data Stores window

The Data Stores window contains all the SQL Servers, data stores (workspaces), and imports done in those
data stores, along with a listing of all work performed on the imports.
The top-down hierarchy of the display is SQL server > Data store (workspace) / Groups > Project >
Custodian > Import > Jobs performed on an import.

Relativity Processing Console Guide 113

 Note: When you delete a workspace that corresponds to a data store in the RPC, that data store is not
automatically removed from the RPC.

13.1.1 Right-click Data Store options

The following actions are available when you right-click a data store and select the Tasks option:

Relativity Processing Console Guide 114

 n Import Data - import a data set to the selected data store.
The following options are only available after you've imported data and when you perform them, they're
applied to all imports in the data store. If you wish to perform any of these tasks on an individual import only,
right-click on the import job instead of on the data store:

n Extract Text - perform text extraction on a data set.

n Generate Images - generate images from a data set.
n Export - Export files for review or production.
n Update Search Indexes - update the search indexes of a data set.
n Inspect Multiple - inspect multiple jobs in a data store.
n Re-discover an item - re-discover an item by file ID.
You can right-click on a data store and select Filter(s) or Reports to access those windows for that data
store. You can also right-click on import and run one of these reports:

n Error report - gives a detailed list of errors on import job or any job performed on an import.
n Summary report - lists the frequency of each file type in the job, the total number of discovered doc-
uments and counts of de-NISTed files. It also includes file sizes of all files, de-NISTed files and the

Relativity Processing Console Guide 115

 size of files remaining after de-NISTIng not counting containers. Additionally, there is a brief entry
describing each error encountered during import.
n Bad Containers report - lists all container files that Invariant was unable to pull any documents from.
If a container throws an error on import but Invariant was able to get a single file from it, the container
will not appear on this report however it will appear on the error report.
n QC Report - provides information on who performed various image QC tasks and when.
The Lookup ID section, look at stored files or folders by entering the File ID or Storage ID of a document.
These are the choices available after entering an ID number:

n View Native File

n Explore Native Folder
n View Intermediate File
n Explore Intermediate Folder
n Explore Pages Folder

13.1.2 Hiding a data store

To hide a data store from the list, simply right-click on the store and select Hidden from the menu. The store
will no longer be visible in the list.
To un-hide the store so that it displays in the list, click the Show All option in the upper left of the window to
display all active and hidden stores, locate the hidden, grayed-out data store name, right-click on it and
select Hidden again. The store will now be visible in the list.

13.1.3 Deleting a data store

To delete a data store from the list, right-click on the store and select Delete Store from the menu. Note the
following details about store deletion:

n You can’t delete stores associated with active Relativity workspaces, meaning workspaces that
haven't been deleted.
n You can manually delete a store through the right-click menu as soon as the workspace is marked for
deletion by the Case Manager.
n If you don't manually delete a store and the workspace associate with it is deleted from Relativity, that
store will automatically be removed from the store list in the RPC.

13.2 Filters window

The Filters window contains RPC Filters for use. Right-click a data store > Filters to access the Filters
window or click the Filters tab at the bottom of the window. See Filtering data on page 48 for more
information.

Relativity Processing Console Guide 116

13.3 Reports window
The Reports window contains RPC reports to run on data sets in data stores. Select a data store and click
the Reports tab at the bottom of the window. See Running standard reports on page 108 for more
information.

Relativity Processing Console Guide 117

Search Filter Reports

n Comprehensive Hits - displays a list of search terms and the frequency that they occur in the doc-
ument set. You need to apply a search filter when you run this report.
Summary Reports

n Detailed Summary Report - contains data about the number and size of raw files, number of files
before and after De-NISTing, number of e-mail messages and their attachments, and other inform-
ation.
n File Types - lists the extensions of all the files in the data set, and the number of occurrences for
each one. It also includes summary totals.
n FileTypes and Size - lists the extensions of all the files in the data set, the number of occurrences for
each one, and file sizes. It also includes summary totals.
n OCR Page Count - lists the number of OCR'd pages.
n Page Count- lists the total page count, in addition to pages marked as deleted and pages marked as
delete candidates.
Error Reports

n Exception Report - displays the exception totals for Unprocessable files, Password protected files,
and Errored Files.
n Password Report - displays the exception totals for Password protected files and total file size
count.

Relativity Processing Console Guide 118

13.4 Worker Activity window
The Worker Activity window contains a list of all workers along with information on their current tasks per
minute, group assignment, and CPU usage. Click View > Workers window to access the Workers window.
See Managing workers and jobs on page 88 for more information.

13.4.1 Worker Activity menu options

Menu Options Description

All Workers drop-down list Displays list of all workers.
Refreshes the Worker Activity window.
Refresh
Starts the selected worker.
Start
Stops the selected worker.
Stop
Takes the selected worker offline.
Offline
Brings the selected worker online.
Online

Relativity Processing Console Guide 119

 Menu Options Description
Open a remote desktop connection to the worker.
Remote Desktop
Log in to a remote machine.
Remote Logon
Log off of a remote machine.
Remote Logoff
Reboot a worker.
Reboot Worker
Remote command Execute a remote command.

Workgroup drop-down list Displays the selected Worker's workgroup.

13.5 Matter Inspector window

The Matter Inspector window displays a list of imported data files and a fixed set of metadata associated
with them. See Performing Quality Control tasks on page 99 for more information.

13.5.1 Matter Browser menu options

Menu options Description
Click this icon to display the Select Filters dialog box. See Filtering data on
Filter page 48.
Use this option to show or hide duplicate documents. Duplicate documents
Show Duplicates are shaded in pink.
Use this option to show or hide containers, including ZIP, RAR, and PDF
Show Containers portfolios. (PDF portfolios contain other PDFs, but do not have content of their
own.)
Use these options when you want to check out a document so no other users
Check out can perform QC tasks on it, and then check the document in when the QC is
completed. (Invariant will skip documents that are checked out when you
Check in browse through the list.)

Relativity Processing Console Guide 120

 Menu options Description
Highlight documents that you want marked as QC completed, and click this
Finalize QC icon. The Matter Browser then displays these documents with green shading.
Click this option to automatically flag documents as being QC’d, and then
Automatic QC click the Play button in the Image Viewer to page through the documents.
Promote Delete Can- Click this option to change pages flagged as Delete Candidate to Delete.
didates Note that this action will be applied to all documents currently displayed in the
Matter inspector. Pages with a delete flag will not be exported.
Mark Integrity Check Click this option to mark this QC task as being completed. This verification
step is listed in the QC Report.
Completed
Mark Placeholder Check Click this option to mark this QC task as being completed. This verification
step is listed in the QC Report.
Completed
Mark Page by Page QC Click this option to mark this QC task as being completed. This verification
step is listed in the QC Report.
Completed
Generates a report that displays who completed the above QC steps and
Generate QC Report when they were completed.
Exclude the Following Use this option to exclude documents that have been flagged with any of the
options listed below.
Deleted, Followup, Has Use these checkboxes to filter on documents with these flags that you do or
OCR, Has Images, do not want displayed in the grid based on the setting selected for ‘Exclude
Placeholder, Redaction, the Following’. For example, you can select Delete Candidate if you want to
Hidden Text, Any Flag, QC display documents that have pages with the Delete Candidate flag so that you
Done, Passes Integrity can QC them before promoting them to Deleted.
Check
n The Redaction flag identifies files that have been named with a naming
convention of "storageID_R.PDF" and copied into the intermediate
folder. The RPC supports using these redacted PDF's, but has no
redaction capability itself.

Note: The Matter Inspector is not necessarily reflective of the complete document list that you would be
exporting. This is because the inspector isn’t concerned with document relationships, as it is merely a
place for you to inspect documents that hit on, for example, any keywords you specified in your report
settings. This means that if you don’t flag parents or children and you review the resulting documents in
the Matter Inspector, you may see a different number of results than you’d see in the comprehensive hit
report. Likewise, the number of documents that you export might be different than what you viewed in the
Matter Inspector because the exporter has to honor document relationships, and thus it will not export
child documents without their parents.

Relativity Processing Console Guide 121

13.5.2 Right-click Matter Browser options
The following options are available to you when you right-click on a record in the Matter Browser:

n Copy Selected Cells - copies the cells you’ve selected to the clipboard. There is no subsequent win-
dow or explicit indication that you’ve taken action on the cells you’ve selected, but they have been
copied once you select this option. This is the only right click option that can be used on more than a
single record. Everything below will only work on the document where the mouse is when the right
click was executed regardless of how many documents are highlighted.
n Regenerate Images/OCR - opens the Job Settings window, where you can switch to the Image or
OCR tab to reconfigure your OCR and image settings and perform an imaging job on the selected
document. Once you have made the desired settings click start and the document will be placed in
the job queue.
n Extract Text - queues the selected document for text extraction. It does not first bring up the Job Set-
tings window as the Regenerate Images/OCR option does. If you wish to make changes to the set-
tings before executing extracting text you will need to right click on the appropriate import job in the
Data Stores window and select ‘Settings’. If the selected document has already been text extracted
nothing will happen unless the ‘Overwrite intermediate files’ setting on the Text tab is checked.
n Generate Placeholder - creates a generate placeholder job in the queue for the selected document.
If you’ve already specified a file for the Custom placeholder PDF setting in the OCR/Image tab of the
Job Settings window, the placeholder that you generate through this right-click option will display as
that file. Thus, there is no option for you to override a custom placeholder PDF in the Matter Browser.

Relativity Processing Console Guide 122

 n Rediscover - starts a rediscover job for that file. This job appears in the Job Activity window as a
ReImport type. Note that if you’ve already exported this file, you will receive the following confirmation
message. Click Yes to continue.

n Convert to Container - takes what is currently a document in the job and converts it to a container
file.
o Containers are not exported or published, and their children are promoted up a level. For
example, if a parent Word document with two child attachments is converted to a container file,
then the two child attachments would become parents instead of children. They would then dis-
play as coming from a Word document parent container.
o This option is useful when the RPC misidentifies a container file as a document, for example if
it misidentifies a ZIP file and treats it like a document. In this case, the option to convert the doc-
ument to a container allows you to correct the relationship.
o When you select this option, the record is no longer available for individual inspection, as you
aren't able to inspect containers.
o You also have the option to convert a container to a document, which does the reverse of con-
tainer conversion.
n Reports - this option currently provides no functionality.

13.6 Image Viewer window

The Image viewer window contains the selected document's images. See Performing Quality Control tasks
on page 99 for more information.

Relativity Processing Console Guide 123

13.6.1 Image viewer menu options
Menu options Description
Use this option to perform a visual inspection of the images by paging through
Play them. (You can use this option in combination with the Automatic QC button
in the Matter Browser to confirm that the images are displayed correctly.)
Show All Shows all pages of the current document regardless of what page level filters
may be applied (e.g. Deleted, Delete Candidates).
Set the number of images that you want to advance by. Regardless of what
this number is set to, the image viewer will always show at least the first page
of each document.
Options These options are available when using the Image Viewer window:

n Skip QC’d items during auto-advance. This option is on by default.

n Apply current flags to the remaining pages in this doc. Useful most
often when you want to add or remove the Deleted or Delete Candidate
flag from the remaining pages in the document.
n Apply current flags until stopped. Useful when needing to apply or
remove a flag on many consecutive pages but not necessarily to the
remainder of the document.
n Apply rotation settings when applying flags.
Anti-Alias Toggles on or off anti-aliasing to the displayed document to sharpen the
image. Having this turned on will slow down the auto-advance of the images.
Use this option to show or hide the image layer of the intermediate PDF
Show Images currently being displayed in the image viewer.

Relativity Processing Console Guide 124

 Menu options Description
Use this option to show or hide the line art layer of the intermediate PDF
Show Line Art currently being displayed in the image viewer. (Line art is most commonly
associated with CAD drawings.)
Show Text Use this option to show or hide the text layer of the intermediate PDF cur-
rently being displayed in the image viewer.
Use this option to display an image in black and white.
Black and White
Delete Candidate Use this option to flag pages for deletion from the export. This is especially
useful for when you encounter blank pages among a set of images and you
want to exclude them from the export file.
Deleted, Followup, Has Use these checkboxes to flag images. The setting for the flag is stored in the
OCR, Has Images, database.
Placeholder, Redaction,
Hidden Text Note: Images marked as delete are not exported, but they remain in the
database.

13.7 Extracted Text Viewer window

The Extracted Text Viewer window contains a selected item's extracted text.

Relativity Processing Console Guide 125

13.7.1 Extracted Text Viewer menu options
Menu Options Description
Lock/Unlock Grid Locks/unlocks the grid for editing.

Toggle Word Wrap Toggle the word wrapping on and off.

Select a different extracted text font to use.

Select Font
Save your changes.
Save Changes

13.8 Page Text Viewer window

The Page Text Viewer window contains a selected item's page text. Click the horizontal Page Text Viewer
button on the right to open the Page Text Viewer window.

Relativity Processing Console Guide 126

The Page Text Viewer window menu options are the same as the Extracted Text Viewer menu options on
the previous page.

13.9 Job Activity window

The Job Activity window contains all RPC active jobs and corresponding statuses.

13.9.1 Job Activity menu options

Menu
Description
Options
All Stores Allows the user to select a specific data store to display jobs only from the selected store.
drop-down
list
Filter Icon Displays a list of job statuses to be displayed in the Job Activity window with these options:

n Not Started
n Picked Up
n Pending Worker
n Started
n Finished
n Error

Relativity Processing Console Guide 127

 Menu
Description
Options

n Interrupted
n Stopped
n Paused
n Waiting
n Canceling
n Canceled

Note: The ‘Finished’ status is unchecked by default. It is strongly recommended to

change the All Stores drop-down list to a specific data store before checking the ‘Finished’
option.
Not doing so will result in every job from every data store being displayed which can
significantly slow the RPC or cause the program to lock up.

Refreshes the Job Activity window.

Refresh
Starts the selected job.
Start
Stops the selected job.
Stop
Delete Deletes the selected job.
Workgroup Displays the selected job workgroup.
Priority Displays the selected job priority.

Note: Making changes to the workgroup or priority, or starting, stopping or deleting a job can have
serious consequences, especially to jobs that originated from Relativity processing. For more information
on what will occur when using these features see Managing workers and jobs on page 88.

13.10 Configuring job settings

All of the tabs in the Settings pane are available at all times while you are in the RPC, regardless of what
operation you’re currently configuring. For example, the import settings are available even when you’re
preparing a data extraction job, even though import settings aren’t logically applicable to extraction. Some of
the Settings overlap, in that they are applicable to more than once type of job.
If you anticipate having to repeatedly perform jobs with identical settings, you have the option of saving a
group of settings as the default for those future jobs.
To save your settings as the default:

Relativity Processing Console Guide 128

1. Select the job from the Data Stores pane, right-click, and select Settings…

2. Specify settings for all applicable tabs in the Job Settings window.

Relativity Processing Console Guide 129

3. Click Save as Default at the bottom left of the window.

4. Select whether to save your settings to a (Data) Store or a Project profile by clicking Store or Project.
If you haven't specified a project name, you'll only have the option to save the defaults to a store. You
aren't able to save the settings of the following areas as defaults:
n The General tab
n The File Handling section of the Import tab
n The E-Mail | Ignore PST/OST CRC errors in the Import tab
5. Note the confirmation screen, which states that your settings have been saved and are now the
defaults for the store or project.

Relativity Processing Console Guide 130

13.11 Properties window
The properties window is a dynamic display of information related to whatever item is currently highlighted
in the RPC. It displays information about a specific worker, job, data store or document, as well as any
information stored in the underlying SQL tables on the given item being examined. You can also edit some
information here.

n For workers, information about what types of work it the worker is set up to do is displayed, and you
can edit the name of the worker , along with setting the max threads the worker can utilize. For a
breakdown of the worker properties, see Viewing basic worker properties on page 94.
n For jobs, information about doc counts, error counts, Start/Finished/Last Progress times are dis-
played.
n The properties of a given data store are also available.
n The properties window can be especially useful when used in conjunction with the matter inspector.
Here, you can click a specific document and find all available metadata on that document. You can
also edit metadata here, even though this isn’t something that would be done in a typical workflow.
You can click and highlight the following items to display their properties:

n Data Stores in the Data Stores window

n Jobs in either the Data Stores or Job Activity windows
n Workers or the Queue Manager in the Worker Activity window
n Individual documents in the Matter Inspector
Clicking on the Project or Custodian level item in the Data Stores window or any item in the Reports or
Filters windows results in nothing being displayed in the Properties window.

Relativity Processing Console Guide 131

n
The Categorized or Alphabetical buttons will change the sort order of the various properties
displayed. The default is Categorized.

n
The Property Pages and blank drop-down above these buttons are currently non-functional.

Relativity Processing Console Guide 132

14 RPC known issues
The following columns are included in the table:

n Item ID - the identifying number applied to an issue when it's reported.

n Affected feature - the RPC feature to which the defect applies.
n Description - a description of the issue.
o Resolution - a resolution for the defect, if known.
Click the Item ID or Affected feature columns to sort defects.

Item Affected fea-

Description
ID ture
REL- Data Stores The Data Stores tree doesn't automatically update after you run an import job.
17034 Resolution
Manually collapse and expand the tree to update the Data Stores tree.
REL- Configuration The RPC doesn't display workstation information (such as the Data Store tree or
17036 jobs) when the AppSettings table isn't populated.
Resolution
Perform the required post-upgrade step in the Worker manager server
installation guide.
REL- Exporting The Job Settings options accepts bogus data. For example, Advanced
17037 tab.DestinationPath.
Resolution
There's currently no resolution for this issue.
REL- Importing The Additional Settings job dialog appears first time with "null _applyTo" object
17039 and displays a System Exception for NULL object reference.
Resolution
There's currently no resolution for this issue.
REL- UI behavior The comment box clears out when you paste in long text paragraphs exceeding
17049 the maximum character limit. This occurs in the Export Wizard, Import Job
Project, and Custodian dialog comment boxes.
Resolution
There's currently no resolution for this issue.
REL- Exporting The Task metadata in the Task window clears out when you click Show All
17054 Metrics. If you click Show All Metrics again, the metadata remains cleared out.
Resolution
There's currently no resolution for this issue.
REL- Importing When creating a new Data Store, the Additional Settings button on the Import
17082 dialog window is disabled.
Resolution
There's currently no resolution for this issue.
REL- UI behavior The Next button on the Task window is enabled without any task selected.

Relativity Processing Console Guide 133

Item Affected fea-
Description
ID ture
17161 Resolution
There's currently no resolution for this issue.
REL- UI behavior The Context menu choice View Tracking Log is disabled.
17201 Resolution
There's currently no resolution for this issue.
REL- Exporting The Want Containers check box option has no effect when selected in Export
17266 options wizard.
Resolution
There's currently no resolution for this issue.
REL- Exporting The ZIP file isn't included in .LFP output file when selected in the Export options
17322 wizard.
Resolution
There's currently no resolution for this issue.
REL- Exporting In the Export Wizard, the dialog boxes allow invalid inputs such as negative
17346 numbers, decimals, and letters.
Resolution
There's currently no resolution for this issue.
REL- Exporting In the Export Field Editor, the Array Merge functionality Write Metatdata and
17577 Copy PDF tasks export results are unexpected.
Resolution
There's currently no resolution for this issue.
REL- Exporting The Descending setting has no effect when selected in the Export options
17633 wizard.
Resolution
There's currently no resolution for this issue.
REL- UI behavior The Field and Text delimiters don't auto-fill in the Field Editor.
17762 Resolution
There's currently no resolution for this issue.
REL- UI behavior The Any Flag button doesn't display items with a flag, even when items are
17820 flagged.
Resolution
There's currently no resolution for this issue.
REL- UI behavior The QC Finalize button doesn't display items that have been QC'd.
17821 Resolution
There's currently no resolution for this issue.
REL- UI behavior The Promote Delete Candidates button doesn't promote items to be deleted.
17822 Resolution
There's currently no resolution for this issue.
REL- UI behavior The Delete button isn't enabled for jobs with a status of Waiting.
17853

Relativity Processing Console Guide 134

Item Affected fea-
Description
ID ture
Resolution
There's currently no resolution for this issue.
REL- Filtering The Remove Filter button doesn't always work. This usually occurs when you
17942 create two identical filters and then select the "wrong one" to delete.
Resolution
There's currently no resolution for this issue.
REL- UI behavior The RPC crashes when you attempt to rediscover an invalid item.
18199 Resolution
There's currently no resolution for this issue.
REL- UI behavior The RPC crashes when you attempt to rediscover an invalid item with an input
18200 string.
Resolution
There's currently no resolution for this issue.
REL- UI behavior The Rediscover matter doesn't update file document messages with errors.
18210 Resolution
There's currently no resolution for this issue.
REL- Filtering Extra fields aren't populated with input values when you create a method in the
18416 Expression filter.
Resolution
There's currently no resolution for this issue.
REL- UI behavior The RPC crashes when you cancel an Extract text job on an import job that
18596 already has text extraction job.
Resolution
There's currently no resolution for this issue.
REL- UI behavior The ActivityTimer window sorts numbers in the Average and Max columns as if
18625 they were strings instead of numbers.
Resolution
There's currently no resolution for this issue.
REL- UI behavior The RPC crashes when you submit a job without an ActivationKey.
18666 Resolution
Submit one processing or inventory job from Relativity processing before using
the RPC.
REL- Filtering An error occurs when you enter a file location in the Cross reference file window.
19049 Resolution
There's currently no resolution for this issue.
REL- Filtering An error doesn't occur when you enter an invalid file location in the Cross
19052 reference file window.
Resolution
There's currently no resolution for this issue.
REL- Reporting The Job Error Report incorrectly states that no errors are found, even when

Relativity Processing Console Guide 135

Item Affected fea-
Description
ID ture
19194 errors exist.
Resolution
There's currently no resolution for this issue.
REL- Importing When you create an import job with the Extract Text option set to true, The
19216 export job doesn't create.
Resolution
There's currently no resolution for this issue.
REL- UI behavior The Matter Browser opens incorrect or unexpected selected files. For example,
19221 the Matter Browser opens native files when you select intermediate files and
vice versa.
Resolution
There's currently no resolution for this issue.
REL- Filtering Enter the exact Filter name doesn't submit the Update Index job.
19222 Resolution
There's currently no resolution for this issue.
REL- Filtering When using the Cross Reference Import filter, the importer will ignore the 1st line
19255 (Header row) and import the rest of the content.
Resolution
There's currently no resolution for this issue.
REL- Filtering The Cross References Import filter isn't able to parse filter results properly, when
19257 there are spaces in front of "".
Resolution
There's currently no resolution for this issue.
REL- Filtering Unicode characters display as number or letters (incorrectly) in the Filter Results
19261 table when you run a Full Text search on the characters .
Resolution
There's currently no resolution for this issue.
REL- Filtering The Regular Expression option in the Full Text filter doesn't display accurate
19403 results in the Filter Results table.
Resolution
There's currently no resolution for this issue.
REL- Filtering The Phonic option in Full Text filter doesn't display correct results for similar
19404 sounds words.
Resolution
There's currently no resolution for this issue.
REL- Data Stores An Invalid path in the Import Data Source displays files from an unknown
19609 location.
Resolution
There's currently no resolution for this issue.

Relativity Processing Console Guide 136

15 Field mappings
The following table provides a breakdown of which processing fields match up with the most common fields
in the RPC.

Processing
Relativity Field Is
RPC field name Description
field/source type Unicode?
name
Attachment ChildRelativityControlNumbers Long Yes Attachment document IDs
Document IDs Text of all child items in family
group, delimited by semi-
colon, only present on par-
ent items.
Attachment List ChildFileNames Long Yes Attachment file names of
Text all child items in a family
group, delimited by semi-
colon, only present on par-
ent items.
Author Author Fixed- Yes Original composer of doc-
Length ument or sender of email
Text (50) message.
BCC Address EmailBCCSmtp Long Yes The full SMTP value for
Text the email address entered
as a recipient of the Blind
Carbon Copy of an email
message.
CC Address EmailCCSmtp Long Yes The full SMTP value for
Text the email address entered
as a recipient of the Car-
bon Copy of an email mes-
sage.
Child MD5 Hash ChildMD5Hashes Long Yes Attachment MD5 Hash val-
Values Text ues of all child items in a
family group, delimited by
semicolon, only present
on parent items. The RPC
can't calculate this value if
you have FIPS (Federal
Information Processing
Standards cryptography)
enabled for the worker
manager server.
Child SHA1 ChildSHA1Hashes Long Yes Attachment SHA1 Hash

Relativity Processing Console Guide 137

 Processing
Relativity Field Is
RPC field name Description
field/source type Unicode?
name
Hash Values Text values of all child items in
a family group, delimited
by semicolon, only
present on parent items.
Child SHA256 ChildSHA256Hashes Long Yes Attachment SHA256 Hash
Hash Values Text values of all child items in
a family group, delimited
by semicolon, only
present on parent items.
Comments Comments Long Yes Comments extracted from
Text the metadata of the native
file.
Company Company Fixed- Yes The internal value entered
Length for the company asso-
Text ciated with a Microsoft
(255) Office document.
Contains Office/EmbeddedItems Yes/No N/A The yes/no indicator of
Embedded Files whether a file such as a
Microsoft Word document
has additional files embed-
ded in it.
Control Number BatesBeginAttach Fixed- Yes The identifier of the first
Beg Attach Length page of the first document
Text (50) in a family group. This is
used for page-level num-
bering schemes.
Control Number BatesEndAttach Fixed- Yes The identifier of the last
End Attach Length page of the first document
Text (50) in a family group. This is
used for page-level num-
bering schemes.
Conversation EmailConversation Long Yes Normalized subject of
Text email messages. This is
the subject line of the
email after removing the
RE and FW that are
added by the system
when emails are for-
warded or replied to.

Relativity Processing Console Guide 138

 Processing
Relativity Field Is
RPC field name Description
field/source type Unicode?
name
Conversation ConversationFamily Fixed- Yes Relational field for con-
Family Length versation threads. This is
Text (44) a 44-character string of
numbers and letters that is
created in the initial email.
Conversation EmailConversationIndex Long Yes Email thread created by
Index Text the email system. This is a
44-character string of num-
bers and letters that is cre-
ated in the initial email and
has 10 characters added
for each reply or forward
of an email.
Date Created CreatedOn Date N/A Date and time from the
Date Created property
extracted from the original
file or email message.
Date Last Modi- LastModified Date N/A Date and time from the
fied Modified property of a doc-
ument, representing the
date and time that
changes to the document
were last saved.
Date Last Prin- LastPrinted Date N/A Date and time that the doc-
ted ument was last printed.
Date Received EmailReceivedOn Date N/A Date and time that the
email message was
received (according to ori-
ginal time zones).
Date Sent EmailSentOn Date N/A Date and time that the
email message was sent
(according to original time
zones).
DeDuped Cus- DeDupedCustodians Multiple N/A Custodians associated
todians Object with the de-duped records
of this document.
DeDuped Paths DeDupedPaths Long Yes Folder structure and paths
Text to this document?s duplic-
ates. Each path will con-

Relativity Processing Console Guide 139

 Processing
Relativity Field Is
RPC field name Description
field/source type Unicode?
name
tain the associated cus-
todian.
Delivery Receipt EmailDeliveryReceiptRequested Yes/No N/A The yes/no indicator of
whether a delivery receipt
was requested for an e-
mail.
Document RelativityDocumentClass Single N/A A single choice field that
Class Choice can be one of: Email,
Edoc, or Attach.
Document FileExtension Fixed- Yes Three (or more) character
Extension Length extension of the document
Text (25) that represents the file
type to the Windows Oper-
ating System. Examples
are PDF, DOC, or DOCX.
Document Sub- DocumentSubject Long Yes Subject of the document
ject Text extracted from the prop-
erties of the native file.
Domains (Email Email/DomainParsedBCC Multiple N/A Domains of 'Blind Carbon
BCC) Object Copy' recipients of the
email message.
Domains (Email Email/DomainParsedCC Multiple N/A Domains of 'Carbon Copy'
CC) Object recipients of the email
message.
Domains (Email Email/DomainParsedFrom Multiple N/A Domains of Originator of
From) Object the email message.
Domains (Email Email/DomainParsedTo Multiple N/A Domains of 'To' recipients
To) Object of the email message.
Email BCC EmailDisplayBCC Long Yes Recipients of 'Blind Car-
Text bon Copies' of the email
message.
Email Cat- EmailKeywords Long Yes Category assigned to an
egories Text email message.
Email CC EmailDisplayCC Long Yes Recipients of 'Carbon Cop-
Text ies' of the email message.
Email From EmailDisplaySender Fixed- Yes Originator of the email
Length message.

Relativity Processing Console Guide 140

 Processing
Relativity Field Is
RPC field name Description
field/source type Unicode?
name
Text
(255)
Email In Reply Email/In_Reply_To Long Yes The internal metadata
To ID Text value within an email for
the reply to ID.
Email Store EmailContainer Fixed- Yes The identifier of the top-
Name Length level container of an email
Text message. For example,
(255) 'jdoe.nsf.' If a document
comes from a rar/zip file
attached to the email, the
container is referred to in
that file.
Email Subject EmailSubject Long Yes Subject of the email mes-
Text sage.
Email To EmailDisplayTo Long Yes List of recipients or
Text addressees of the email
message.
File Name FileName Fixed- Yes Original name of the file.
Length
Text (25)
File Size FileSize Decimal N/A Generally a decimal num-
ber indicating the size in
bytes of a file.
File Type FileType Fixed- Yes Description that rep-
Length resents the file type to the
Text Windows Operating Sys-
(255) tem. Examples are Adobe
Portable Document
Format, Microsoft Word
97 - 2003 Document, or
Microsoft Office Word
Open XML Format.
From Address EmailSenderSmtp Long Yes The full SMTP value for
Text the sender of an email
message.
Group Identifier RelativityGroupId Fixed- Yes Group the file belongs to
Length (used to identify the group
Text (40) if attachment fields are not

Relativity Processing Console Guide 141

 Processing
Relativity Field Is
RPC field name Description
field/source type Unicode?
name
used).
Has Hidden HiddenText Yes/No N/A Indication of the existence
Data of hidden document data
such as hidden text in a
Word document, hidden
columns, rows, or work-
sheets in Excel, or slide
notes in PowerPoint.
Importance EmailImportance Single N/A Notation created for email
Choice messages to note a higher
level of importance than
other email messages
added by the email ori-
ginator.
Keywords Office/Keywords Long The internal value entered
Text for keywords associated
with a Microsoft Office doc-
ument.
Last Accessed LastAccessed Date N/A The date and time at
Date/Time which the loose file was
last accessed.
Last Saved By Office/LastAuthor Fixed- Yes The internal value indic-
Length ating the last user to save
Text a document.
(255)
Last Saved LastSaved Date N/A The internal value entered
Date/Time for the date and time at
which a document was
last saved.
Lotus Notes Lotus/OtherFolders Long Yes A semi-colon delimited list-
Other Folders Text ing of all non-primary
folders that a Lotus Notes
message or document
was included.
MD5 Hash MD5Hash Fixed- Yes Identifying value of an
Length electronic record that can
Text (40) be used for deduplication
and authentication gen-
erated using the MD5

Relativity Processing Console Guide 142

 Processing
Relativity Field Is
RPC field name Description
field/source type Unicode?
name
hash algorithm.
Meeting End Email/EndDate Date N/A The date and time at
Date/Time which a meeting item in
Outlook or Lotus Notes
ended.
Meeting Start Email/StartDate Date N/A The date and time at
Date/Time which a meeting item in
Outlook or Lotus Notes
began.
Message Email/MessageHeader Long Yes The full string of values
Header Text contained in an email mes-
sage header.
Message ID EmailMessageID Fixed- The message number cre-
Length ated by an email applic-
Text ation and extracted from
(255) the email's metadata.
Message Type Email/MessageType Single N/A An indication of the email
Choice system message type.
Possible values include
Appointment, Contact, Dis-
tribution List, Delivery
Report, Message, or
Task. The value may be
appended with '(Encryp-
ted)' or 'Digitally Signed'
where appropriate.
Native File StoredAs Long Yes The path to a copy of a file
Text for loading into Relativity.
Number of AttachmentCount Whole N/A Number of files attached
Attachments Number to a parent document.
OCR Text HasOcrText Yes/No N/A The yes/no indicator of
whether the extracted text
field contains OCR text.
Office Docu- Office/Manager Fixed- Yes The internal value entered
ment Manager Length for the manager of a doc-
Text ument.
(255)
Office Revision Office/Revision Fixed- Yes The internal value for the

Relativity Processing Console Guide 143

 Processing
Relativity Field Is
RPC field name Description
field/source type Unicode?
name
Number Length revision number within a
Text Microsoft Office doc-
(255) ument.
Other Props OtherProps Long Yes Metadata extracted during
Text processing for additional
fields beyond the list of
processing fields available
for mapping. Field names
and their corresponding
values are delimited by
semicolon.
Parent Docu- ParentRelativityControlNumber Fixed- Yes Document ID of the parent
ment ID Length document. This field will
Text (50) only be available on child
items.
Password Pro- PasswordProtected Single N/A An indication of doc-
tected Choice uments that were pass-
word protected. It will
contain the value 'Decryp-
ted' if the password was
identified; 'Encrypted' if
the password was not
identified; or no value if
the file was not password
protected.
Primary Date DocDate Date N/A Date taken from Email
Sent Date, Email
Received Date, or Last
Modified Date (in order of
precedence).
Read Receipt EmailReadReceiptRequested Yes/No N/A The yes/no indicator of
whether a read receipt
was requested for an e-
mail.
Sensitivity EmailSensitivity Single N/A The indicator set on an
Choice email to denote the
email's level of privacy.
SHA1 Hash SHA1Hash Fixed- Yes Identifying value of an
Length electronic record that can
Text (50)

Relativity Processing Console Guide 144

 Processing
Relativity Field Is
RPC field name Description
field/source type Unicode?
name
be used for deduplication
and authentication gen-
erated using the SHA1
hash algorithm.
SHA256 Hash SHA256Hash Fixed- Yes Identifying value of an
Length electronic record that can
Text (70) be used for deduplication
and authentication gen-
erated using the SHA256
hash algorithm.
Sort Date SortDate Date N/A Date taken from Email
Sent Date, Email
Received Date, or Last
Modified Date (in order of
precedence) repeated for
parent document and all
children items to allow for
date sorting.
Speaker Notes PowerPoint/HasSpeakerNotes Yes/No N/A The yes/no indicator of
whether a powerpoint file
has speaker notes asso-
ciated with its slides.
To Address EmailToSmtp Long Yes The full SMTP value for
Text the recipient of an email
message. For example,
'[email protected]'
Track Changes TrackChanges Yes/No N/A The yes/no indicator of
whether tracked changes
exist in the document.
Unified Title UnifiedSubject Long Yes Subject of the document.
Text If the document is an
email, this field contains
the email subject. If the
document is not an email,
this field contains the doc-
ument's file name.
Unprocessable Unprocessable Yes/No N/A The yes/no value indic-
ating if a file was able to
be processed. If the file

Relativity Processing Console Guide 145

 Processing
Relativity Field Is
RPC field name Description
field/source type Unicode?
name
could not be processed,
this field is set to Yes.
Unread Flag EmailIsUnread Yes/No N/A The yes/no indicator of
whether an e-mail was not
read.

Relativity Processing Console Guide 146

Proprietary Rights
This documentation (“Documentation”) and the software to which it relates (“Software”) belongs to
Relativity ODA LLC and/or Relativity’s third party software vendors. Relativity grants written license
agreements which contain restrictions. All parties accessing the Documentation or Software must: respect
proprietary rights of Relativity and third parties; comply with your organization’s license agreement,
including but not limited to license restrictions on use, copying, modifications, reverse engineering, and
derivative products; and refrain from any misuse or misappropriation of this Documentation or Software in
whole or in part. The Software and Documentation is protected by the Copyright Act of 1976, as amended,
and the Software code is protected by the Illinois Trade Secrets Act. Violations can involve substantial
civil liabilities, exemplary damages, and criminal penalties, including fines and possible imprisonment.
©2025. Relativity ODA LLC. All rights reserved. Relativity® is a registered trademark of Relativity
ODA LLC.

Relativity Processing Console Guide 147