Jump to: navigation, search

max-backup-index

Section: log4j
Default Value: 10
Valid Values: Any positive integer
Changes Take Effect: Immediately
Dependencies: None

Specifies the maximum number of backup log files that are kept in addition to the active log file.

days-to-keep-output-files

Section: gim-export
Default Value: 14
Valid Values: Any positive integer
Changes Take Effect: On the next job run
Dependencies: None
Introduced: 8.5.005

Specifies how many days to store exported files before deleting them.

days-to-keep-discards-and-job-history

Section: gim-etl
Default Value: 600
Valid Values: Any positive integer
Changes Take Effect: At the next run of Job_MaintainGIM
Dependencies: None

Specifies the number of days to retain data in the discard tables and audit and history tables.

more...

days-to-keep-cfg-facts

Section: gim-etl
Default Value: 400 or days-to-keep-gim-facts
Valid Values: Greater than, or equal to, days-to-keep-gim-facts
Changes Take Effect: At the next run of Job_MaintainGIM
Dependencies: days-to-keep-gim-facts
Introduced: 8.5.003

Specifies the number of days to retain deleted configuration fact data in GIDB and relevant fact tables. Facts that have an end time that is earlier than the retention period are eligible to be purged. The value of the days-to-keep-cfg-facts configuration options must always be greater than, or equal to, days-to-keep-gim-facts.

days-to-keep-gdpr-history

Section: gim-etl
Default Value: 15
Valid Values: An integer in the range 0-30
Changes Take Effect: At the next run of Job_MaintainGIM
Dependencies: None
Introduced: 8.5.010

Specifies the number of days to retain data in the CTL_GDPR_HISTORY table.

The table stores the actual personally identifiable information (PII) that relates to a General Data Protection Regulation (GDPR) request. For Right to Access (export) requests, the table stores the data that was requested for export. For Right of Erasure ("forget me") requests, the table stores the data that was redacted.

To ensure that this data is ephemeral, the retention period for records in the CTL_GDPR_HISTORY table is restricted to a maximum of 30 days.

days-to-keep-gidb-facts

Section: gim-etl
Default Value: 14
Valid Values: Any positive integer
Changes Take Effect: At the next run of Job_MaintainGIM
Dependencies: None in releases 8.1.0 and 8.1.1; starting with release 8.1.2, days-to-keep-gim-facts
Related Options: days-to-keep-active-facts

Starting with release 8.1.1, specifies the number of days to retain fact data in GIDB. Facts that have a start time that is earlier than the retention period are eligible to be purged.

For multimedia interactions, this option specifies the retention period for completed facts, provided that there are no active facts that have the same (or earlier) start timestamps. If there are active facts with the same (or earlier) start timestamps, the eligible completed facts will not actually be purged until these active interactions have been terminated and, therefore, also become eligible to be purged, as described in days-to-keep-active-facts.

days-to-keep-active-facts

Section: gim-etl
Default Value: 30
Valid Values: Any positive integer
Changes Take Effect: At the next run of Job_TransformGIM or Job_MaintainGIM
Dependencies: None in releases 8.1.0 and 8.1.1; starting with release 8.1.2, days-to-keep-gim-facts
Modified: 8.1.1 (behavior and default value changed); 8.1.2 (dependencies introduced)

Starting with release 8.1.1, specifies the maximum number of days to retain active multimedia interactions in GIDB and the dimensional model, including certain Staging tables, after which the interactions become eligible for artificial termination.

more...

days-to-keep-gim-facts

Section: gim-etl
Default Value: 400
Valid Values: Any positive integer
Changes Take Effect: At the next run of Job_MaintainGIM
Dependencies: None

Starting with release 8.1.1, specifies the number of days to retain fact data in the dimensional model. Facts that have a start time that is earlier than the retention period are eligible to be purged. Job_MaintainGIM does not purge active fact data, dimension data, or aggregate tables (if aggregation is enabled).

Starting with release 8.1.2, Genesys Info Mart enforces days-to-keep-active-facts < days-to-keep-gim-facts and days-to-keep-gidb-facts < days-to-keep-gim-facts.

more...

run-maintain

Section: schedule
Default Value: true
Valid Values: true, false
Changes Take Effect: Immediately
Dependencies: None

Specifies whether to run Job_MaintainGIM at the scheduled time, as specified by the maintain-start-time option.

Genesys Info Mart Support for GDPR

This page describes product-specific aspects of Genesys Info Mart support for the European Union's General Data Protection Regulation (GDPR) in premise deployments. For general information about Genesys support for GDPR compliance, see General Data Protection Regulation.

Warning
Disclaimer: The information contained here is not considered final. This document will be updated with additional technical information.


Overview of Genesys Info Mart support for GDPR compliance

Starting with the initial 8.5.010 release, Genesys Info Mart processes input JSON files that customers provide to comply with Right to Access ("export") or Right of Erasure ("forget") requests from their customers (consumers). Starting with release 8.5.010.16, Genesys Info Mart also processes GDPR requests that customers provide from their employees, who are contact center agents and supervisors.

Genesys Info Mart uses the JSON files customers place in configured, tenant-specific locations (see Input and output file location). The JSON files identify the consumers or employees who have made GDPR requests to either "export" or "forget" their personally identifiable information (PII). The Info Mart database potentially stores requesting consumers' and employees' PII.

The daily Info Mart database maintenance job, Job_MaintainGIM, processes any "export" or "forget" JSON files that have been added or modified since the last processed JSON file. To execute the requests, the maintenance job uses a SQL script, gdpr.sql or gdpr_partitioned.sql, located in the sql_scripts folder in the installation directory.

  • For "export" requests, the applicable PII is reported in a database table, CTL_GDPR_HISTORY.
  • For "forget" requests, the data is redacted in Info Mart tables (primarily fact tables), and the PII that was searched for (in other words, the data prior to redaction) is reported in the CTL_GDPR_HISTORY table.

The PII is stored in the CTL_GDPR_HISTORY table for a configurable amount of time (maximum 30 days). You can query the CTL_GDPR_HISTORY table to obtain:

To query the CTL_GDPR_HISTORY table, you must have read-only privileges for the Info Mart schema.

Important
No special configuration is required on the Genesys Info Mart side to enable support for GDPR. Equally, no special actions are required as long as the maintenance job runs regularly. Genesys recommends that you do not disable the run-maintain option, which enables the maintenance job to run on an automatic schedule.

Data retention policies

A number of configuration options control how long data is retained in the Info Mart database. The following table summarizes the applicability of GDPR to different types of Info Mart data. For more information about the categories of data in the Info Mart database, see the "Genesys Info Mart Database Schema" page in the Genesys Info Mart Physical Data Model for your RDBMS.

Important
Genesys Info Mart support for GDPR compliance is based on default configuration settings and typical application usage. For example, given the recommended values of the days-to-keep-* options, fact data in GIDB tables and logs is short-lived (ephemeral) data that is automatically deleted within a short period of time, and Genesys Info Mart does not redact this data outside of the automated deletion.

Particularly in multimedia deployments with long-living interactions, Genesys recommends that you reduce the values of certain options that affect purge behavior. See Recommendations for purge-related options for multimedia for more information.
Configuration option Controls what type of data? Contains PII? Included in GDPR processing?
Never deleted—not configurable Configuration object data Yes, for employees Yes (starting with release 8.5.010.16)
gim-etl section
days-to-keep-gim-facts Fact data in the dimensional model Yes Yes.

Note: Dimension tables in the dimensional model are never purged. Dimensions store low-cardinality data that does not contain PII.

days-to-keep-active-facts Active multimedia interaction data in the dimensional model, GIDB, and Staging tables Yes Active interaction data in dimensional model fact tables is processed. Active interaction data in GIDB and Staging tables is short-lived and is not processed.
days-to-keep-gidb-facts Fact data in GIDB Yes No. See Important note, above.
days-to-keep-gdpr-history
(introduced to support GDPR)
The CTL_GDPR_HISTORY table Yes Not considered for GDPR reprocessing, because the data is short-lived (maximum 30 days).
days-to-keep-cfg-facts Deleted configuration fact data No consumer- or employee-related PII No
days-to-keep-discards-and-job-history Discard tables, and audit and history tables No No
gim-export section
days-to-keep-output-files Files generated by Job_ExportGIM to provide Info Mart data to customers who use Genesys Info Mart's Data Export feature Yes No. See Important note, above.

Note: Customers are responsible for implementing adequate processes to ensure that any PII in their imported data is handled in accordance with GDPR requirements, including using suitable retention periods or redacting data to comply with "forget" requests.

log4j section
max-backup-index Log files Yes No. See Important note, above.

Recommendations for purge-related options for multimedia

Particularly in multimedia deployments with long-living interactions, ensure that the days-to-keep-* options that control retention policies for short-lived data that is not included in GDPR processing provide a sufficient buffer for unexpected delays—for example, if a database purge was not executed because the maintenance job was interrupted. In deployments where the Info Mart database is partitioned, you must also factor in partition sizes, since a partition is not purged until all the data it contains is eligible to be purged.

Genesys recommends that you base option settings on the following calculations:

  • days-to-keep-active-facts = 30 – <buffer> – "partitioning-interval-size-gidb-mm"/24/3600
  • days-to-keep-gidb-facts = 30 – <buffer> – max("partitioning-interval-size-gidb", “partitioning-interval-size-gidb-mm”, “partitioning-interval-size-gidb-ocs”)/24/3600

Example

If the partition size for multimedia interactions in GIDB has been increased to the nondefault value of one week (partitioning-interval-size-gidb-mm=604800), the partition size for Outbound Contact–related data in GIDB has been increased to the nondefault value of ten days (partitioning-interval-size-gidb-ocs=864000), and you want to allow a buffer of three days for unexpected delays, set:

  • days-to-keep-active-facts = 30 – 3 – 604800/24/3600 = 20
  • days-to-keep-gidb-facts = 30 – 3 – 864000/24/3600 = 17

Right to Access ("export") requests

Genesys Info Mart uses the GDPR input files named export-<DDMMYYYY>-<any optional content>.json as the input for "export" processing. See Input and output files for full details about the JSON file requirements. The PII that Genesys Info Mart will report is specified in the input JSON files in:

  • The phone and email attributes that identify the requesting consumer
  • The username attribute that identifies the requesting employee
  • Custom user data KVPs and custom Outbound Contact Server (OCS) fields customers might specify in the "gim-attached-data" element

While custom KVPs and fields are included in the GDPR output, Genesys Info Mart searches only on the phone or email address in order to find fact table records associated with the requesting consumers. Similarly, while employee ID, name, and other attributes in the configuration record (for example, email address) are included in the GDPR output, Genesys Info Mart searches only on the username in order to find configuration object and fact table records associated with the requesting employees. For details about the specific tables and fields that are searched, see the description of the CTL_GDPR_HISTORY table in the Genesys Info Mart Physical Data Model for your RDBMS.

In the initial implementation, Genesys Info Mart does not provide an output JSON file. Instead, the PII data is reported in the CTL_GDPR_HISTORY table.

Example: SQL query for "export"

The following is an example of SQL you can use to retrieve PII for export. The example returns distinct occurrences of consumer PII data for phone number 5551212. The same query specifying CONSUMER_ID = 'jsmith' will return distinct occurrences of employee PII data for the agent with username jsmith.

SELECT TENANT_KEY,
       FORGET, 
       CONSUMER_ID, 
       TABLE_NAME, 
       COLUMN_NAME, 
       KEY_VALUE 
FROM CTL_GDPR_HISTORY 
WHERE TENANT_KEY = <tenant>
AND FORGET = 0
AND CONSUMER_ID = '5551212' 
AND KEY_VALUE IS NOT NULL 
AND CREATED_TS BETWEEN <TS_1> AND <TS_2>
GROUP BY TENANT_KEY, FORGET, CONSUMER_ID, TABLE_NAME, COLUMN_NAME, KEY_VALUE 
ORDER BY TENANT_KEY, FORGET, CONSUMER_ID, TABLE_NAME, COLUMN_NAME, KEY_VALUE

Right of Erasure ("forget") requests

Genesys Info Mart uses the GDPR input files named forget-<DDMMYYYY>-<any optional content>.json as the input for "forget" processing. See Input and output files for full details about the JSON file requirements. Genesys Info Mart processing of "forget" requests parallels "export" requests, except that any PII that is found is redacted in the applicable Info Mart database tables, with the PII that was searched for (in other words, the data prior to redaction) recorded in the CTL_GDPR_HISTORY table.

As is the case for "export" requests, the PII that Genesys Info Mart will redact is specified in the input JSON files in:

  • The phone and email attributes that identify the requesting consumer
  • The username attribute that identifies the requesting employee
  • Custom user data KVPs and custom Outbound Contact Server (OCS) fields customers might specify in the "gim-attached-data" element
Warning
Usage of Genesys products requires processing of employees' Personal Data (user’s name, work phone number, and work email) for proper functioning of the Genesys solution. Without storing this Personal Data associated with an employee, Genesys Info Mart could stop performing its function. Thus, for current employees, the processing of their Personal Data is necessary for the purposes of the legitimate interests pursued by the customer. Further, the customer may be required to keep employee interaction records in order to meet other regulatory requirements. Based on the lawfulness of this processing and the design of Genesys products, Genesys does not recommend erasing Personal Data associated with an ongoing user.

Audit and tracking

The Genesys Info Mart log tracks the progress of GDPR processing.

Sample Genesys Info Mart log (click to enlarge)

In addition, there are two log events that report the status of GDPR request processing, by tenant: 55-20172 (GIM_ETL_GDPR_ERROR) and 55-31406 (GIM_ETL_GDPR_SUCCESS). Genesys recommends setting an alarm on message 55-20172, to alert you to problems that are preventing completion of GDPR processing. You can use log message 55-31406 to cancel the alarm.

The CTL_GDPR_HISTORY table as audit report

Genesys Info Mart does not provide an execution report in the form of an output file. Instead, the CTL_GDPR_HISTORY table provides details about the PII data associated with “export” or “forget” requests. The table also provides a detailed audit trail of all the fields that were interrogated to satisfy a particular GDPR request. NULL values indicate that the field was evaluated for a particular instance of PII and was found to be empty.

Consider the following example:

  • An input JSON file for a specific tenant indicates that the consumer with phone number 5551212 wishes to see PII data related to that phone number. The input JSON file also specifies custom user data keys, which, based on business practices in the customer's environment, might be associated with PII data.
  • To satisfy this request, Genesys Info Mart interrogates interactions to or from 5551212, as well as related facts, such as user data related to those interactions. CTL_GDPR_HISTORY would have a row for each table/column searched.
  • Say there were no interactions to or from 5551212. In this case, the row for each table searched for this phone number would have a NULL FACT_ID and a NULL KEY_VALUE.
  • Say there were interactions to or from 5551212. However, in an associated record, a field populated by one of the keys specified in the JSON file is empty. In this case, the row for that table and column would have a NULL KEY_VALUE.

Example: SQL query for audit

The following is an example of SQL you can use to see the PII data for phone number 5551212, along with a full audit trail of the search for this PII, showing both the presence and the absence of this instance of PII data in tables and columns included in the search.

SELECT * FROM CTL_GDPR_HISTORY 
WHERE TENANT_KEY = <tenant>
AND CONSUMER_ID = '5551212' 
AND CREATED_TS BETWEEN <TS_1> AND <TS_2>
ORDER BY TABLE_NAME, COLUMN_NAME, FACT_ID

Feedback

Comment on this article:

blog comments powered by Disqus
This page was last modified on August 15, 2018, at 11:35.