Washington DC Patch 3

The Washington DC Patch 3 release contains important problem fixes.

Washington DC Patch 3 was released on May 9, 2024.

• Build date: 05-06-2024_1747
• Build tag: glide-washingtondc-12-20-2023__patch3-04-25-2024

Important:

For more information about how to upgrade an instance, see ServiceNow Upgrades.

For more information about the release cycle, see the ServiceNow Release Cycle.

Note:

This ServiceNow AI Platform major family release is now available in ServiceNow's Regulated Market environments. For more information about services available in isolated environments, see KB0743854.

For a downloadable, sortable version of the fixed problems in this release, click here.

Overview

Washington DC Patch 3 includes 293 problem fixes in various categories. The chart below shows the top 10 problem categories included in this patch.

Figure 1. Top 10 problem categories

Security-related fixes

Washington DC Patch 3 includes fixes for security-related problems that affected certain ServiceNow® applications and the ServiceNow AI Platform®. We recommend that customers upgrade to this release for the most secure and up-to-date features. For more details on security problems fixed in Washington DC Patch 3, refer to KB1641423.

Changes in Washington DC Patch 3

Application Manager release notes

One-click experience to update dependent applications: Use the Update Dependencies button to view all available updates for dependent applications of a parent application.

Search results tab recommendations: A message is displayed when using the search bar if your application or plugin doesn't display under the currently selected tab. The message recommends selecting a different tab to locate your search result.

Available system properties

com.glide.cs.field.boolean.show.choice_picker: Enables (false) or disables (true) a Boolean option to requesters whenever a dynamic choice node only has a single value associated. When set to true, the dynamic choice node displays to requesters with only the single value selection option and no Boolean option.

Dynamic Choice user input control

This script enables requesters to see their previously selected choices if they choose to edit their choices later in the conversation. This script also displays secondary fields to requesters, such as an email address, gender marker, or both, to help them further differentiate their available choices. Displaying the secondary field to requesters can be especially helpful when dealing with similar values, such as selecting the correct employee name when working with the same or similar names (that is, Christopher Smith or Chris Smith).

When selecting multiple choices, you can search within the Search for options field for filtering search results. If you input the script to enable secondary fields found in this topic, the search filter results apply to the secondary fields, too. Select or enter all choices that apply, and then select the send icon

Digital signature for PDF documents using CAC or PIV smart cards

Review the com.snc.pdfsigning.validation.method property.

Property value for digital signature using CAC or PIV smart cards

Specify the method that will be used to identify and validate the user who has signed the PDF using the com.snc.pdfsigning.validation.method property.

Installing applications in Application Manager

Three substeps are new to WP3: 1. To update dependencies if available, select Update Dependencies. 2. Select the dependency applications and versions to update and select Proceed 3. Review the installation details and select Install.

Service Portal release notes

Additionally, field-level read ACLs are enforced for filter conditions in Simple List widget instances by default. A new system property, glide.service_portal.enable_acls_for_encoded_query_in_list, enforces these ACLs regardless of whether the Enforce field-level Read ACLs on Filter query terms option is selected for Simple List widget instances. To use the Enforce field-level Read ACLs on Filter query terms option, change the value of glide.service_portal.enable_acls_for_encoded_query_in_list to false. For more information, see Simple List widget.

Service Portal properties

Option to always enforce read ACLs for filter conditions specified in Simple List widget instances. If false, the Enforce field-level Read ACLs on Filter query terms option in Simple List widget instances determines whether to enforce the ACLs.

Simple List widget

When the glide.service_portal.enable_acls_for_encoded_query_in_list system property is set to true, read ACLs are enforced for Simple List filter conditions always.

Virtual Agent release notes

The com.glide.cs.field.boolean.show.choice_picker system property determines whether a Boolean node or a Dynamic Choice node with a single value displays to users whenever the Dynamic Choice node only has one single value associated.

Update labels in the Virtual Agent Designer Properties tab so that both LLM and NLU topics have the same Topic description field label.

Notable fixes

The following problems and their fixes are ordered by potential impact to customers, starting with the most significant fixes.

Problem	Short description	Description	Steps to reproduce
REST API Explorer PRB1744904	A user with the rest_api_explorer role isn't able to access REST API Explorer	Users with the rest_api_explorer role aren't able to access the REST API Explorer and they need the web_service_admin role to access.	Log in to the instance. Impersonate as a user who only has the rest_api_explorer role and not the web_service_admin role. The fields are unselectable because of the access control.
Session Management PRB1730421	org.apache.coyote. http11.upgrade. UpgradeProcessorInternal consumes high memory and causes memory contention on the node	One node has more frequent out of memory or semaphore exhaustion than any other nodes. Heap dump displays a large object: 'org.apache.coyote. http11.Http11 Nio2Protocol'.
Condition Builder PRB1721656	When upgrading from Tokyo to Vancouver, the Add custom conditions button in the UI Builder 'Data visualization' component 'Add data source page' isn't working properly	A set condition doesn't display and the Add custom conditions button disappears.	Upgrade to Vancouver. Open the Agent Workspace in UI Builder. Add the 'Data visualization' component. Click Add data source. Select a source. In the filter area, try to click the Add custom conditions button.
CMDB Data Manager PRB1717464	The condition filter count, 'records match condition' link, and 'Preview' list in CMDB Data Manager may use an incorrect retirement definition	The 'Policy' filter query ignores the enabled retirement definition.
Database Persistence PRB1749923	The table join query fails to execute due to an incorrect alias name in SQL

All other fixes

Problem	Short description	Description	Steps to reproduce
Access Control List (ACL) Rules PRB1713026	tableLevelACLAllow throws an error for few tables on an instance		Navigate to CMDB Workspace. Try to enter anything on the intelligent search. Sometimes there's an internal server error.
Access Control PRB1703976	Users are unable to add a user to a group after elevating them to a role when the system property 'glide.ui.schedule_ slushbucket_save_ for_group_roles' is set to true	When the system property 'glide.ui.schedule_ slushbucket_save_for _group_roles' is set to true, a user can't add users to groups if it's required that the user elevate to a role. The background job that's created doesn't add the user to the group.
Advanced Work Assignment PRB1745788	isLiveAgentAvailable isn't working for domain-separated instances when the requester and the agent are in different domains	The issue occurs intermittently when a user is trying to connect to a live agent through Virtual Agent. They're receiving the message 'no agents available at the moment, please try again later', even though agents are available.	Activate Agent Chat and Domain Separation plugins. In the 'Chat' queue, add an agent from Domain A from a group also in Domain A. Make the agent available. A requester in Domain B initiates the live agent request, but it returns 'no agent available'.
AI Search for Service Portal PRB1753189	Sort options in the AI search results page aren't displaying after an upgrade	Sort options aren't displayed after an upgrade. On clicking the 'Most relevant' menu, nothing is happening and users can't choose the other sorting options.	Log in to the 'NL UAT Now Create' portal. Click the Search button under 'Search for an asset'. Under the 'Asset/Success Pack' navigation tab, find the 'Most relevant' sort option.
AI Search Glide PRB1594303	Incremental updates aren't working for dot-walked reference fields		Set up an indexed source with a dot-walk field. Add a few PRBs to incident records, with PRBs that don't have category=hardware. Create a search source filter for problem.category=hardware. Update the PRB from another category to hardware. Search for an incident. Expected behavior: The record matches a filter and is returned for the search. Actual behavior: AI Search doesn't return an incident.
AI Search PRB1706402	AI Search doesn't index-translated text for 'Reference' type columns	Currently, the AI Search index reference column value only has one of the languages, even though translations in multiple languages are available.
AI Search PRB1738159	On Service Portal, users aren't able to see search results related to NowAssist QnA	Results related to QnA and NowAssisted QnA should be returned in the results on Service Portal and global search.	Navigate to an instance. Navigate to Service Portal. Search for the term 'What is spam?'. No results related to QnA or NowAssisted QnA are returned. Repeat the steps on global search. No results related to QnA or NowAssisted QnA are returned.
AI Search PRB1739538	Currently, there's no option to toggle the display of the List view button on global search results		Migrate global search to use AI search/use an instance where the migration is done. Perform a search. Click any navigation tabs. Notice that there's no dev configuration way to hide to the List view button. The sn-search-result-wrapper should have a configuration to hide the List view button from the display.
AI Search PRB1739541	Currently, there's no option to toggle the display of the Back to results button on global search results		Migrate global search to use AI Search/use an instance where the migration is done. Perform a search. Click any search results. Verify if the back button is displaying. Expected behavior: The sn-search-result-wrapper should have a way to hide the button from the display. Actual behavior: There's no way to hide to the Back to results button.
AI Search PRB1741947	A few AI Search (AIS) tables don't let users create, update, and delete access for other scopes	Some AIS tables should allow for CRUD access for other scopes.	On the table record, under 'Application access', the fields Can read, Can create, Can update, and Can delete should be enabled.
AI Search PRB1748854	No results for any search terms are returned on global, search portal, and ESC	Results are seen from the search preview, but not the portal. There's no log errors.
AI Search PRB1751394	Some queries on attachment documents result in an error	There's an empty page and an error: 'Cannot invoke "String.isEmpty()" because the return value of "com.glide.search.core.models. AISearchResult.getSearchTerm()" is null...'	Open an instance. Confirm that there's Now Assist for Search version 4.0.3. Add dpr config field mappings on a KB index source. Attach a document to the KB. Confirm that index_attachments and index_attachments_with_dpr attributes are set to true. Search for the term 'research' on the portal.
AI Search PRB1751700	The 'Order' value for genius results should be returned	The 'Order' value should be returned for every genius result, both async and sync.
AI Search PRB1752918	'geniusResultsMode' should be added to requests	This is a product update.
AI Search PRB1752919	There should be split steps for running a genius result through Entity View Action Mapper (EVAM) into a separate composite data broker	This is a product update.
AI Search PRB1752920	A pass-search string should be added in a clickthrough payload for both Genius and regular results	This is a product update.
AI Search PRB1752921	Pass details are needed for post-processing on context objects to genius result scripts	This is a product update.
AI Search PRB1752922	The Asynchronous Message Bus (AMB) channel name should be added to the GQL/data broker response	This is a product update.
AI Search PRB1752923	Result improvement rules should be able to be shared across search profiles	This is a product update.
AI Search PRB1752924	A scriptable should be created to generically handle post-processing	This is a product update.
AI Search PRB1752925	Pending GRs should be added to the data broker response	This is a product update.
AI Search PRB1752926	New GENIUS_RESULT _TRIGGERED signal events and logger should be created	This is a product update.
AI Search PRB1758066	An indexed source attribute 'index_translated_reference_fields' is missing		Create an instance from the latest track/aisearchsandbox. Enable AI Search. Navigate to any index source. Try adding the advance configuration attribute 'index_translated_reference_fields' to turn off the default behavior. The attribute isn't found in the list.
AI Search PRB1763179	The new parameter 'geniusResultMode' is added to the ScriptableSearchAPI in the incorrect position due to a merge conflict resolution
Application Portfolio Management PRB1715173	Application Portfolio Management (APM) Performance Analytics issue	When a user refreshes impacted services, the impacted business applications are calculated based on direct CI relationships for 'Incident' and 'Change', but not for 'Problem'.
Appsee - Platform Infrastructure PRB1716290	An empty value in the Analytics consent policy field of the core_country table record causes an error	An empty value in the Analytics consent policy field of the core_country table record causes a 401 unauthorized error and breaks the navigation bar.	Create a country in the core_country table (or have an existing country) with an empty Analytics consent policy field. Confirm that in the Consent policy provider field, only 'GeoIP' is enabled and the rest are turned off. Log in from a location for which the consent policy is set as empty. On log in, ux-metrics flow triggers and executes the 'Get consent' API. Expected behavior: 'Get consent' should return the expected consent policy to ux-metrics. Actual behavior: A null pointer error is thrown and it's not handled to return the correct response to ux-metrics.
Archive Rules PRB1669506	The 'Archive Destroy' rule isn't working and displays 0 records	The 'Archive Destroy' rule isn't working and displays 0 records, and doesn't delete records from the 'AR' table.
Asynchronous Message Bus (AMB) PRB1737142	Asynchronous Message Bus (AMB) Logger logs an entire message when it fails to deserialize, causing an out of memory error	When there's a large message that's not deserializable, it throws an error that is logged by AMB. It causes an out of memory error. These messages are coming through ActivityEventsAMBProcessor, triggered by an update on a change request. The message doesn't seem to have the correct JSON structure.
Automated Test Framework (ATF) PRB1717536	Setting a value in the select2 element fails on a Service Portal page	The select menu 'How satisfied were you with the response time to your incident?' remains in a '-- Choose --' state instead of being set.
Automated Test Framework (ATF) PRB1749925	Automated Test Framework (ATF) server-side steps don't display transaction details when compared in a performance comparison		Have an instance configured for ATF performance testing. Navigate to Automated Test Framework (ATF) > Tests > New. Enter any name. Click Save. Click Add Test Step. Navigate to Server > Record Insert > Next. For table, choose 'Incident'. Enter any values for 'Short description' and 'Caller'. Click Submit. Refresh the test page. Click Run Performance Test. Check the acknowledgment. Click Run Test. Wait for the run to complete. When it's finished, navigate back to the test form. Repeat step 11 to generate a second performance run. Navigate back to the test form. Open the 'Performance Test Results' related list. Check the two performance results created in the prior steps. Choose the 'Compare performance results' UI action from the 'Actions on selected rows...' dropdown menu. In the 'Performance Statistics' related list, select the 'Record Insert' statistic record. Expected behavior: There should be a record in the 'Aggregated Transactions' related list. Actual behavior: There are no records in the 'Aggregated Transactions' related list.
Case and Knowledge Management for HR Service Delivery PRB1752005	There's a cross-scope access privilege issue when checking for canRead for the sn_hr_core_case table		Navigate to an HR case table. Select a case record. Scroll down to the related lists for that case record. Select the tab labeled 'Attached knowledge'. Select the button New to create a KB. Once the page loads for a new KB record, verify if there's a cross-scope access error.
Case and Knowledge Management for HR Service Delivery PRB1752405	RCAs are generated for 'Apply for Parental Leave'		Activate Now Assist for HR Service Delivery and Journey Designer. Execute the topic 'Parental Leave of Absence Request'. RCAs are generated.
Condition Builder PRB1735699	There's multiple issues with the reference list condition builder	The following issues occur: When running a query, the default conditions are appended to the query (for example, running with 2 default conditions results in 4 conditions shown). 'Sort By' conditions don't persist (The user can add a sort to a column, but when they open the filter again, it's not there). The 'Name' column is sorted as 'Ascending' and can't be changed.	Navigate to Service Operations Workspace. Open an incident record. Open the 'Details' tab. Click the magnifying glass for a service reference input. Observe the following issues when running a query: The default conditions are appended to the query, the 'Sort By' conditions don't persist, and the 'Name' column is sorted in 'Ascending' order.
Condition Builder PRB1750793	There's an incorrect label in the 'is one of' comparison type	Users are facing this issue in the Now Assist Admin > ITSM > Incident summarization > Choose input screen.	Add a condition builder component in a UI Builder page. Select 'Incident' as the table name. In the 'Runtime' page, select 'state' as the field. Select 'is one of' as the comparison. Select any value. Check the label of the selected value in the input field.
Configuration Management Database (CMDB) PRB1702705	During performance testing, similar symptoms to PRB1369519 were noticed	During performance testing, similar symptoms to PRB1369519 were noticed, 'Contention and continuous updates on the cmdb_qb_result_base table'.
Core Platform PRB1594189	Decompiling JavaScript to pass compiled functions, as strings can lead to high memory churn	When a large script (over the interpreted threshold of 20000) that converts a function code into a string is invoked, the script is decompiled from its source. If that source is large, many large strings are created, which are then thrown away as garbage.
Core Platform PRB1704668	A node log is missing the SessionID and txID in the localhost log, causing an incorrect date format	At startup, if reverse DNS lookups take longer than 250 ms, orbit tmp-cleaner logs a message to the logs and uses the wrong formatter, because it's the only formatter it has access to while shading the Log.java class.	Set the 'glide.debug.trace_threshold' property to 1. Restart the node. Start the node.
Core Platform PRB1725401	Some fields in syslog_transaction aren't logged to the node log file	The PageTiming log messages contain the information, but aren't enabled by default.	Open a form. Modify a field value within that form. Select and hold (or right-click) the form menu bar. Save. Observe that the syslog_transaction table contains 'Client Response Time', 'Client Network Time', and 'Browser Time' for this event. Expected behavior: Node logs/splunk do contain 'Client Response Time', 'Client Network Time', and 'Browser Time' metrics for this event. Actual behavior: Node logs/splunk don't contain 'Client Response Time', 'Client Network Time', and 'Browser Time' metrics for this event.
Core Platform PRB1732249	Double escaping occurs and breaks a query when carets and forward slashes are both used	This is an issue that occurs for any query that includes both '/' and '^' characters. While a '^' in a query is escaped correctly, when combined with '/', double escaping occurs which then breaks the query. GlideRecord doesn't escape carets consistently.	Navigate to any instance. Navigate to any list. Create an 'Incident' record. Set a short description to '/test/!/^/test/!'. Navigate to the 'Incident' list. Group by the 'Short Description'. Clear filters. Click 'Show matching' on the record cells with the descriptions. Clear filters. 'Filter Out' on the record with cells with descriptions. Expected behavior: The list should remove all the records with the description '/test/!/^/test/!'. Actual behavior: Records with description '/test/!/^/test/!' aren't removed.
Core Platform PRB1755783	A high response time is observed for the Data Push Rest API in the 'schema' table	For various payload sizes, the Data Push API has seen an incremental increase with the response times, with an increase in payload sizes going from ~5 seconds for 50 KB to ~123 seconds for 1000 KB.	Execute the Data Push Rest API with a payload size of 50 KB. Navigate to the syslog_transaction_list.do. Filter with the URL to find the breakup of the transaction.
Database Persistence - Data Management PRB1732094	The 'Physical Table Stats Gatherer' job impacts the database load	A job called 'Physical Table Stats Gatherer' that hosts a large number of instances can impact the database load on the hosted database.
Database Persistence - Data Scale PRB1701011	The prepared statement skips if the input predicate size is >1000
Database Persistence - Data Scale PRB1734637	A newly added node on an instance has '{}' added to the glide_build_tag and some of the node has a null/empty glide_build_tag	This can cause timeout errors.
Database Persistence - Data Scale PRB1737524	Execution time per SQL statement should be captured	Having the SQL execution time at the aggregate-level isn't good enough to compare and assess the performance of individual queries.
Database Persistence PRB1710202	There should be a JDBC driver that can work with MySQL 8 using the sha256_password/ caching_sha2 _password authentication plugin	Replacing the built-in driver with a driver from MySQL and renaming it threw an error.
Database Persistence PRB1749083	On Tembo migration, automatic activation of the 'Low connection' mode should be turned off	As far nodes start up, they come up by default in 'Low connection' mode, putting the max connection pool size to 4. This connection pool size isn't large enough for a node to bootstrap and to report the status as online.	Have a far node with a large number of ACL rules, like 100 K. Restart the far node in low connection mode. Observe the 'connection busy' message in the logs and see the XMLStats transaction take a long time to complete.
Declarative Actions PRB1695335	The Configuration Item field filter condition doesn't get applied to the pop-up modal	This occurs in Service Operations Workspace when the user opens any incident and navigates to the Configuration Item field. When the user types a search, for example, 'SAP' and clicks the magnifying glass, the conditions aren't applied to the pop-up modal.	Navigate to any base instance. Navigate to Service Operations Workspace. Open any incident. Navigate to the Configuration Item field. Type 'SAP'. Notice that the list is filtered. Click the magnifying glass. Notice that the results in the modal aren't filtered by those characters.
Discovery PRB1728253	There's high memory consumption by Discovery when 'Orchestrator' pages are split	When this issue is reproduced, failing to cancel Discovery prior to its completion results in severe memory shortage and instance nodes crashing. The instance also becomes unresponsive for ~30 minutes. Stats.do displays several workers busy with 'ASYNC: Discovery - MultiPage Sensors Pat' jobs.
Discovery PRB1746323	The 'Shazzam' batch size is hard-coded to 5 K instead of 1 K	The batch size for an initial 'Shazzam' scan is correctly set to 1,000 IPs. However, when the sensor calculates the batch size for the second 'Shazzam' scan, instead of reading the default value from the properties table, it employs a hard-coded value of 5,000 IPs and initiates the second IP range from 5,001. Since the total IP range consists of only 3,000 IPs, the second 'Shazzam' scan won't be launched.	Run a quick discovery on a large IP range crossing 1-K IPs. 'Shazzam' is launched only on the first batch of 1-K IPs.
Document Intelligence Unified Backend PRB1732364	A task failed under the copy of an imported uC, which isn't upgrading to the latest version	A message displays that the task processing failed.
Document Management Services PRB1742666	In GovCommunityCloud (GCC), users are getting an error for the Personal Identity Verification (PIV)/Common Access Card (CAC) signing feature when clicking 'Validate/Submit' on ESC Portal	Users are unable to use CAC/PIV cards for signing since the signing certificates aren't registered.
Encryption Support PRB1727700	GlideEncrypter API's deprecation process (instance scan) only searches 'script' type fields	The 'Deprecated API: GlideEncrypter' check-in instance scan isn't comprehensive, as it only searches fields of the 'script' type. GlideEncrypter can be present in other areas, such as (non-script type) fields and variables.
Event Management PRB1714485	Navigating to the Service Dashboard by clicking from the Service Operations Workspace home takes 26+ seconds on a large synthetic data set	Service Dashboard loads slowly on a large synthetic data set with 15,000 application services on the instance. It takes 26+ seconds on both the first and repeat view.	Log in with the evt_mgmt_admin role. Click the 'Service Dashboard' icon.
Event Management PRB1750681	A script include wrapper should be included for ServiceAnalyticsProcessor	A script include wrapper for ServiceAnalyticsProcessor should be included in the global scope to access the API from any scoped application. The script include should be open for all applications.
Event Management PRB1753924	Most of the tables are missing in 'Reference table' in em_mapping_rule		As a logged-in user, change the scope to some application scope. Create a em_mapping_rule record. Select the mapping type 'Map field from reference table'. Only a few tables are listed in the Reference table field.
Event Management PRB1758320	UTF-8 characters aren't supported in System Center Operations Manager (SCOM) Client	SCOM Client doesn't support special characters that are encoded in UTF-8.	Fetch events with special characters in any field from the SCOM Pull Connector. Navigate to 'All Events'. Special characters are replaced by '?'.
Fenix (Family) PRB1751391	The business rule on Glide credentials (discovery_credentials) triggers a force restart of the Fenix job via deployments (fenix_manifest)		Have a discovery_credential updated and referred from a Fenix deployment. The Fenix job with the updated Glide credential used in the program is restarted.
Flow Designer (Family Channel) PRB1723249	The plugin 'com.glide.pad.core.runtime' causes excessive 'Process Automation Event Handler' jobs to be recreated	On larger instances with high node counts, restarting several nodes when upgrading can result in a large number of 'Process Automation Event Handler' jobs being deleted and created, sometimes in the thousands.	Navigate to a 2 node instance. Open the sys_trigger table. Locate those with name='Process Automation Event Handler'. Customize to show System ID, Created, and Updated fields. Sort by Created. Verify that there are 2 with the System ID 'ACTIVE NODES', and they end in 8080 and 8081. Make note of the 'Created times'. Restart one node. Refresh the list. Expected behavior: Only the records connected with the restarted node should be removed. After a short period, all nodes should be readded. Actual behavior: All non-active nodes are removed.
Flow Designer (Family Channel) PRB1732445	In Process Automation Designer (PAD), there's an infinite loop situation when SourceAlias and ParentSourceAlias are the same	If any of the PAD lane name sand PAD activity names are the same, PAD generates the same value in sourceAlias and parentSourceAlias fields of a compiled flow in a table 'sys_flow_compiled_flow_chunk'. An infinite loop situation is hit and the instance goes down. This only happens when PAD is running with Flow Engine V2.	Create a PAD process with a lane name that matches an activity name. Navigate to the sys_pd_activity table to see the names of the activities. Use the same name for the lane. Click the Test button. The instance is restarted.
Flow Designer (Family Channel) PRB1750937	Flows become read only when feature access is enabled	A user must click 'Edit Subflow/Flow' when first opening a flow. However, when feature access is enabled, the Edit Subflow/Flow button isn't rendered, so the user can never make changes.
Flow Designer (Family Channel) PRB1755683 KB1640735	There's unexpected triggering of 'Run Once' flows due to the data cleaner job in 'sys_flow_plan_context_binding'	The data cleaner job is deleting records older than a year from the 'sys_flow_plan_context_binding' table. This is causing 'Run Once' flows to be triggered again, which is an unexpected behavior.	Refer to the listed KB article for details.
Flow Designer (Family Channel) PRB1757401	Users are unable to get the flow preview when testing with a Flow Designer role	There's an empty screen. 'Open Flow' does nothing.	Log in to an instance with a flow_designer role. Navigate to an existing flow. Toggle to 'Diagram' view. There's an empty screen and the diagram doesn't load.
Flow Designer (Family Channel) PRB1758376	If a flow is in preview mode, it shouldn't display in pickup where a user left off or the recent activities		Create a flow from Now Assist. A preview page opens. Change the prompt. Rebuild the flow. Navigate to the 'Workflow Studio' home page. The recent activities user can see the preview flow details.
Flow Designer (Family Channel) PRB1760148	A flow preview doesn't load with delegated developer due to flow ACLs	A generated flow of an example doesn't make sense.	Create a delegated developer with flow_designer. Switch to the user who has delegated developer. Attempt to create a flow preview Expected behavior: The flow preview renders correctly Actual behavior: The flow preview doesn't load correctly and has some missing nodes.
Flow Designer (Family Channel) PRB1761976	Aligning Admin Console content with branding standards	This is a product update.
Flow Designer (Family Channel) PRB1761977	Add a new status preview to the 'Flow' table to indicate that the flow is in preview mode and isn't committed	This is a product update.
Flow Engine PRB1706393	A flow priority isn't propagated to subflows	This behavior is the same regardless of the following settings of the subflow: 'Wait Condition is Checked', 'Wait condition is unchecked', and 'Dynamic Flows'.	Create a flow that calls a subflow. Set the priority of the flow to 'LOW' in the flow's settings. Trigger the flow. Expected behavior: Check the flow context details of the flow. It should be set to 'P7', which is low. Actual behavior: Check the flow context details of the subflow. It's set to P5 'Medium' instead of 'P7'.
Flow Engine PRB1752916	Enable quickAPI for async http requests for a Virtual Agent perf issue	This is a product update.
Flow Engine PRB1757176	Accessing subflow context fields via an inline script doesn't work	An error encountered when processing a script: 'Invalid reference to: fd_data._1__s1_ with_3_outputs.context...'	Create a flow that calls a subflow. In the calling flow, add a log action. In the log message input, add an inline script.
Flow Engine PRB1757198	'Retry policy' without an interval results in an error	The flow ends in the state of 'ERROR'.
Form Controller PRB1731381	In an issue record page, 'Activity' and 'Setting' sections display as empty		Activate the following scoped apps: irm-shared-commom-components, app-grc, app-risk, app-risk-advanced, app-grc-base-workspace, and app-risk-workspace. Navigate to any workspace. Navigate to List > Issues > Create new. Create a new one. In 'Details', the 'Activity' and 'Settings' sections display as empty.
GlideRecord PRB1755685	'Chat Summary' doesn't display the updated message as 'Now Assist (AI) has modified this field value'	The message isn't displayed correctly for summary and the sparkle icon is missing for 'Short Description' and 'Chat Summary'.	Start an interaction. Exchange over 7 messages between the agent and user. As an agent, click End chat > Short description. Chat summary fields are populated.
Health Log Analytics (Family) PRB1753824	Health Log Analytics (HLA) business rules (BR) are being processed by the Discovery BR	Three HLA BRs (Process data input connector port check, process data input preprocessor example, and update data input markers) are processed by Discovery BRs.	Navigate to /ecc_queue_list.do. Filter by one of the HLA BRs that are impacted. Open the BR debugger. Select one of the 'error' records. Rerun the logic. Observe that the 'Discovery - Sensors' rule isn't skipped.
Hermes (Family) PRB1748050	Hermes is installed but not configured	The Hermes endpoints are available but the callback doesn't get called.	On a clean instance, request the IDR plugin install. Locate the 'Enable Hermes Service' change that's triggered. Locate the workflow context. Find the log entries associated with 'Setup Hermes Service Endpoints'. It's expected that 'hermes' is the service for those endpoints. The Hermes cluster configuration table isn't configured, as that activity returns 'hermes-internal'.
Hermes (Family) PRB1751021	IDRHermesConsumerJob intermittently doesn't terminate		Turn off one of the sysauto_script IDRHermesConsumerJobs before doing the test. Create a Hermes bidirectional replication ret. Activate a producer and consumer replication set. Replicate a large number of records both ways. While messages are still producing, delete both producer and consumer replication sets. As this happens, notice that the IDRHermesConsumerJob gets stuck.
Hermes (Family) PRB1755317	Topic Inspector fails to load consumer group info when 'subscribe by pattern' is used in consumer client		Create 2 topics on the same instance. When users subscribe by pattern, there's 2 different topic names in the same group description.
Instance Data Replication (IDR) PRB1736093	The Track in Update Set button doesn't add transformations to the update set	The Track in Update Set button doesn't work as the user may expect. It adds IDR-related tables to the update set (idr_entry_set, idr_consumer_replication_entry), but doesn't include the transformations.
Instance Data Replication (IDR) PRB1745789	Instance Data Replication (IDR) Record Transformer must be clear when it's in a draft state	When users modify a consumer replication entry, it isn't clear that users must click 'Update' or 'Save'. It should be clear that the transformations are in a draft state, because users think they have errors with their replication.	Create a replication set. On a consumer replication entry, add or modify a transformation, but don't save it. It should be clear that it's in draft state and doesn't apply to the current replication set.
Instance Data Replication (IDR) PRB1745827	There's an IDR conditional vault check when 'VAULT' isn't available	It errors, indicating that the vault is no properly configured, even though the vault isn't available in the environment.
Instance Data Replication (IDR) PRB1753899	When adding a transformation mapping, a desired message doesn't display unless refreshed	When users remove a transformation mapping, there's a refresh that happens. That refresh is required for the desired info message regarding the state of transformers.	Add a transformation mapping to an entry that has its transformation enabled. Expected behavior: After a refresh, there should be a yellow info message indicating the transformation is still being edited. Actual behavior: The info message doesn't appear.
Integration Hub PRB1710047	A REST step is appending charset=UTF-8 to the content type	The issue is rising due to the content type sent in the REST step, where charset=UTF-8 is appended to the content type, like 'Content-Type: application/ json;charset=UTF-8;'.	Create an action with a REST step. Select the HTTP method as 'POST'. Select the request type as 'Multipart' The content type is sent as 'application/json;charset=UTF-8'. As a result, some third-party servers are erroring out when the passing content type is set as 'application/json; charset=UTF-8'.
JVM at Scale PRB1736766	Memory watcher doesn't log most pauses from ParallelOld because they are under 2 seconds	The pause duration for ParallelOld is typically shorter than the garbage collector, but may happen more frequently. Memory watcher should be more responsive to the shorter pauses of ParallelOld.	Review the logs.
Key Management Framework (KMF) PRB1754276	KMFKeySecureAPI may be initialized in an improper immutable state	In KMFKeySecureAPI, fKMFEnv and fPropertyRepo are static final. If they're improperly initialized, then KMFKeySecureAPI is initialized using an improper state from which it never recovers.
Knowledge Management PRB1740935	A Customer Service Management (CSM) consumer is unable to comment on knowledge articles from the '$knowledge.do' page	The issue occurs even if the consumer has read access.	Access any Vancouver and new release instance. Activate the customer service demo data plugin, which activates the necessary plugins along with the demo data to reproduce the issue. Impersonate a CSM user. Navigate to '$knowledge.do'. Open any knowledge article. Attempt to leave a comment. Currently, this feature isn't working and no comment is attached to the article.
Knowledge Management PRB1742640	The Source task field should be populated when a KB is generated	Users expect the Source task field to be filled with the HR case task they were previously on.	Navigate to an HR case task table. Select a case task record. Scroll down to the related lists for that case task record. Select the tab labeled 'Attached knowledge'. Select the button New to create a KB. Once the page loads for a new KB record, observe that the field Source task is empty.
Knowledge Management PRB1747293	UI defects are in the Knowledge Base general article modal	The heading isn't displaying the proper font. It should already be part of the modal component. The heading font color should be 'Primary', the same color as the Primary button. In the body text, replace all circled words with 'incident'.	Log in to an instance. In UI16, navigate to the 'Incidents' table. Open an incident that isn't resolved or closed. Resolve the incident. Click the Create Knowledge button. The 'Use AI to draft this article?' modal appears. UI issues display.
Knowledge Management PRB1753175	There's warnings that a variable has been passed to getMessage()	There's L10n warnings under the 'Merge verify' report.
Knowledge Management PRB1755358	Setting the sys property 'sn_customerservice.kcs.enable _template_on_case_workspace' as false isn't triggering the GenAI flow	For UI16, the 'Create Knowledge' UI action doesn't trigger the article draft modal when the Customer Service Management (CSM) Knowledge Base (KB) generation skill is active. For workspace, 'Create Knowledge' UI actions doesn't trigger the article draft modal, when CSM KB generation skill is active and the sys property 'sn_customerservice.kcs.enable _template_on_case_workspace' is false.	Activate the Customer Service Management demo data plugin and the GenAI CSM plugin in an instance. As an admin, navigate to Now Assist Admin Console > Feature > Customer. Click Knowledge > Activate KB generation skill. Choose any close case. Create a knowledge article. Expected behavior: It should trigger a GenAI flow. Actual behavior: Setting the sys property 'sn_customerservice.kcs.enable _template_on_case_workspace' as false isn't triggering the GenAI flow and doesn't display the modal for the user to choose GenAI or draft on their own. It prepopulates the draft.
Knowledge Management PRB1758503	True-up knowledge apps for Washington Patch 3	True-up knowledge apps app-kb-uib and app-knowledge-gen-ai.
Knowledge Management PRB1759061	Knowledge Management instrumentation changes are missing
Knowledge Management PRB1760623	Fix dependencies for Now Assist in Knowledge Management	'Now Assist for Knowledge Management' must be installed as part of 'Now Assist for Platform'.
List Administration PRB1735472	A related list doesn't function when the 'Applies To' table in the relationship record is in a database view	Related lists don't function as expected if the underlying sys_relationship record uses a database view for the 'Applies To' table. Pagination doesn't work and filters become stuck.	Create a database view. Add a view table where the table is: Requested Item [sc_req_item]. Provide a variable prefix. Submit the record. Add another view table where table: Variable Ownership[sc_item_option_mtom], Prefix: var, and Clause: req.sys_id=var.request_item. Submit the record. Create a sys_relationship record. Provide a name. Make the 'Applies To' table the database view created in step 1. For 'Queries from table', select the 'Incident' table. Submit the record. In the filter navigator, type 'YOUR_DB_VIEW_NAME_HERE.list'. Click the 'i' icon next to a record to open. Select and hold (or right-click) the form header. Choose: Configure > Related Lists. Add the relationship created in step 7. Save. Back at the form, try to page through the related list. It doesn't paginate. Click the 'magnifying glass' icon so the search boxes appear under the column headers. In the short description box, enter a search string. Execute the search. Try to dismiss the filter by clicking the chevron before the condition. The filter isn't removed.
MetricBase PRB1753796	Trigger flows with a filter condition and an auto-promoted metric throw an error	Users can identify this error in the MetricBase trigger logs by looking for a stack trace that mentions the 'Preconditions' class.	Create a table hierarchy in Glide and have the metrics associated with those tables have 'auto-promote' set to true Create trigger definitions and flows for the auto-promoted metrics. Make sure that the trigger flows have the condition field set to something. Confirm that the trigger configuration is synced. Start posting lots of data that would cause trigger transitions. On Glide, navigate to metric base trigger logs. Look for the preconditions stack trace.
MID Server PRB1732428	MID disconnects agents on a null pointer exception
MID Server PRB1743685	There's a failure during key exchange, as SNCSSH doesn't support ED25519		Configure the SSH server to use only ED25519 for host key verification. Run Discovery/test credential from an instance to this SSH server. An error displays: 'Cannot connect, status is SSH_CONNECTION_FAILURE. Could not agree on signature algorithm Client:...'
Mobile Platform PRB1728314 KB1633400	In the 'Questionnaires' survey, the Boolean type variable isn't working in Mobile	The issue isn't reproducible on desktop.	Refer to the listed KB article for details.
Mobile Platform PRB1743040	Legacy lists aren't displaying the update indicator (the 'Cloud' icon) in offline mode when a record is updated	Legacy lists aren't displaying the update indicator (the 'Cloud' icon) in offline mode when a record is updated if the data item is declarative with a parameter unresolved. A document with a parameterized data item (the parameter not substituted) is sent to the client without any modification. This would miss the update if some record has been modified.	Access the Agent Mobile application. Switch to offline mode after downloading the cache. Update a record in a legacy list. There's no indication on a record that it has been updated.
Next Experience Unified Navigation PRB1743909 KB1635281	There's a trailing space in sys_app_module records that cause a 500 error on Next Experience	This isn't specific to Next Experience and also affects the UI16 menu API.	Refer to the listed KB article for details.
Next Experience Unified Navigation PRB1747769	Clicking Next Experience menus ('All', 'Favorites', 'History') causes a CPU spike and locks the browser	When some users are interacting with the Next Experience menus, such as 'All', 'Favorite' and 'History', the user experience has lag and slowness. The CPU spikes to over 100% usage. This only occurs when doing a navigation command in the menu.	Set up a large number of modules. Open the 'All' menu. Enter 'incident.list'. Click Enter. Try to open the 'All' menu. Expected behavior: The menu opens with no lag. Actual behavior: There's a delay when opening the menu.
Notification Preferences PRB1701164	The 'Notification Preferences' link on an email opens two tabs on the browser		Navigate to System Logs > Emails. Open an existing email. Click the 'Preview Email' related link. Click Notification Preferences. Expected behavior: The new tab opens, linking to the notification preferences. Actual behavior: Two new tabs open, one with the home page and the other with the notification preferences.
Notification Preferences PRB1757886	Issue when trying to send a push notification to users whose Firebase token has expired	Android push notifications for Mobile Publishing are impacted. If there's at least one inactive token in the payload, the whole notification would fail.
Now Assist Panel PRB1756599	Dynamic choice control theming with multi-select isn't consistent with NAP	There's errors on submission.	Create a dynamic choice control topic with the multi-select check box enabled. Make it available in NAP. Run the topic in NAP. Expected behavior: The theming is consistent with NAP theming. Actual behavior: The theming is the same as on Web Client.
Performance Analytics PRB1732618	Users should be able to view saved data visualizations from UIB pages/technical board without sharing the data	Saved data visualizations should be visible to the end users with the correct ACLs without the need of sharing.
Performance Analytics PRB1746162	A list filter isn't migrated in a BT1 clone		Navigate to a BT1 clone. Trigger a bulk migration. Navigate to the migrated Next Experience dashboard 'Out of Band Release Dashboard'. In the first tab, scroll to the list filter 'Release Channel - OOB'. Expected behavior: The choice filter should be migrated. Actual behavior: The choice filter isn't migrated and displays as an unsupported filter.
Performance Analytics PRB1749825	Dashboard-level group/role/user permissions aren't migrated during bulk migration		Create a dashboard. Create a dashboard group. In the group, give the user Abel Tutor access. Do a bulk migration. Check permissions for the dashboard. There's no permission record for Abel Tutor.
Performance Analytics PRB1750501	A layout isn't retained in a dashboard copy after duplication	The issue occurs if a dashboard is in the scope. Admins can reproduce the issue.	Impersonate a user. Navigate to the IT Agent Dashboard. Duplicate the dashboard. Expected behavior: The dashboard is duplicated and all the widgets are present. Actual behavior: The dashboard is duplicated, but all the widgets aren't present.
Platform Analytics Migration Center PRB1722203	Users must check in the UI for the sys_trigger status to inform when activation is completed	Users should be kept on the Migration Center and have a loading information that an action is happening.	Run the bulk migration from Migration Center. Click Activate. The user is redirected to the 'Dashboard library' page but not all the actions on activation are performed.
Platform Analytics Migration Center PRB1742143	There should be a message in the Migration Center about performance degradation for compatibility mode	Dashboards with visualizations or other widgets migrated in compatibility mode may take slightly longer to load. In Migration Center, an info alert should display when viewing the compatibility mode content.	Navigate to the Migration Center. Confirm that there's a message about the performance degradation for compatibility mode dashboards.
Platform Analytics Migration Center PRB1743812	When activating Next Experience from the migration center, the 'Initializing all your content' state is never removed		Navigate to the migration center. Trigger a bulk migration on the dashboards. Activate Next Experience. A spinner appears: 'Initializing all your content'. Expected behavior: After this process is done, the state should automatically change to the button Take me to dashboard library. Actual behavior: It wouldn't move from this state until the page is manually refreshed.
Platform Analytics PRB1729810	There's a chart title issue in data visualizations	Updating the chart title doesn't trigger the Save button in the visual designer.	Create a data visualization. Save. Update the chart title. Don't make any other changes. Expected behavior: The Save button is enabled. Actual behavior: The Save button isn't enabled.
Platform Analytics PRB1736557	A saved visualization and filter modal in a dashboard must use a new API	A saved visualization and filter modal in a dashboard must use a new API so lists don't share every visualization.	Impersonate an itil user. Open a dashboard. Add a new visualization. Select the saved visualization. The list displays all visualizations. It should only list visualizations shared with the itil user. List items should be the same as the 'Data visualizations' library.
Predictive Intelligence PRB1754231	The default ml_dc_vip_list incorrectly maps AMS20 and DUB0 to AMS200 and DUB200 predictor infrastructure, which doesn't exist	Insert records exist in Glide that are populating ml_dc_vip_list with incorrect mapping for AMS20 and DUB20. Records are added inserting this mapping on instances: ams20 > ams200 and dub20 > dub200. This is incorrect. AMS20 and DUB20 are regular capacity pods that were brought online recently. They should point to ams100 and dub100 infrastructure pods.	Provision an instance in AMS20x or DUB20x capacity pods. Invoke ML.
Project Management PRB1745099	The baseline created on a planning console isn't listed on the 'Financials' tab in Project Workspace (PW)	As a project, demand, or product manager, when users create a financial baseline from the classic experience, they need the financial baseline also to be supported in the new experience. This is so users can compare costs in the new experience.	Create a project on PW. Open the 'Financials' tab. Create a baseline. Open the planning console. Create another baseline. Come back to the 'Financials' tab to check the baselines. The baseline created on the planning console isn't listed.
Project Portfolio Management PRB1743629	A confirmation modal pop-up isn't displayed on a resource plan when the date is updated on the allocated resource plan	A pop-up message should be displayed, but it isn't displayed when edited from Project Workspace. A confirmation modal isn't displayed when a resource plan is created on the closed projects.	Create a project. Create a resource plan. Keep it in an allocated state. Navigate to Project Workspace. Navigate to the 'Resource plan' tab. Edit the details. Click Save.
Scheduled Jobs PRB1754295	Long-running jobs during Scheduler V2 to V3 switchover are requeued and run twice	Any jobs running when the instance upgrade is marked as fully completed and take longer than 5 minutes to complete end up being requeued.	Update 'glide.central_scheduler.enable' to false. Create a job in sys_trigger with the name 'testjob' and script - gs.sleep(600000). Update 'glide.central_scheduler.enable' to true. Wait for 5 minutes. Navigate to the localhost logs. Expected behavior: testjob shouldn't be requeued and not run twice. Actual behavior: testjob is requeued and run twice.
Schedule Optimization PRB1740617	Validation must be added against the 'intraday day events' table for the type 'wm_work'	Currently, agents that aren't part of a group with the 'wm_work' type can create intraday events. That shouldn't happen.
Schedule Optimization PRB1748610	In batches, a locked task query isn't sending the closed completed tasks	This was observed during TridentCare testing.	Add the 'Created on' filter in the map provider logs module in Dispatcher Workspace.
Schedule Optimization PRB1750396	The 'Home' location of the agent should be sent to optimization engine	The optimization engine has been operating under the assumption that the agent location sent from the Glide side is the 'Home' location of an agent. The 'Home' location for the agent should be matched.
Schedule Optimization PRB1751727	The scheduling_resolution property default value should be changed from 100 to 300 to round off the scheduled start time to the nearest 5 minutes	Currently, the engine returns a scheduled start as 11:01:03. With the new default value, the engine returns it as 11:05:00, rounding it to the closest fifth minute.
Schedule Optimization PRB1751745	There's a policy override issue	Intraday run wasn't respecting an overridden policy and fell back to the default policy.	Create a policy without an optimization value objective. Set this policy as a default policy for intraday optimization. Create a different policy with the optimization value objective. Add an optimization value in eligible tasks. As a dispatcher, override the default policy with the policy with an optimization value objective. Run optimization. The optimization value isn't being considered during optimization even if the selected policy has the objective. This shouldn't happen. Once the dispatcher overrides the policy, optimization should consider the overridden policy.
Schedule Optimization PRB1756570	The scope qualifiers and Intraday qualifiers related lists are still displaying the Assignment group field if a non-admin user enables territory		Activate Territory, FSM, and SO plugins on the instance. As a non-admin user, navigate to Territory planning > Territory Model. Select a model. Check the 'Active' check box. Save. Navigate to Schedule Optimization > Scopes. Click the New button to create a scope. In the scope creation form, notice that the related list displays qualifiers even if users have territories enabled. The same happens while creating a new Intraday configuration. This shouldn't happen. Users should see territories as qualifiers instead of assignment groups after enabling territories.
Secondary Database Management PRB1754273 KB1641157	There's gateway password decryption failures due to a premature attempt to use Key Management Framework (KMF)	An error displays in the logs.	Refer to the listed KB article for details.
Seismic Framework PRB1751888	Cache busting doesn't occur on pages routed through the UxPageProcessor	Cache busting doesn't occur on pages routed through the UxPageProcessor, resulting in state assets being served.	Navigate to an instance. Clear the cache with /cache.do. Navigate to a page that goes through the UxPageProcessor. Look at Dev Tools > Application > Cache Storage. Take note of how many POLARIS_STATIC_ASSET_CACHE entries exist and their timestamps. Visit /cache.do again in a new tab. Refresh the tab with the page that goes through UxPageProcessor. Expected behavior: A new POLARIS_STATIC_ASSET_CACHE entry is created with the new cache buster. The timestamp/uxasset/set-cache-buster /timestamp.js header tooltip is present in the document response. Actual behavior: POLARIS_STATIC_ASSET_CACHE entries all have previous values. The cache busting header tooltip isn't present in the document response.
Service Catalog Builder PRB1758640	There's family changes to support Text2Catalog	This is a product update.
Service Catalog PRB1751251	Attachments are disappearing when attached to the service catalog forms	Attachments added on the catalog item when submitting in the backend aren't displayed on the form but the same attachment is added to the created requested item (RITM) after submission.	Log in to an instance. Impersonate a user. All > Service Catalog > Internal Catalog > Open any catalog item. Attach the attachment. Observe that the attachment isn't attached. Fill in the required variables. Submit the item. Observe that the created RITM has the attachment.
Service Catalog PRB1751304	Variable details aren't displayed after upgrading to the latest patch releases on the Agent Mobile app	The issue occurs on both iOS and Android.
Service Catalog PRB1752902	There's family changes for large language module (LLM) Virtual Agent May support	This is a product update.
ServiceNow Studio (Legacy) PRB1737701	Pages in legacy Dev Studio aren't honoring the values set in Next Experience	The issue is also applicable when creating an app/file or updating an existing app/file.	On an instance that has Next Experience enabled, navigate to System Application > Studio. Open Studio. Expected behavior: Users should get Next Experience. Actual behavior: Users get the classic UI experience.
Service Portfolio Management PRB1731498 KB1587599	There's long loading times for Service Portal on a cold cache of catalog items	A large amount of data in the Service Catalog is needed to reproduce the loading times. The initial load time takes 40+ seconds to render the categories and catalog items available to the user. It's a few seconds in subsequent loads with caching.	Refer to the listed KB article for details.
Service Portfolio Management PRB1750898	The Digital Portfolio Management (DPM) true-up for Washington DC (WDC) can't be installed on a WDC instance due to Enterprise Portfolios applications	Enterprise Portfolios applications haven't been trued-up. An offline user isn't able to install DPM or the Enterprise Portfolios application.	Navigate to an instance. Navigate to plugins. Try the instance DPM true-up version for Washington DC.
Service Reliability Management (Family release) PRB1747909	Site Reliability Metrics (SRM) responders/managers are unable to add an integration	In Vancouver, the usage of [].includes() appears not to be supported in the global scope.	Log in as an SRM responder or manager. Navigate to the 'Services' page (/now/sow/srm-services). Create a service or select an existing one. Navigate to the 'Details' tab. Expected behavior: The user should have write access to the service. Actual behavior: The user doesn't have write access.
Service Reliability Management (Family release) PRB1749826	Service Reliability Management (SRM) admins don't have write access on incidents assigned to SRM teams that they aren't part of		Create an SRM team. Add users. Impersonate one of those users. Create an incident. Impersonate as an SRM admin who isn't part of that team. Open the incident. Verify that the admin doesn't have write access to the incident.
Session Management PRB1741142	An uncaught exception causes stale records in the 'Active Transactions (All Nodes)' table	This issue started happening when users turned on the glide.guest.active .session.life_p feature. A race condition happens when the guest session is destroyed at the same time as the HTTPSessionTimeoutManager, which invalidates that session due to the guest active timeout configuration.	Run 'IT: ActiveSessionTimeoutConcurrentIT'.
Session Management PRB1748598	Stale IDs are leading to ~8 MB/week being leaked on JPMC, causing the memory floor for the node to rise gradually	Heap dump analysis shows 200 K+ entries, most of which are related to a single user. Similar stale IDs are found in the other two maps.	Observable in a heap dump.
System Scheduler PRB1746390	Scheduler v3 doesn't run a job when the state is manually changed from 'Running' to 'Ready'	Historically, users can force a running job to execute again by setting the state back to 'Ready' without clearing its claimed_by field. In Scheduler V2, the job is picked up again for execution by another node. In Central scheduler, this job isn't picked up for execution by any node.	Enable the central scheduler. Create a job with: name = anything next_action = now trigger_type = repeat with the specified interval state = Running claimed_by = anything Set the job to a 'Ready' state. Expected behavior: The job should be picked up for execution. Actual behavior: If central scheduler is enabled, the job is never picked up for execution. If the v2 scheduler is enabled, the job is picked up by some other node.
Table Administration and Data Management PRB1755862	There should be default ordering by sys_id on all metadata tables and a few non-metadata tables	There's differences in forms' layouts.
Time Card Management PRB1708797	A warning isn't displayed when a time card is recalled	After recalling a time sheet, the 'merge duplicate time cards' icon is no longer displayed to the user. The user is expected to have the same behavior before submitting the time sheet.​	Navigate to the time sheet portal. Add time on a record type. Add that same record to the time sheet again and add the time. There should be a triangle on each time card 'saying merge duplicate records' instead of 'merging submit the time sheet'. Recall the time sheet. The merge duplicate time sheets aren't there.
UI Field Administration PRB1674332	Screen readers don't narrate any information when the focus moves to link content	Users who rely on screen readers face difficulty if the screen readers aren't narrating any information when the focus moves to link content.	Open any instance in an Edge/Chrome browser. Enable the NVDA/JAWS screen reader. Navigate to a 'cmdb_consumable_product_model' record. Navigate to the 'Click to Add' link in the 'Product Catalog' section. Verify whether the screen readers are narrating any information when the focus is on the 'Click to Add' link or not. Expected behavior: Screen readers must narrate the name and the role of the link when the focus lands on the link. Actual behavior: Both the screen readers aren't narrating any information when the focus lands on the 'Click to Add' link.
UI Form Administration PRB1723383	A missing empty table check is causing an undefined page route on 'Create article'		Create a knowledge feedback task. Confirm that there's the latest version of sn_sow_itsm_cont. Navigate to Workspaces > Service Operations Workspace. Navigate to List > Knowledge > All Articles. Open the knowledge article with the knowledge feedback task. Click the Edit button. Click Related records > Feedback Tasks > Open the Feedback. Click Create Article. Expected behavior: A new tab to create an article is opened. Actual behavior: A new tab is opened with the following error message: 'The page you are looking for could not be found'.
UI Form Administration PRB1738564	A ribbon component displays the page title as 'Contact'	360 components display the title as 'Contact' if the primary field is empty.	Navigate to any Vancouver instance. Create UX ribbon settings for the sn_customerservice_case table for any field. Don't add a secondary user. Create ribbon configuration settings. Open a case record in CSM Workspace where the Contact/Assigned To field is empty. Notice that the ribbon component displays the title of 'Contact'.
UI Form Administration PRB1746032	kb_knowledge routing opens a record twice due to sys_ux_data_broker_ transform being executed twice	There's a missing empty table check.	Create a knowledge feedback task. Confirm that the latest version of sn_sow_itsm_cont is installed. Create a feedback task. Navigate to Workspaces > Service Operations Workspace. Navigate to List > Knowledge > All Articles. Open the knowledge article with the knowledge feedback task from step 1. Click the Edit button. Click Related records > Feedback Tasks > Open the Feedback. Click Create Article. It opens the record twice.
Unified Filter PRB1731318	Unified filters in the Seismic dashboard don't apply on 'compatibility' mode Performance Analytics (PA) widgets		Create a Core UI dashboard with a breakdown source filter on the 'Incident' priority. Create a PA widget pareto chart on the dashboard. Migrate the dashboard to Platform Analytics experience. The filter isn't applied to the PA widget.
UX Framework PRB1713439	Associating records in Service Operations Workspace isn't working as expected	When a record is associated with a new interaction record, the associated record should open beneath the existing interaction. Instead, it's creating an interaction parent tab and displaying the associated record beneath it.	Log in to a Vancouver instance. Navigate to Service Operations Workspace > + > New Interaction. Fill in the field type as 'Chat'. Click the 3 dots. Associate the record. Expected behavior: The associated record opens as a subtab in the current interaction. Actual behavior: The associated record opens as a subtab in a new interaction.
Virtual Agent Actionable Notifications PRB1760636	Non-actionable notifications are sent during the middle of conversation in LLM and are loading non-LLM topics	If a user selects any topic, it errors out.	Create a non-actionable notification. Navigate to Teams in 'LLM' mode. Enter 'hi' to get the topic picker in Teams or web client. Trigger a non-actionable notification. On the previous topic picker, check the menu again to continue. Expected behavior: The user should see LLM greetings and topics in the topic picker. Actual behavior: The user sees non-LLM greetings and non-LLM topics in the topic picker.
Virtual Agent Designer PRB1756187	Internal text is displayed in the record summary on NAP	The AI response contains '[insert resolution notes here]' text.
Virtual Agent Designer PRB1756739	There's a zBoot error related to VADesignerUtils	An error displays: 'Glide SYSTEM SysLog WARNING *** WARNING *** Evaluator: com.glide.script.RhinoEcmaError: undefined is not a function...'
Virtual Agent Designer PRB1759073	The 'Now Assist Live agent' topic isn't being executed	An error displays in the logs.	In NAVA, type 'Connect to Live Agent'. The 'Now Assist Live agent' topic isn't executed.
Virtual Agent Designer PRB1759951	Users are unable to duplicate the Virtual Agent (VA) LLM topics in track/ washnowassiststable or track/ xnowassist	There's an error in the console.	Navigate to VA designer. Try to create a duplicate VA topic LLM. Add a deflection into the graph. Attempt to duplicate the topic. Enter a duplicate topic name. Expected behavior: The user should able to create a duplicate LLM topic. Actual behavior: The user is unable to create a topic.
Virtual Agent PRB1720318	A Virtual Agent topic with a text response node that contains a rich text message with a text link generates a mismatched token when an ATF test step is executed, causing a test step failure	Running Automated Test Framework (ATF) to test Virtual Agent topics may encounter a step failure. This can happen when a topic's node has a text response containing rich text with a text link. A test step of a topic with text response node that contains rich text message with text link is saved with a token. When a test case is executed, a new token is generated and it doesn't match with the expected message, which has a different token.
Virtual Agent PRB1746241	Two records are set to default as true for Virtual Agent 'Skill Search Capability'	Only one record should default as true for Virtual Agent 'Skill Search Capability'.
Virtual Agent PRB1746867	Virtual Agent's (VA) 'Abandoned chat' scheduled job short-circuits if it's unable to fault a conversation		Start a few VA conversations. Wait for the idle time to expire. Manufacture some kind of problem with one of the conversations, something that causes the 'Abandoned' job to fail catastrophically. Manufacture a problem with the corresponding interaction that prevents it from being closed. Run the 'Time Out Abandoned VA Conversations' scheduled job. Expected behavior: The job should close all conversations. Actual behavior: An uncaught exception causes this job to stop early.
Virtual Agent PRB1749109	Skipping the secure text input in the topic, the next normal text input is rendered as masked	Skipping the secure text input in the topic, the next normal text input is rendered as masked and throws an error when a user clicks Enter for that variable question.	Create a new topic block of type 'LLM'. Add the Secure text input control. Add a prompt, like 'Enter password'. For 'Hide or skip this node', return true. Add another input control of type 'text'. Test this topic. Skip the first question. Enter some text for the second question. Expected behavior: It should render as plain text, as it's a single-line text. Actual behavior: The second question is rendered as masked.
Virtual Agent PRB1751841	Non-actionable notifications aren't sent in 'LLM' mode on Teams	The consumer account doesn't change to 'LLM'.	Create a non-actionable notification for Teams. Navigate to Teams in 'LLM' mode. Link a user to Teams. Trigger a notification. Expected behavior: The notification should be received on Teams. Actual behavior: The notification isn't sent on Teams.
Virtual Agent PRB1753464	An initial search text is encoded in Agent Chat and the conversation context	There's no functional issue with the normal conversation flow. There are 2 main issues: 1. The first message in the conversation transcript is illegible. 2. The first message displayed to the agent is illegible, especially with a link. If there's a link, the agent isn't able to click it.	Using any Conversational Adapter, start a conversation with complex search text. Expected behavior: The search text should be displayed as sent when viewed by an agent in Agent Chat or in the conversation transcript. Actual behavior: The entire search text is URL encoded.
Virtual Agent PRB1754166	If the correct version of NAVAD isn't installed, 'Test/Publish' should be turned off	There should be a check to confirm that users have the minimum installed level. If not, 'Test/Publish' should be turned off for those topics with some explanation of why those features are turned off.	Create an instance using an older version of the store app. Upgrade the instance to confirm that the NAVAD version is still 1.2.1. Expected behavior: 'Test/Publish' should be turned off, displaying some sort of information about why. Actual behavior: The user can still 'Test/Publish' even though the features necessary aren't available.
Virtual Agent PRB1758634	There's an error setting integer values in vaVars	A dynamic list with pagination is failing on large language model (LLM) conversations because it tries to set an integer to vaVars. There's an error in the logs.
Virtual Agent PRB1759564	Unable to get and set global conversation context variables using vaVars		Create an LLM topic that displays the dynamic choice picker using a script. Add the VirtualAgentChoicePagingUtil script for paging through the options. Expected behavior: The dynamic choice picker displays 'Show 10 more...' at the end of loading options and Clicking 'Show 10 more...' should display the next 10 rows. Actual behavior: The Dynamic choice picker shows 'Show 2 more...'. Clicking 'Show 10 more...' didn't load the next 10 rows.
Virtual Agent PRB1760598	There's zBoot errors in Washington	There's an error in the logs: 'Syntax Error or Access Rule Violation detected by database (ERROR: column "definition" does not exist)'.
Virtual Agent Web Client PRB1754600	Users aren't able to send a query when a pre-chat survey is kept on		Set up the pre-chat survey. Set up the LLM conversation. Open the portal. The pre-chat survey displays with the input box. Enter the query. Click >. Users aren't able to send a query.
Work Order Management PRB1747917	The travel line indicator for 'Closed Complete' tasks and 'Incomplete' tasks aren't visible on Dispatcher Workspace (DWS)	Even if 'actual_travel_start' isn't empty, the condition is using 'Expected travel start'. If the StartTimeField uses the 'expected_travel_start', the event body's start_time is made equal to mid_time. Therefore, no travel indicator appears.

Fixes included

Unless any exceptions are noted, you can safely upgrade to this release version from any of the versions listed below. These prior versions contain PRB fixes that are also included with this release. Be sure to upgrade to the latest listed patch that includes all of the PRB fixes you are interested in.

• Washington DC Patch 2 Hotfix 1
• Washington DC Patch 2
• Washington DC Patch 1
• Washington DC security and notable fixes
• All other Washington DC fixes