2025.3 Release Notes

Core

Blocked Reports Due to Envoy Restrictions

Issue: Some reports failed to load because updated Envoy configurations blocked access to the /reportAction.do endpoint.

Fix: Updated backend configurations to restore access to all affected reports.

Core

Stale Service Monitoring from Probes

Issue: Monitoring data appeared outdated because of interrupted probe communication.

Fix: Corrected probe-side handling to ensure continuous and accurate service monitoring.

Core

GSuite Login Redirect Restored

Issue: After upgrading to 2025.2.1.8, users logging in via GSuite were not redirected to the Google SSO page as expected.

Fix: We corrected the redirect logic to ensure GSuite users are seamlessly routed to Google SSO after entering their email without needing to manually click a “Sign in with Google” button.

Core

Analytics Reporting – NCOD Instance

Issue: Some hosted environments lacked analytics data, which prevented users from accessing dashboards and reports.

Fix: Enabled analytics data capture and availability for this instance.

Core

Slow Device Deletion Performance

Issue: Deleting a large number of devices caused performance slowdowns due to database locking.

Fix: Optimized the deletion process to reduce database contention.

Core

Launch Darkly Crash on Legacy Customers

Issue: The Launch Darkly client crashed due to missing timestamps for older customers.

Fix: Added fallback logic to handle missing creation time.

Core

All Devices View Filtering Incorrectly

Issue: The All-Devices view did not respect Service Organization (SO) boundaries during filtering.

Fix: Corrected the filter logic to properly account for SO segregation.

Core

UI Performance – All Devices page slow performance

Issue: The "All Devices" page became slow when using the “More Devices” filter option.

Fix: Improved page performance by optimizing how the filter interacts with large datasets.

Core

Take Control reinstalls triggered Too Early

Issue: Some devices were flagged for reinstall prematurely due to recovery logic errors.

Fix: Improved recovery checks to avoid unnecessary reinstalls.

Core

Device Selection UX in Custom Property Updates

Issue: Device selections remained active after custom property edits, leading to UI confusion.

Fix: UI now clears selections properly after applying custom property changes.

Slow performance seen for some hosted instances

Issue: Some hosted servers suffered from lag and degraded performance.

Fix: Refined background job processing and cache handling.

Data Access Object (DAO) Design Bottleneck Causing System Hang and Event Processing Delays

Issue: In large environments, especially NCOD, systems were intermittently unresponsive (up to 15 minutes). This was caused by too many background tasks trying to access the same part of the system at once (called the DAO layer), which created a traffic jam and slowed everything down — especially when the system was busy creating or sending event updates. 

Fix: Improved how these tasks access the system, allowing them to take different paths and work in parallel, eliminating the bottleneck and keeping the system responsive.

Inconsistent Patch Data Eventing

Issue: Patch data events were consistently delayed, requiring manual intervention  to scrape missed data while the patch data is expected to be live-evented as it is added to the N-central server.

Fix: Improved the eventing mechanism to ensure patch data is emitted in real time, eliminating the need for manual rescrapes.

High API Load from Active Issues Scroll Behavior

Issue: Hierarchy and customer data were occasionally missing from the reporting warehouse. 

Fix: Addressed ETL gaps and enforced data completeness checks. 

Excessive API Calls When Scrolling in All Devices View

Issue: Selecting devices triggered API calls per device instead of per page, causing UI slowdown in large environments.

Fix: Updated pagination logic to batch API calls per page, improving performance and reducing network load.

Core

Autotask: Ticket closures issues

Issue: Tickets failed to close automatically during maintenance events.

Fix: Updated closure logic to ensure consistent handling with Autotask integration.

Core

System Error When Creating Customers with Correlated Notifications

Issue: When trying to create a new customer while setting up correlated notifications, the system would show an error and fail to complete the setup. This happened because the system was incorrectly looking for a "profile ID" that didn’t exist, instead of using the correct "correlated profile ID".

Fix: The system now properly uses the correlated profile ID when setting up correlated notifications, preventing the error and allowing customers to be created successfully.

Core

PSA Integration Stability (ConnectWise / Autotask)

Issue: PSA export profile was unintentionally deleted after customer removal from ConnectWise or Autotask.

Fix: Adjusted deletion logic to retain export profiles when associated customers are removed.

Core

Data Type Handling in ConnectWise Exports

Issue: Exports to ConnectWise did not honor distinctions between “Text” and “Number” fields.

Fix: Field export logic now correctly respects data type definitions.

Core

Asset Scans not running according to schedule

Issue: Scheduled asset scans were not triggering at expected intervals.

Fix: Fixed internal scheduling mechanisms to restore consistent scan execution.

Core

User Access Control / Permissions

Issue: Misconfigured permission logic allowed users unauthorized access to privileged pages.

Fix: Applied proper permission restrictions to prevent escalated access.

Core

LDAP to SSO Account Migration Logic

Issue: Standard LDAP accounts were incorrectly migrated to MSP SSO, causing login and account management issues.

Fix: Refined account migration rules to exclude standard LDAP users from MSP SSO transitions.

Core

Endpoint Detection and Response (EDR) Service Assignment

Issue: EDR monitoring services were not automatically assigned to newly onboarded devices.

Fix: Ensured automatic assignment of EDR services during device provisioning.

Core

Server Connectivity – EDS Communication

Issue: N-central server experienced periodic disconnections from the External Data Store (EDS).

Fix: Improved connection stability and retry mechanisms to maintain persistent communication.

Core

SNMP Parsing – Linux Custom Types

Issue: SNMP failed to correctly parse custom entries of type "String" on Linux systems.

Fix: Fixed string-type handling logic in SNMP implementations.

Ecoverse

LDAP User Login Name Validation Appears Incorrectly After Import

Issue: LDAP user accounts displayed a validation error on the Login Name field when editing, despite this field being non-editable for LDAP users. This prevented any updates to the account.

Fix: The Login Name field for imported LDAP accounts is now correctly disabled, and the erroneous validation error has been removed.

Ecoverse/

Core

Device Provisioning & Remote Access (Tools / Take Control)

Issue: After upgrading to 2024.6, devices failed to receive configuration, causing tools and Take Control to stop working. This was due to missing or unavailable Ecoverse URLs.

Fix: Introduced fallback logic and improved error handling to ensure devices can still be provisioned even when Ecoverse URLs are missing, restoring functionality for tools and Take Control.

PSA-System Error with Halo Customer Mapping

Issue: Errors surfaced when mapping customers using Halo PSA integration. 

Fix: Adjusted field validation to resolve mapping inconsistency.  Android N-central Mobile App – 2FA Entry Clears Login Fields-

Android N-central Mobile App – 2FA Entry Clears Login Fields

Issue: After 2025.2 release, users (primarily on Samsung devices running Android OS 15) experienced an issue where switching to an authenticator app to retrieve a 2FA code caused the N-central login fields (server, username, password) to be cleared.

Fix: Issue resolved in the latest version of the Android N-central mobile app. Users can now retrieve 2FA codes from external apps without losing login field data.

NC Agent/Probe startup fails due to missing AM Agent executable

Issue: The NC Agent or Probe fails to start on systems where the AM Agent service executable is not present. This resulted in service startup errors, preventing normal operation.

Fix: Updated the installer to ensure the AM Agent service executable is properly deployed and verified during startup.

N-central Incorrectly Invalidating Agent Sessions

ssue: Agent sessions were being incorrectly invalidated because they were misclassified as UI sessions.

Fix: Updated logic to correctly identify agent-originated sessions and prevent unnecessary termination.

Devices Not Loading Due to Incorrect Cove Backup Check

Issue: An incorrect detection method for Cove Backup Manager caused all devices to fail loading.

Fix: Replaced the faulty logic with a stricter Cove detection approach and after jetty start, the Cove sync maintenance task will auto-correct affected devices.

Missing Audit Logs on NCOD Servers

Issue: Certain audit events were not being logged in NCOD-hosted environments. 

Fix: Restored events capture paths and ensure reliable log generation.