26.1.6
Queue failover calls failing
26.1.5
OpenAPI additions
26.1.4
BroadWorks improvements
Additional security for scheduled reports
Callback enters loop and causes agent freeze
26.1.3
Unable to edit users
26.1.2
Update to Call Recording Behavior
We are introducing an update to call recording to further strengthen compliance and data governance. Going forward, recordings will automatically stop when a call is transferred to an endpoint (such as a direct phone number). Recording will continue for agent-to-agent and agent-to-queue transfers. This adjustment ensures clearer alignment between recording scope and organizational boundaries. To accompany this, we are exploring more flexible options that will allow customers to designate specific phone numbers as eligible for continued recording in the future.
Removal of 118.dk Caller Lookup Integration
We are removing the integration to the external 118.dk caller lookup service, as the provider is discontinuing API access. Following this change, external caller information from 118.dk will no longer be displayed in Zylinc. This update ensures a predictable and stable experience by eliminating dependency on a service that is no longer supported. Customers using caller lookup functionality should review their setup and consider alternative caller lookup sources if needed.
Also please note that we have release a new statistics
26.1
The 26.1 release includes several important fixes that reduce the risk of call disruptions, improve callback reliability, and make agent state handling more predictable during busy periods. It also introduces new features that support better coordination and staffing throughout the day.
Agent notes - new feature
One new feature has been added to this release. Read more below:
Telenor presence
We have added Telenor presences status, giving Telenor users real-time visibility into colleagues’ availability in Zylinc Cloud.
This helps agents coordinate outreach and reduces unnecessary call attempts when colleagues are unavailable.
Agent - bug fixes
Both minor and major bugs have been resolved in this release - including some calls freezing, ghots calls and wrap-up getting stuck. Read more below:
Call handling and queues
Callback stuck after decline or postpone: An issue was found where callbacks could get stuck if the customer declined or postponed the callback request, preventing proper handling in the client. This has been fixed, ensuring callbacks now update and clear correctly in all scenarios, including declines, IVR exits, and postponed calls. [27343]
Unclear callback error message: Agents experienced the error message “Callback activation failed, dial failed” when customers did not confirm their callback by pressing ‘1’ or hung up too soon. The issue has been addressed, which now includes clearer and more informative callback error messages for agents. [17600,19115]
Repeated notifications for closed queues: An issue was identified where agents received repeated notifications whenever another agent logged in, logged out, or changed their queue monitoring settings, if the affected queue was closed. The problem has been resolved, and notifications are now only shown when relevant queue status changes occur. [21363]
Unable to answer calls in queue: Agents experienced an issue where calls remained in the queue for over 30 minutes without agents being able to answer, despite being visible in the interface. We have redesigned our session manager, preventing this issue from reoccurring. [23873]
Calls freezing and multiple offers to agents: An issue was identified where calls offered to agents could not be answered, and in some cases, multiple calls were offered simultaneously. Fixes and optimizations have been applied to prevent offer loops and ensure stable call handling. [27133]
Ghost calls ringing without incoming call display: An issue caused agents to hear an incoming ringtone without any call being presented in the WebAgent. The problem has been resolved, where the agent-to-backend interface was made more robust, preventing ghost calls and ensuring calls are properly registered and handled. [21526]
Transfer completed popup shown to all agents: A bug was identified where the “Transfer Done” popup was shown to all agents in the same queue, even if they were not involved in the transfer. The problem has been fixed, and the popup is now shown only to the agent who handled the call and performed the transfer. [27788]
Agent stuck busy after transfer to queue with announcement: An issue was identified where agents remained in a Busy state after transferring a call to a queue with an announcement. The problem has been fixed, and agents are now immediately released from the call and shown as available as soon as the transfer is completed. [27980]
Wrap-up getting stuck after attended transfer: An issue was identified where agents could get stuck in wrap-up after handling an attended transfer, requiring a page refresh to continue. The issue has been fixed, and wrap-up now completes correctly after attended transfers. [27344]
Agent stuck offline with error message: A bug was identified where agents could appear offline and stop receiving calls, even though they were logged in. The handling has been improved, so only one active session is allowed, and any previous session is automatically closed. This prevents agents from getting stuck as inactive and ensures calls are delivered correctly. [27589]
Email Queues
Email queue not displaying messages with attachments: An issue was identified where emails containing attachments, such as PDFs or embedded files, did not display content in the Zylinc client. The problem has been resolved, and the Zylinc client now correctly displays both email text and attachments across all tested scenarios, including replies and multiple attachments. [27307]
- Incorrect “waiting conversations” count in email queues: An issue was observed where the number of waiting conversations in an email queue view was incorrect for some email queues. The problem was linked to synchronization issues when emails were manually moved or deleted from the inbox. Corrections have been made to improve synchronization. [27948]
- Email queue not receiving emails: A bug was identified where an email queue stopped showing incoming emails even though messages were present in the mailbox. The recovery handling has been improved so the service can restore access automatically instead of failing, and the documentation has been updated. [27962]
Calendar and scheduling
Calendar week starts on Sunday for en-US: An issue was found where the calendar view started the week on Sunday when the client was set to en-US, ignoring the computer’s weekday start setting. The problem has been fixed by explicitly setting Monday as the first day of the week in the localization settings. [27699]
Calendar not updating: An issue was identified where calendar events were not always updated correctly. The problem has been fixed, and calendars are now kept in sync as expected. [27368]
UI and usability
- Missing tooltips on contact card icons: The Web Agent contact card was missing tooltips on the dial and message icons, causing users to lack guidance on their functions. Tooltips have now been added, and language tokens have been created to ensure consistent localization across all interfaces. [27008,26956,27158]
- Configuration manager link not readable in Icelandic: In the Icelandic (is-IS) version of WebAgent, the Configuration Manager link text appeared invisible due to white-on-white or black-on-black contrast in different themes. The token has been corrected in the translation tool, and the link now displays correctly in all themes. [27044]
Calendar font size: An issue was identified where the text in the Calendar component in the Web Agent appeared significantly larger than before due to incorrect styling being applied. The problem has been fixed, and the Calendar view now uses the correct font size again. [27256]
- Missing tooltips in Web Agent: An issue was identified where tooltips were not shown on several icons in the Web Agent, including icons highlighted in the main view and within a popup window. This has been corrected, and tooltips are now displayed correctly across the affected areas. [28182]
Agent name not displayed correctly: An issue was identified where long agent names did not display properly in the agent view, causing layout problems. This has been fixed, and agent names now adjust correctly regardless of length. [27588]
Reason codes not syncing correctly in the Web Agent: A bug was identified where reason code configurations did not stay in sync with the server when supervisors switched between queues quickly. The problem has been resolved, and the view now correctly reflects the selected queue and saves changes reliably. The reason codes experience has also been improved with clearer on/off behavior, immediate saving when selecting a profile, better error handling with rollback to the last known state, and clearer feedback to the user. [28035]
Warm transfer layout incorrect in Agent View: An issue was identified where warm transfers in the agent view displayed call parties in narrow columns instead of being stacked horizontally as intended. The correct horizontal layout will now be included. [28020]
Misc.
Federation to Teams not updating agent state: An issue was identified where the federation of agent state to Microsoft Teams stopped working, causing Teams presence to remain “Available” during active calls. Logging has been improved, initialization corrected, and error handling added to detect and alert if federation settings are missing. The fix ensures agent state now correctly federates to Teams, including automatic “Do Not Disturb” updates during calls. [26825]
Microphone permission not prompted in Navigator: Users relying solely on the Navigator client were not prompted to allow microphone access, preventing device selection for ringtone and headset. The boot logic has been improved, so the Navigator now requests microphone and headset permissions like the Web Agent. Once access is granted, devices appear instantly in the UI without needing a reload. [24174]
Mobile app crash when typing certain letters: The mobile app could crash when users typed specific letter combinations during search. This only affected tenants where one or more users had an empty display name. The app failed instead of handling the missing name. The issue has been fixed, and the app now uses the user’s email when no display name is available. [21038]
Admin - new features
Two new features have been added to this release. Read more below:
Mandatory secret expiration for improved integration reliability
To help prevent unexpected integration issues, we’ve made secret expiration dates mandatory. This ensures that every secret has a defined lifespan, enabling the system to notify you proactively when a secret is nearing expiration. By enforcing expiration dates, we strengthen integration reliability and make it easier for you to manage credentials securely and predictably.
Open API improvements for transcription retrieval
We’re improving our Open API so integrators can more efficiently fetch transcription content. The updated Get Transcriptions endpoint offers finer filtering, flexible detail levels, and higher retrieval limits, making it easier to build reliable downstream services like putting transcriptions and summaries into CRM records
See: https://zylinc.zendesk.com/hc/en-us/articles/22849114318876-Zylinc-Open-API-User-Guide
Admin - bug fixes
Both minor and major bugs have been resolved in this release - including some being unable to delete queues and creating new calendars. Read more below:
Call handling and queues
- Media stack error when editing or creating failovers: Administrators experienced a “Media stack error” when attempting to edit or create new failovers in the Configuration Manager. The issue has been corrected, ensuring failover changes and creation now work reliably without errors. [27416]
- Announcement profile not updating through callflow: Calls kept the initial announcement profile throughout their journey, even after transfers, failovers, or IVR routing. This has been corrected so digit collection now follows the active queue’s announcement profile. [27439]
-
Unable to delete queues: An issue prevented queues from being deleted because they were still linked to inactive or leftover agent settings. The underlying cleanup problem has been fixed, and the configuration view has been improved, so remaining references are now visible and can be resolved. [27383]
Email Queues
-
Configuration email queues blank page: An issue occurred where editing or creating new email queues in the configuration view resulted in a blank page. The fix has now been fully applied to all tenants, ensuring the configuration page loads correctly when editing or creating email queues. [26583]
Users
- Reimported users remained deleted: An issue was identified where reimporting a previously deleted user via CSV did not reactivate the user. The user stayed marked as deleted in the Configuration Manager, shown in red. This has been fixed so that reimporting a deleted user now correctly restores and enables the user. [26901]
- Timeout when editing user information: Editing a user’s tags or keywords caused the system to freeze, and saving changes ended with a timeout error. In some cases, this also prevented users from logging in. The issue has been fixed by improving how the system handles user updates, so changes now save quickly and reliably. [27427]
- Unable to save user edits in configuration manager: Editing a user and saving changes caused the system to time out, leaving the save button spinning and preventing users from logging in. The issue was caused by slow database handling when saving user updates. The problem has been fixed by improving performance in how user data is processed, ensuring edits now save quickly and reliably. [27526]
Calendar and scheduling
- Unable to create new calendars: An issue was identified where creating new calendars failed with a “Failed to save configuration” error. Over time, repeated CSV uploads caused the system log to grow excessively, slowing down configuration checks until they timed out. The issue has been fixed so CSV file contents are no longer stored in the log, preventing the problem from reoccurring. [28024]
- Calendar showing detailed events: An issue was identified where private calendar events could be shown with full details instead of being hidden. This happened in rare cases when many calendar access rules were involved. The issue has been fixed, and private events are now always shown as private. [27422]
Transcription and recording
- Transcription language defaulting to English: Customers experienced outbound call transcriptions appearing in English, even though Danish was set as the tenant’s default language. The issue has been resolved by setting the default transcription language based on the configured language, ensuring correct localization for all tenants. [25896]
- Recording cut when call is placed on hold: An issue was identified where call recordings stopped when a recorded callback call was placed on hold. This caused the recording to be cut short even though the conversation continued. The problem has been fixed, and recordings now continue correctly when callback calls are put on hold. [27969]
Statistics and reports
- Work mode missing from statistics: Work mode data disappeared from the statistics view after being renamed to AudioWorkMode during the introduction of email functionality. This caused missing work mode information in reports. The data source has been updated to include AudioWorkMode, and AudioStatus has been added to the table view for improved visibility. [27184]
- Scheduled report missing attachments: An issue was identified where scheduled activity reports sometimes did not include the CSV or PDF attachment. The retry logic has been improved by increasing both the number of retries and the interval between them, ensuring reports that take longer to generate are now correctly attached. [24779]
Misc.
- Error message prefixed with “AxiosError”: When entering an invalid phone number in the GDPR anonymization field, the error message was incorrectly prefixed with “AxiosError,” causing confusion. The prefix has been removed, and the message now displays clearly without technical details. [27166]
-
Logout from Zylinc Cloud landing page logs out Mobile Agent: An issue caused mobile agents to be set offline when a user logged out from the Zylinc Cloud landing page. This happened if the user had logged in on both WebAgent and mobile, and then logged out from the landing page while the mobile app was inactive. The problem has been fixed. [23979]