Channelio 장애 기록

Temporary issue has been detected with Customer Support (Inbox) Snippet feature. We are currently investigating the cause and will provide updates shortly. We apologize for any inconvenience this may cause.

The temporary issue has been resolved, and the service is now fully operational. We apologize for any inconvenience this may have caused. - Incident Time: July 10, 2026 04:47 ~ 14:31 (KST) (approximately 9 hours 44 minutes) - Affected Scope: Channels using snippets - Cause: A styling issue caused dropdown and snippet colors to not display correctly. - Resolution: The system was restored to its previous state and the root cause has been fixed. Pre-release checks will be strengthened to prevent recurrence. We will continue to strengthen monitoring to prevent similar issues from occurring again.

The temporary issue has been resolved, and the service is now fully operational. We apologize for any inconvenience this may have caused. - Incident Time: June 24, 2026 20:48 ~ 21:23 (KST) (approximately 35 minutes) - Affected Scope: Phone feature (inbound/outbound calls) - Cause: A change in the internal data structure caused the Meet feature to stop functioning properly. - Resolution: The system has been restored to its previous state, and we will be improving our deployment process to prevent recurrence. We will continue to strengthen monitoring to prevent similar issues from occurring again. Temporary issue has been detected with Phone feature. We are currently investigating the cause and will provide updates shortly. We apologize for any inconvenience this may cause.

Temporary issue has been detected with Kakao Talk Integration Service. We are currently investigating the cause and will provide updates shortly. We apologize for any inconvenience this may cause.

The temporary issue has been resolved, and the service is now fully operational. We apologize for any inconvenience this may have caused. - Incident Time: June 23, 2026 15:00 ~ 17:00 (KST) (approximately 2 hours) - Affected Scope: File and image messages sent by customers via KakaoTalk. - Cause: File and image messages were not properly received due to a Kakao service outage. - Resolution: The service was restored following the recovery of Kakao's service outage.

Channel Talk system is experiencing an issue. We are currently investigating the cause and will provide updates shortly. We apologize for any inconvenience this may cause.

The temporary issue has been resolved, and the service is now fully operational. We apologize for any inconvenience this may have caused. - Incident Time: June 22, 2026 14:50 ~ 15:15 (KST) (approximately 25 minutes) - Affected Scope: All channels accessing the Channel Talk app and pages - Cause: An error occurred upon access due to a display configuration issue within the admin app. - Resolution: The affected configuration has been corrected and the service has been restored.

The temporary issue has been resolved, and the service is now fully operational. We apologize for any inconvenience this may have caused. Temporary issue has been detected with AI ALF feature. We are currently investigating the cause and will provide updates shortly. We apologize for any inconvenience this may cause. - Incident Time: June 5, 2026 14:34:45 ~ 14:38:30 (KST)

A fix has been implemented and we are monitoring the results.

Temporary issue has been detected with SMS Sending Service. We are currently investigating the cause and will provide updates shortly. We apologize for any inconvenience this may cause.

The temporary issue has been resolved, and the service is now fully operational. We apologize for any inconvenience this may have caused. - Incident Time: May 24, 2026 04:00 ~ May 26, 2026 10:50 (KST) (approximately 2 days 6 hours 50 minutes) - Affected Scope: Domestic SMS sending via the marketing and follow-up notification features - Cause: During an internal system maintenance, SMS processing was not resumed, resulting in a failure to deliver text messages. - Resolution: SMS processing has been resumed and the service has been restored. Two-stage monitoring has been added to prevent recurrence.

We have identified the cause of the current outage and have taken corrective action. We are currently monitoring the service for stabilization. We will provide you with an update as soon as we confirm the detailed cause. We apologize for any inconvenience this may have caused.

Temporary issue has been detected with Kakao Talk Integration Service. We are currently investigating the cause and will provide updates shortly. We apologize for any inconvenience this may cause.

The temporary issue has been resolved, and the service is now fully operational. We apologize for any inconvenience this may have caused. - Incident Time: May 21, 2026 19:00 ~ 22:00 (KST) (approximately 3 hours) - Affected Scope: Channels using messaging integrations (KakaoTalk, WeChat, etc.) and shopping mall integrations (GodoMall, ShopBy, etc.) - Incident Details: 1) Messages could not be sent through messaging apps such as KakaoTalk and WeChat 2) Order information was not saved for integrated shopping malls such as Godo Mall and ShopBy - Cause: A compatibility issue between some servers occurred during a system update. - Resolution: The affected servers were updated to restore compatibility and resolve the issue.

Channel Talk system is experiencing an issue. We are currently investigating the cause and will provide updates shortly. We apologize for any inconvenience this may cause.

A fix has been implemented and we are monitoring the results.

The temporary issue has been resolved, and the service is now fully operational. We apologize for any inconvenience this may have caused. - Incident Time: May 18, 2026 10:03 ~ 11:14 (KST) (approximately 71 minutes) - Affected Scope: Delays in inbox loading, IVR response, and App Store settings screen access in some environments during the incident period - Cause: Repeated data lookup requests in App Store-related features caused a temporary surge in load on the server managing the data, which also delayed other requests using the same server. - Resolution: The processing path causing the load was temporarily disabled to restore service. Going forward, the affected processing will be separated into its own flow, and frequently used information will be stored in advance for faster access. We have also adjusted request volume and processing time to prevent similar load from accumulating again. We will continue to strengthen monitoring to prevent similar issues from occurring again.

Channel Talk system is experiencing an issue. We are currently investigating the cause and will provide updates shortly. We apologize for any inconvenience this may cause.

We have identified the cause of the current outage and have taken corrective action. We are currently monitoring the service for stabilization. We will provide you with an update as soon as we confirm the detailed cause. We apologize for any inconvenience this may have caused.

The temporary issue has been resolved, and the service is now fully operational. We apologize for any inconvenience this may have caused. Incident Time: May 14, 2026, 19:47 – 20:03 (KST) Scope of Impact: Temporary unavailability of the Channel Talk admin application, including disruptions to auto-assignment and agent status updates. Cause: Temporary system overload during a deployment for product improvements. Resolution: Resolved the overload by scaling up the database servers and performing a rollback.

The temporary issue has been resolved, and the service is now fully operational. We apologize for any inconvenience this may have caused. - Incident Time: May 4, 2026 09:59 ~ 13:17 (KST) (approximately 3 hours 18 minutes) - Affected Scope: Intermittent delays and failures in customer contact lookups and marketing recipient lookups in some channels during the incident period - Cause: Load was concentrated on some servers handling data, causing temporary delays and intermittent failures - Resolution: Initial load balancing has been completed, and additional stabilization work will follow. Temporary issue has been detected with Contacts feature. We are currently investigating the cause and will provide updates shortly. We apologize for any inconvenience this may cause.

Temporary issue has been detected with AI ALF Statistics feature. We are currently investigating the cause and will provide updates shortly. We apologize for any inconvenience this may cause.

The temporary issue has been resolved, and the service is now fully operational. We apologize for any inconvenience this may have caused. - Incident Time: April 24, 2026 15:00 ~ April 27, 2026 23:50 (KST) (approximately 3 days 8 hours 50 minutes) - Affected Scope: Channels provided with the renewed ALF analytics feature - Cause: Data from some periods was omitted from aggregation, resulting in inaccurate results being displayed - Resolution: Aggregation has been corrected to include all previously omitted data from the affected periods.

We are currently experiencing a temporary issue with the Channel Talk service. We have identified an issue where users are unable to log in on Android mobile devices after a new login or logging out. We are urgently investigating the cause and will provide updates as soon as possible. We sincerely apologize for any inconvenience this may cause.

The temporary issue has been resolved, and the service is now fully operational. We apologize for any inconvenience this may have caused. -Incident Time: 2026-04-23 19:44 ~ 2026-04-24 11:05 (KST) -Affected Scope: Users who launched the Android app while in a logged-out state. -Cause: Following a server update, an issue occurred where the app entered an unexpected path during the launch process while not logged in. -Resolution: Reverted the server to its previous state and updated the Android app. -Measures to Prevent Recurrence: Strengthen pre-deployment testing for scenarios that could cause the app to freeze or hang during the launch process.

Temporary issue has been detected with AI ALF feature. We are currently investigating the cause and will provide updates shortly. We apologize for any inconvenience this may cause.

The temporary issue has been resolved, and the service is now fully operational. We apologize for any inconvenience this may have caused. - Incident Time: April 22, 2026 21:30 ~ April 23, 2026 12:00 (KST) (approximately 14 hours 30 minutes) - Affected Scope: User chats newly created during the incident in which tasks were triggered - Cause: A temporary disruption occurred after a change in a specific processing logic of the system - Resolution: The affected logic has been rolled back, and internal inspection procedures have been strengthened to prevent recurrence. We will continue to strengthen monitoring to prevent similar issues from occurring again.

Temporary issue has been detected with AI ALF feature. We are currently investigating the cause and will provide updates shortly. We apologize for any inconvenience this may cause.

The temporary issue has been resolved, and the service is now fully operational. We apologize for any inconvenience this may have caused. - Incident Time: April 20, 2026 10:25:42 ~ 11:08:41 (KST) (approximately 43 minutes) - Affected Scope: All channels 1) Some real-time chat responses were delayed compared to usual 2) The ALF Preview feature was temporarily not functioning properly - Cause: A temporary disruption occurred in some servers supporting the system under specific conditions - Resolution: Completed emergency server inspection and restoration. Preventive measures have been implemented to avoid recurrence. All systems have been fully restored to normal operation following the emergency inspection. We will continue to strengthen monitoring to prevent similar issues from occurring again.

We would like to inform you of an issue where the unread message count in the inbox was inconsistent and inbox loading was slower than usual. All necessary actions have been completed, and the service is now operating normally. - Incident Details: 1) The “unread message count” displayed on the right side of the inbox did not match the actual number 2) Loading speed of the inbox list was slower than usual - Incident Time: April 20, 2026 09:55 ~ 10:39 (approximately 44 minutes) - Cause: A temporary surge in load occurred in a specific segment of the server managing inbox data - Resolution: The overloaded server segment was distributed to balance the load, and all systems have now been fully restored to normal operation.

Temporary issue has been detected with AI ALF feature. We are currently investigating the cause and will provide updates shortly. We apologize for any inconvenience this may cause.

We are continuing to investigate this issue.

A fix has been implemented and we are monitoring the results.

The temporary issue has been resolved, and the service is now fully operational. We apologize for any inconvenience this may have caused. - Incident Time: 2026. 04. 08 15:10 ~ 2026. 04. 08 15:25 (KST) - Affected Scope: Some channels using a specific LLM model (Claude) for ALF responses (List of channels attached below). - Cause: Service error (overload) from the provider of the specific LLM model. - Resolution: Switched to using a different LLM model for generating responses. - Measures to Prevent Recurrence: Plan to reduce the time required to switch LLM models in the event of a service error from the LLM provider.

The temporary issue has been resolved, and the service is now fully operational. We apologize for any inconvenience this may have caused. -Incident Time 1: 2026. 04. 01 00:00 ~ 2026. 04. 01 00:30 (KST) -Incident Time 2: 2026. 04. 01 08:30 ~ 2026. 04. 01 11:30 (KST) -Affected Scope: In some channels, ALF sent error messages instead of responses, after which the chats were transferred to managers. -Cause: System load on the external service (LLM provider) used for generating ALF responses. -Resolution: Swapped to an alternative LLM service model. -Measures to Prevent Recurrence: Establish an automatic fallback system to ensure response continuity by switching to alternative models or services if a specific provider encounters an issue.

Temporary issue has been detected with Customer Support (Inbox) feature. We are currently investigating the cause and will provide updates shortly. We apologize for any inconvenience this may cause. Please note that there is an issue where the 'Send' button in the inbox input field may appear disabled in certain situations. However, messages can still be sent by pressing the 'Enter' key.

The temporary issue has been resolved, and the service is now fully operational. We apologize for any inconvenience this may have caused. Incident Time: 2026. 03. 19 19:41 ~ 2026. 03. 19 21:55 (KST) Affected Scope: An issue where the 'Send' button fails to activate when composing response messages in certain user chats. Cause: While disabling the 'Send' button due to configuration errors in email user chats, regular user chats (non-email) were inadvertently affected. Resolution: Modified the logic to perform the 'Send' button disable check only within the email editor.

Temporary issue has been detected with AI ALF feature. We are currently investigating the cause and will provide updates shortly. We apologize for any inconvenience this may cause. The temporary issue has been resolved, and the service is now fully operational. We apologize for any inconvenience this may have caused. -Incident Time: 2026. 03. 18 14:00 ~ 2026. 03. 23 16:00 (KST) -Affected Scope: ALF providing inaccurate responses due to failure to correctly reference linked knowledge. -Cause: An issue where certain reference information was missing in the RAG (Retrieval-Augmented Generation) logic. -Resolution: Emergency fix and deployment completed after identifying the cause; service has been restored to normal.

The temporary issue has been resolved, and the service is now fully operational. We apologize for any inconvenience this may have caused. Issue 1: Response disruption due to Claude model quota exhaustion - Incident Time: 2026.03.13 11:11 ~ 11:18 (approximately 7 minutes) - Affected Scope: Some channels using ALF - Cause: ALF responses were not functioning properly in certain channels due to the exhaustion of the allocated quota for the Claude model. Issue 2: Service disruption caused by configuration error during model deactivation - Incident Time: 2026.03.13 11:18 ~ 11:40 (approximately 22 minutes) - Affected Scope: All channels using the ALF response feature - Cause: While disabling the Claude model for specific channels to resolve Issue 1, a configuration error caused the model setting to be applied across all channels, resulting in response disruptions. Issue 3: Response disruption due to Claude model usage control configuration - Incident Time:2026.03.13 15:16:20 ~ 15:22:50 (KST) - Affected Scope: Some channels using ALF - Cause: Due to an issue with the vendor-side control configuration for Claude model calls, ALF responses were not functioning properly in certain channels. The issue has now been resolved, and the service has been operating normally since 11:40. We sincerely apologize for the inconvenience caused.