28 Knock App incidents ยท 2025๋ 12์ โ official updates, affected components, duration and resolution details.
์ฐ๋ฆฌ๋ ํ์ฌ ์ฐ๋ฆฌ์ ์๋ฃ ์ฐฝ๊ณ ์์ถ์ ์ง์ฐ์ ํฌ์ํ๊ณ ์์ต๋๋ค. ์ฐ๋ฆฌ๋ ๊ณง ์ ๋ฐ์ดํธ๋ฅผ ๊ฒ์ํฉ๋๋ค.
์ด ์ฌ๊ฑด์ ์ง๊ธ ํด๊ฒฐ๋๊ณ ์๋น์ค๋ ์ง๊ธ ์์ ํ ๊ฐ๋์ ๋๋ค.
๊ณต์ ์ธ์๋ํธ ์ ๋ฐ์ดํธ๋ฅผ ์๋ ๋ฒ์ญํ์ต๋๋ค.
Dashboard ๋ฐ V1 API์ ์ํฌํ๋ก์ฐ ์คํ ๋ก๊ทธ์ ๊ฐ์ฉ์ฑ์ ์กฐ์ฌํ๊ณ ์์ต๋๋ค. ํ์ฌ ์ฐ๋ฆฌ๋ ์ผ๋ถ ์คํ ๋ก๊ทธ์ ๋ํ ์ต๋ 18 ๋ถ์ ์ง์ฐ์ ๋ณผ ์ ์์ต๋๋ค.
์ด ์ฌ๊ฑด์ ์ง๊ธ ํด๊ฒฐ๋๊ณ ์๋น์ค๋ ์ง๊ธ ์์ ํ ๊ฐ๋์ ๋๋ค.
๊ณต์ ์ธ์๋ํธ ์ ๋ฐ์ดํธ๋ฅผ ์๋ ๋ฒ์ญํ์ต๋๋ค.
์ฐ๋ฆฌ๋ ํ์ฌ ์ค๋ฅ ์ฒ๋ฆฌ ์๋ฆผ ์ํฌํ๋ก์ฐ๋ฅผ ์กฐ์ฌํ๊ณ ์ผ๋ถ ๊ณ ๊ฐ์๊ฒ ๋ฉ์์ง๋ฅผ ์ ์กํ๊ณ ์์ต๋๋ค.
์ฐ๋ฆฌ๋ ์์ ์ ๋ฐ์กํ๊ณ ๋ณต๊ตฌ๋ฅผ ์์ํฉ๋๋ค. ์ฐ๋ฆฌ๋ ๋ฐ์ ํ๊ฒ ๊ฐ์ํ๊ณ ์์ต๋๋ค.
Workflow ์ฒ๋ฆฌ์ ๋ฉ์์ง ๋ฉํ์ ๋ค์ ๊ฐ๋๋ฉ๋๋ค. ์ฐ๋ฆฌ๋ ๋ฐ์ ํ๊ฒ ๊ฐ์ํ๊ณ ์์ต๋๋ค.
์ด ์ฌ๊ฑด์ ์ง๊ธ ํด๊ฒฐ๋๊ณ ์๋น์ค๋ ์ง๊ธ ์์ ํ ๊ฐ๋์ ๋๋ค.
๊ณต์ ์ธ์๋ํธ ์ ๋ฐ์ดํธ๋ฅผ ์๋ ๋ฒ์ญํ์ต๋๋ค.
We're currently investigating an issue with our service. We'll have an update soon.
We've identified the issue with the service and are actively working on a fix. We're seeing workflows and message deliveries process again at the usual rate.
A fix has been implemented, and we're currently monitoring the service. All our services seem to be recovering. We're in the process of replaying any failed work that happened during the window. We expect there was no loss of workflows during this incident, but some work may have been delayed.
This incident has now been resolved and the service is now fully operational. We're replaying any jobs that may have failed during the window. All our processing queues are fully caught up. We will share a full post mortem as soon as possible.
Message data synchronization into downstream customer data warehouse destinations is delayed for messages created after 8 PM ET on June 30, 2026. We have implemented a fix. We expect the issue to be fully resolved following the nightly synchronization job scheduled on July 2nd, 2026.
This incident has now been resolved and the service is now fully operational.
We are investigating delays processing workflow run event logs.
We're still investigating delays affecting workflow run event logs.
We've completely recovered any lag on workflow logs and we are back to normal.
We're currently investigating an issue with some workflow logs being delayed.
This incident has now been resolved and the service is now fully operational.
We're currently investigating elevated error rates connecting to the V1 Websocket API. Realtime behavior for in-app feed and guides is degraded as a result.
We have shipped a fix and are seeing resolution in websocket connections. We are continuing to monitor closely.
This incident has been resolved.
We are investigating delays in ingesting workflow run logs and triggering outbound webhooks from workflow run events.
We are seeing recovery in workflow run log and webhook delivery delays and are continuing to monitor.
This incident has been resolved.
We intermittently experienced delayed workflow logs between ~09:00 AM UTC and ~01:45 PM UTC today (2026/06/15). This was caused by some internal changes in how we fan out workflow logs, which caused consumers to get temporarily backlogged. The peak delay we experienced before processing workflow logs was around 15 minutes, but we intermittently saw delays of up to 5-6 minutes. This affected 1. Logs showing up in the dashboard. 2. Workflow batch mappings. We saw complete resolution at 01:43 PM UTC. Logs are being processed within our usual window (within up to ~5s) again.
We did not experience any further issues after we put the fix in place.
We're seeing elevated response times loading some log data. This is leading to slower page loads in the Dashboard and slower responses from some V1 API endpoints.
We have shipped a fix and we are seeing recovery of log data load times. We are continuing to monitor closely before resolving.
This incident has been resolved.
We're currently investigating an issue with our service. We'll have an update soon.
A fix has been implemented, and we're currently monitoring the service.
This incident has now been resolved and the service is now fully operational.
We're currently investigating an issue related to slow notification processing and message delivery. Some workflows will be delayed
We're seeing an extremely high amount of load on our services which is causing workflow and broadcast delays. We're working to relieve the pressure immediately.
We're seeing some recovery on our services but are dealing with delays in some workflow and broadcast processing still. Our logging service is also currently delayed, so may not reflect the latest as we catch up and work through a huge influx of work.
One of our workflow queues is currently still backlogged, but we're seeing good resolution across other areas of our service and we have effectively cordoned off the source of the issue. We're burning through remaining work and are exploring ways to shed the load further to get back to stability.
We've taken steps to shuffle around work in our system which has meant that we're processing new work for all customers again. We still have some queued workflow executions that we're working through and expect to be fully caught up in the next 40 minutes.
We are seeing a recovery in all workflow and broadcast processing delays. We are monitoring closely.
All behavior remains normal, we are considering this incident resolved.
We are investigating errors resolving and redirecting tracked links.
We've shipped a fix and are seeing resolution in tracked link resolution. We are monitoring closely.
This incident has now been resolved and the service is now fully operational.
We're currently investigating an issue with delayed ingestion of our API logs, putting pressure on our ingestion pipeline and resulting in some degraded state for browsing API logs as well.
Weโve shipped a fix and weโre seeing initial signs of recovery. We are continuing to monitor closely.
We are seeing full recovery of API log delays. We are continuing to monitor before closing out this incident
This incident has now been resolved and the service is now fully operational.
We are investigating delays in creating and processing message events. Some customers may experience delays seeing message events in the Knock Dashboard. Some customers may also experience delays in outbound webhook events triggered by message events.
We've implemented a fix and are seeing message event lag begin to recover.
Webhook delivery lag has resolved. We are still waiting for message event Dashboard availability lag to recover.
All message event lag has recovered. We are considering this incident resolved.
We're currently investigating an issue with promoting workflows. We're working on a fix now. This does not impact notification delivery.
We've identified the issue with workflow promotion and are deploying a fix for it now.
The fix has been deployed and workflows are able to be promoted again.
We're currently investigating an issue with log processing. There are no issues currently processing workflows and messages, but logs are affected.
We've identified the issue with the service and are actively working on a fix.
A fix has been implemented, and we're currently monitoring the service.
This incident has now been resolved and the service is now fully operational.
We're currently investigating an issue with our service. We'll have an update soon.
A fix has been implemented, and we're currently monitoring the service.
This incident has now been resolved and the service is now fully operational.
We're currently investigating an issue with subscriptions in our service.
Our service has caught up on the backlog of subscription workflow executions.
This incident has now been resolved and the service is now fully operational.
We're currently investigating an issue with delayed workflow run logs. Workflow execution and message delivery are unaffected, but our log queues are currently behind.
Weโve shipped a fix to improve throughput for our consumer that processes workflow run logs and are beginning to see the consumer catch up on its backlog. Weโre continuing to monitor closely.
Workflow run logs are no longer delayed. We are monitoring closely before closing this out.
This incident has now been resolved and the service is now fully operational.