Instructure Data Breach & Product Outage
Affected: Canvas
Edlink Outage History
Edlink is up right nowEdlink had 8 outages in the last 2 years totaling 378h 21m of downtime — averaging 0.3 incidents per month.
There were 8 Edlink outages since November 12, 2024 totaling 378h 21m of downtime. Each is summarised below — incident details, duration, and resolution information.
Degraded API Performance
Affected: Edlink APIEdlink Dashboard
Timeline · 4 updates
Meta API request timeouts
Timeline · 1 update
- resolved Mar 02, 2026, 08:41 PM UTC
For 172 seconds, our meta API (including SSO endpoints) were unresponsive. This incident took place around 2:30pm CT. The incident was caused by a lock on a table related to single sign on. The lock was the result of a database maintenance query that we were running at the time. The query was expected to complete in less than one second (and therefore, not cause any service disruption), but maintained the lock longer than expected. We terminated the execution of the query approximately two minutes later.
Schoology API Elevated Error Rates
Affected: Schoology
Timeline · 8 updates
- investigating Dec 09, 2025, 05:41 PM UTC
We are currently investigating reports of an unusual number of School API errors. This may affect any Schoology integrations, including ones where you are only accessing class roster data via Edlink. We are receiving random (intermittent) 403 errors from AWS Cloudfront (Schoology's CDN provider). These errors do not actually correlate to permissions issues in Schoology as they appear to be random and will succeed after several retries or a significant time delay. Our system, by default, replays external requests up to 5 times, but they are in relatively short succession so this retry logic does not provide a workaround in this situation. Roster data is all still available (the Edlink API itself still returns data as usual), but Edlink may not receive subsequent roster updates from Schoology during this time. We are attempting to work with Schoology to help them determine the root cause.
- identified Dec 11, 2025, 09:14 PM UTC
We have a deployed an update to our roster sync code to try to workaround this ongoing issue. This should help roster data begin to flow again. This change does not impact the assignments, grades, and coursework APIs, so you may still experience random, intermittent 403 errors from Schoology.
- identified Dec 16, 2025, 02:56 PM UTC
The issue with the Schoology API has been escalated to "major". At this point, it appears that close to 100% of API requests to Schoology are failing. This includes SSO, syncs, coursework, and grade requests. We are temporarily pausing all Schoology syncs in order to free up queue space for other sources.
- monitoring Dec 17, 2025, 01:10 AM UTC
We have noticed some API requests begin to succeed again. We are not sure if this is temporary (or if it applies to 100% of API requests) but we are monitoring the situation.
- monitoring Dec 17, 2025, 04:53 PM UTC
Everything appears to be back to normal, we're waiting on a few more districts to sync.
- monitoring Dec 17, 2025, 10:21 PM UTC
...and we're back to seeing elevated error rates again after a relatively smooth morning and early afternoon.
- monitoring Dec 22, 2025, 03:37 PM UTC
Error rates have dropped again and we are begging to resume syncs. Last word from Powerschool is that they are "looking into it" (as of last Friday).
- resolved Dec 23, 2025, 05:14 AM UTC
Although we have not heard any official messages from Schoology, it appears that the API issues have been resolved on their end. Sync have resumed.
Database Write Limitation
Affected: Edlink APIEdlink Dashboard
Timeline · 3 updates
Google Cloud Storage Outage
Affected: Edlink APIEdlink DashboardGoogle Cloud Platform Google Cloud StorageSFTP
Timeline · 4 updates
- identified Jun 12, 2025, 06:59 PM UTC
Currently Google Cloud Storage is down globally. This is an insanely rare event. Edlink's API is currently functional, but some requests may fail. The SFTP transfer (Edlink's SFTP servers) are currently down. So new roster data can't be transferred into Edlink via SFTP. Existing roster data is unaffected. Functions of the Edlink Dashboard that involve SFTP (i.e. CSV mapping) may be affected. In some instances, when loading the Administrator Onboarding Flow, Edlink's clients' logos / icons may not load.
- identified Jun 12, 2025, 07:00 PM UTC
Here's a link to the Google Cloud page showing their status: https://status.cloud.google.com/regional/americas
- monitoring Jun 12, 2025, 08:40 PM UTC
A fix has been implemented by Google Cloud and we are monitoring the results. We are now able to connect to the Edlink SFTP servers and API requests are stable.
- resolved Jun 12, 2025, 10:30 PM UTC
Google services are back up and running.
Dashboard not loading ~10 minutes
Timeline · 1 update
- resolved Mar 05, 2025, 10:41 PM UTC
Issue: - We updated our Kubernetes HTTP route configuration at approximately 4:16pm CT. - This route configuration was slightly out of date and contained an invalid "catchall" route that used to route traffic to the Edlink Dashboard. - The issue was identified and fixed at approximately 4:26pm CT. Impact: - This issue only affected the Edlink Dashboard, not our API, SSO, Widgets, Upload, Data Pipeline or other related services. - It was likely not noticed by many users as it would only have impacted people who were attempting to load the dashboard in a new tab or window. - If you already had a dashboard page open, it would have continued to work uninterrupted (as our API was unaffected).
Edlink Dashboard Not Responding
Affected: Edlink Dashboard
Timeline · 2 updates
- investigating Nov 12, 2024, 12:46 AM UTC
Right now the Edlink dashboard is not responding. It is returning a message that says "no healthy upstream." We're currently investigating the issue. Edlink's API is not affected. SSO is not affected - all end-users should be able to launch and sign in as normal. Only users attempting to interact with Edlink via the dashboard (i.e NOT teachers or students) are potentially affected, as of now, that we know of.
- resolved Nov 12, 2024, 12:47 AM UTC
The incident has been resolved. We will issue a full post-mortem writeup during business hours tomorrow (10am CST - 6pm CST).