Changes / Trello / 500 Errors being experienced by Trello users
Trello: Critical incident: 500 Errors being experienced by Trello users
Summary
Trello reported a critical disruption starting 2025-12-12 06:40 UTC and lasting 1 h 1 min, with 500 errors and service disruptions affecting web and mobile users. The postmortem attributed errors and slow loading to a dependency library upgrade causing high resource usage; rolling back the library restored service.
Machine-written from the diff below (Sol; type: incident outage). Check the diff before relying on it.
What changed
Before: not present
After: 500 Errors being experienced by Trello users
Before: not present
After: critical
Before: not present
After: postmortem
Before: not present
After: 2025-12-12 06:40 UTC
Before: not present
After: 2025-12-12 07:41 UTC
Before: not present
After: API, Atlassian Support Knowledge Base, Atlassian Support - Support Portal, Atlassian Support Ticketing, Trello.com
Before: not present
After: We are actively investigating reports of a service disruption affecting Trello users. We'll share updates here as they become available or within the hour.
Before: not present
After: Our engineering team has implemented a fix, and services have now recovered. We will continue to closely monitor the situation.
Before: not present
After: On December 12, 2025, from 05:51 UTC to 08:09 UTC, Trello users may have experienced service disruptions on the web page and mobile apps. The issue has now been resolved, and the service is operating normally for all affected customers.
Before: not present
After: ### Summary
On December 12, 2025, between 05:31 UTC and 07:11 UTC, Trello customers experienced errors or slow loading times due to a dependency library upgrade that caused high resource usage. The incident was detected within 1 minute by the automated monitoring system and mitigated by rolling back the version of the dependency library which put Atlassian systems into a known good state. The total time to resolution was 1 hour and 41 minutes.
### **IMPACT**
The issue caused service disruption for all Trello customers on December 12, 2025 between 05:31 UTC and 07:11 UTC. Users experienced degraded service between 05:31 UTC and 06:32 UTC, and most users were unable to load Trello between 06:32 UTC to 07:07 UTC.
### **ROOT CAUSE**
The issue was caused by an upgrade to a dependency library. As a result, there was an increase in system memory consumption causing startups to take longer than expected. When the startup time exceeded the allowed threshold, the worker process was flagged as unresponsive and restarted as an automated recovery action. Because every worker on the instance failed to startup within the expected time, the instance entered a startup loop that required intervention.
### **REMEDIAL ACTIONS PLAN & NEXT STEPS**
We know that outages impact your productivity. While we have several testing and preventive processes in place, we didn’t anticipate that the upgrade would consume more memory, which ultimately caused the startup failure loop and service disruption.
We are prioritizing the following improvement actions designed to avoid this issue in the future:
* Review our automated detection of unresponsive workers and adjust the detection thresholds during initialization.
* Improve monitoring during workers startup and restarts due to unresponsiveness.
* Enhance the stricter standard process for confirming memory/other data processing resource requirements for planned updates.
We apologize to customers whose services were impacted during this incident; we are taking immediate steps to improve the platform’s performance and availability.
Thanks,
Atlassian Customer Support
First version of this incident archived by Rewindex. View the live incident (opens in new tab).
History of this incident
| When (UTC) | Change | Kind |
|---|---|---|
| Critical incident: 500 Errors being experienced by Trello users (this change) Trello reported a critical disruption starting 2025-12-12 06:40 UTC and lasting 1 h 1 min, with 500 errors and service disruptions affecting web and mobile users. The postmortem attributed errors and slow loading to a dependency library upgrade causing high resource usage; rolling back the library restored service. | Major incident |