Changes / Salesforce / Agentforce Service Disruption: Agentforce
Salesforce: Major incident: Agentforce Service Disruption: Agentforce
Summary
Summary pending. The diff below shows exactly what changed.
Major incidentLabel: outage (model (Jev), 97%)Changed Detected Source: status.salesforce.com (opens in new tab)
Incident page: full timeline of updates →
What changed
Before: not present
After: Agentforce Service Disruption: Agentforce
Before: not present
After: major
Before: not present
After: resolved
Before: not present
After: 2025-10-09 03:18 UTC
Before: not present
After: 2025-10-11 08:09 UTC
Before: not present
After: 4 instances, Agentforce
Before: not present
After: On October 9, at 3:18 UTC, we noted a service disruption impacting Agentforce on multiple instances. Customers may observe that OpenAI permission is currently disabled, which appears to be preventing LLMs, such as Omni and Omni Mini, from loading correctly in Prompt Builder and Agent Builder, and from accessing the Agentforce services. We’re actively exploring multiple strategies to mitigate the issue and restore services as quickly as possible.
We’ll provide an update in 60 minutes or sooner if additional information becomes available.
Before: not present
After: During investigations, we established that the level of impact differs from what was initially understood. We have updated the incident posting to reflect the current impact symptoms accurately.
Before: not present
After: We continue to investigate a potential trigger and explore strategies to mitigate this issue.
Thank you for your patience as we work on resolving this issue. We’ll provide an update in 60 minutes or sooner if additional information becomes available.
Before: not present
After: We’ve identified a recent code change as the potential trigger and are exploring solutions to resolve the issue.
We’ll provide an update in 60 minutes or sooner if additional information becomes available.
Before: not present
After: Investigations have highlighted that the impact radius is broader than initially understood. We have updated the posting to reflect the additional instances experiencing impact, and customers in these instances will receive communications related to this incident going forward.
Before: not present
After: We’ve confirmed the recent code change as the trigger and are preparing an emergency release to fix this issue.
Thank you for your patience as we prepare the fix. We’ll provide an update in 60 minutes or sooner if additional information becomes available.
Before: not present
After: We’re finalizing our preparations for the emergency release before implementing the fix.
Thank you for your patience as we finalize preparations for the fix. We’ll provide an update in 60 minutes or sooner if additional information becomes available.
Before: not present
After: Upon further investigation, the Salesforce Technology team has determined that the start time of impact is different than initially understood. We have now revised the start time of the Trust post to more accurately reflect the time customers may have begun to experience impact. We apologize for any confusion caused by this.
Before: not present
After: We have initiated the emergency release and it will take approximately 7 hours to implement and test the fix. We will monitor the release as it progresses.
Thank you for your patience as we implement the fix. We’ll provide an update by 08:00 UTC or sooner if additional information on the release becomes available.
Before: not present
After: The fix was successfully deployed and validated on USA9018S. We are continuing to deploy it on the remaining instances.
We will provide an update once the deployment is complete or sooner if additional information becomes available.
Before: not present
After: We’ve confirmed that the solution implemented to mitigate the issue causing the service disruption was successful. Engineers have observed that the service is stable and have resolved the incident. Customers should no longer see any impact. We will fully investigate the incident, confirming the technical trigger, the underlying cause, and preventive action to avoid a repeat in the future. We apologize for how you and your business may have been affected by this incident.
First version of this incident archived by Rewindex. View the live incident (opens in new tab).
History of this incident
| When (UTC) | Change | Kind |
|---|---|---|
| Major incident: Agentforce Service Disruption: Agentforce (this change) | Major incident |
All Salesforce changes → · Salesforce profile · Markdown · JSON