Salesforce Outage History

We’re investigating the cause of the issue. We will provide an update in 30 minutes or sooner if we have more information.

We’re aware of an issue affecting .mil network connectivity to some Salesforce GovCloud environments. Customers using affected network paths may be unable to access Salesforce services or may experience intermittent connectivity. We’ve engaged the global infrastructure provider for a subset of government agencies to investigate the issue further and are monitoring for recovery. We’ll provide another update when there is a significant development.

We’re investigating an issue affecting Data Cloud streaming ingestion in a subset of environments. Customers may experience delays or failures when ingesting data through the Ingestion API, Web and Mobile Connector, or MCP Connector. We’ve identified service startup failures in some affected environments and are evaluating targeted restarts and rollback options to restore normal ingestion. We’ll provide an update in 30 minutes or sooner if additional information becomes available.

We’re continuing to investigate the Data Cloud streaming ingestion issue and are moving forward with a rollback in one environment to validate recovery before expanding the action. In parallel, we’re investigating why the service is failing to start normally. We’ll provide another update in 30 minutes or sooner if additional information becomes available.

We’ve started the rollback in an initial affected environment and are seeing early signs of recovery, though service capacity has not yet returned to expected levels. We’ve narrowed the investigation to a small number of production environments and are manually increasing capacity while validating the rollback. Customers on affected environments may continue to experience delays or failures when ingesting data through the Ingestion API, Web and Mobile Connector, or MCP Connector. We’ll provide another update in 30 minutes or sooner if additional information becomes available.

We’ve successfully increased service capacity in one of the affected environments, and initial validation confirms the additional service instances are active. We’re applying the same targeted action to the remaining affected environments while addressing the issue that prevented services from scaling automatically. Customers on affected environments may continue to experience delays or failures when ingesting data through the Ingestion API, Web and Mobile Connector, or MCP Connector. We’ll provide another update in 30 minutes or sooner if additional information becomes available.

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.

We’re aware of an issue affecting Log Center on a subset of PODs, where customers may experience missing application, deployment, and Ecom logs. The impact began at 02:36 UTC and may affect log monitoring, alerting, and third-party log streaming. We are actively investigating the issue. We’ll provide an update in 30 minutes or sooner if additional information becomes available.

We’ve confirmed that the log processing service is operating normally and that a recent planned application update is not related to the issue. Our teams are restarting the affected log collection component to restore full log visibility, beginning with one impacted instance. We’ll continue to monitor recovery and provide an update in 30 minutes or sooner if additional information becomes available.

We’ve restarted the affected log collection components and confirmed that log activity has returned to normal on those instances. We identified an additional impacted instance, completed the restart, and confirmed recovery there as well. We’re continuing to review other instances flagged by monitoring and will restart and validate them as needed. We’ll provide another update in 30 minutes or sooner if additional information becomes available.

We're investigating SSO login failures where customers logging in via single sign-on (SSO), specifically to Anypoint Platform, are unable to authenticate. The impact spans the platform for anyone using SSO login. We've identified a recent change that's preventing customers whose org's stored Identity Provider (IDP) signing certificate isn't formatted as Privacy-Enhanced Mail (PEM) from logging into access management. Users who are already logged in are unaffected and won't lose access or experience disruptions. We're executing a rollback to resolve the ongoing login issues. We'll provide an update in 30 minutes or sooner if additional information becomes available.

We're continuing to work on a rollback to resolve the single sign-on (SSO) login issue impacting Anypoint Platform. The rollback is in progress for one of the regions and we're checking to see if this resolves the issue. Users who are already logged in are unaffected and won't lose access or experience disruptions. We'll provide an update in 60 minutes or sooner if additional information becomes available.

The rollback in the US region is complete without any errors. The rollback is in progress for the remaining impacted regions and should be completed within 60 minutes. Users who are already logged in, continue to remain unaffected and won't lose access or experience any disruptions. We'll provide an update in 30 minutes or sooner if additional information becomes available.

Recurrence of the earlier US2 incident — same root cause. The US2 Redis cache is again at its memory limit and rejecting writes. Impact:logins disruption

We're investigating an issue impacting login access to Stack 401 for Marketing Cloud Engagement. Affected users are unable to log in to this Stack. We've engaged multiple engineering teams to review database and caching services. We'll provide an update in 30 minutes or sooner if additional information becomes available.

As we continue to investigate the login issue impacting Stack 401 in Marketing Cloud Engagement, we identified timeout errors on login token requests and are reviewing the health of the caching and database services connected to this Stack. We’ve also identified a high volume of login requests originating from a small number of integrations. We're implementing rate limits on these requests to see if this resolves the issue. We'll provide an update in 30 minutes or sooner if additional information becomes available.

We've identified the specific integration that's generating a high volume of login requests to Stack 401 in Marketing Cloud Engagement and have applied rate limits to those requests. We’re validating the changes and monitoring login access. We've also confirmed there are no underlying database issues on the Stack. We'll provide another update within 30 minutes, or sooner if new information becomes available.

At 13:31 UTC on September 24, 2026, some Common Runtime customers began experiencing intermittent loss of application logs in the US region. Our engineers are working to resolve the issue.

Engineers are continuing to investigate this issue. The current known impact is intermittent router logs being dropped in a subset of US regions. App logs are unaffected. We will provide another update in the next 60 mins.

Engineers have fixed the issue and are monitoring. All router logs are now being successfully delivered.

We’re investigating the cause of the issue. We will provide an update in 30 minutes or sooner if we have more information.

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.

We're aware of an issue impacting performance where some customers may experience slower response times or delays when using core services on these instances. We have taken steps to increase system capacity on the impacted instances as we continue to investigate the underlying trigger. We'll provide an update in 30 minutes or sooner if additional information becomes available.

Some customers may still experience slower response times or delays when using core services on the impacted instances. We've identified the trigger of this issue as increased database session activity, which is driving elevated cache response times. We're remapping the impacted instances to a newly scaled cache cluster to help restore normal performance, and we're closely monitoring the situation. We'll provide an update in 30 minutes or sooner if additional information becomes available.

We've started remapping to the scaled cache cluster, and the change is still deploying. We expect full propagation across all impacted instances to take up to 60 minutes, after which traffic should stabilize. Some customers may continue to experience slower response times while this rollout completes. We'll provide an update in 30 minutes or sooner if additional information becomes available.

We're continuing to investigate the issue affecting Data Cloud activations for customers in the US-east-1 region on our Prod1-CDP1 environment. We've confirmed the issue stems from a third-party cloud storage service, and we're waiting on their team to trace the cause using system logs. Our engineers are preparing alternative options to reduce impact while this investigation continues. We'll provide an update if additional information becomes available.

We've identified that the issue affecting Data Cloud activations stems from a third-party cloud storage service used in our Prod1 CDP1 environment. We applied a configuration change to that service to help address the errors, and we're preparing an alternate storage option as an additional measure. We're continuing to work directly with the third-party provider and monitoring for improvement. We'll provide an update if additional information becomes available.

We performed further review and analysis and noted that a few other instances may also be impacted as a result of this incident. We have added those instances to the list. Meanwhile, we're continuing to work directly with the third-party provider and monitoring for improvement. We'll provide an update if additional information becomes available.

We're investigating an issue impacting a database supporting Marketing Cloud Engagement, and customers on the impacted database are currently unable to connect. Our investigation points to a potential file lock as the trigger. We've engaged additional support and are preparing a secondary environment in case it is needed. We'll provide an update in 30 minutes or sooner if additional information becomes available.

We're continuing to investigate a disruption impacting a database supporting Marketing Cloud Engagement, and affected customers remain unable to connect. We've identified that a storage-level lock is likely preventing the database from completing recovery, and our engineering teams are working to resolve it. We'll provide an update in 30 minutes or sooner if additional information becomes available.

We've ruled out a storage-level file lock as a factor in the ongoing recovery issue impacting Marketing Cloud Engagement. Affected customers remain unable to connect to services on the impacted database. We've started taking the database offline and bringing it back online in an effort to complete recovery. In parallel, we're preparing a secondary database as a backup option. We'll provide an update in 30 minutes or sooner if additional information becomes available.

We've performed an in-place restart of the impacted Marketing Cloud Engagement database which has moved into a recovery state. Affected customers remain unable to connect to services while recovery completes. We're monitoring progress closely. In parallel, we're preparing a secondary database as a backup option. We'll provide an update in 30 minutes or sooner if additional information becomes available.

Affected customers remain unable to connect to the Marketing Cloud Engagement database affected by this incident. We're now evaluating whether to continue recovery efforts on this database or move forward to our secondary database, which is fully prepared. We'll provide an update in 30 minutes or sooner if additional information becomes available.

We continue to work on restoring the Marketing Cloud Engagement database. A replacement of a failed component on the storage array has been planned, and is expected to take a few hours. The replacement will run parallel to our recovery work without affecting it. We'll provide an update in 30 minutes or sooner if additional information becomes available.

The Salesforce Technology team has completed its investigation and determined that the listed instances and services remained healthy and did not experience impact during the times outlined. We apologize for any inconvenience caused by the false positive notification, and we will undertake a full investigation as we look to improve our impact alerting and assessment configurations.

We are investigating a potential cause of the issue.

We’ve confirmed that the solution implemented to mitigate the issue causing the feature degradation was successful. Engineers have observed that the service is now stable and have declared the incident resolved. 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.

We’re investigating an issue affecting Agentforce session scoring and tagging functionality for a subset of customers. Our teams are actively working to identify the trigger and restore service. We’ll provide an update in 30 minutes or sooner if additional information becomes available.

We’re continuing to investigate an issue affecting Agentforce session scoring and tagging for a subset of customers. Our teams are analyzing processing failures and reviewing affected customer environments to determine the trigger and identify a path to restore functionality. We’ll provide an update in 30 minutes or sooner if additional information becomes available.

Our teams have narrowed the investigation to failures occurring during session processing and are working to identify a path to restore functionality. We’ll provide an update in 60 minutes or sooner if additional information becomes available.

We have identified a potential fix and are preparing to apply it to affected customer environments. We’ll monitor session tagging performance following the change to determine whether functionality recovers while our investigation into the underlying issue continues. We’ll provide another update as significant progress is made.

We're aware of the issue affecting credit card transactions on Commerce Cloud, where customers are unable to complete purchases using credit cards at checkout. We're restarting the affected applications to restore transaction processing and working on a broader fix to resolve this for all impacted customers. We'll provide an update in 30 minutes or sooner if additional information becomes available.

Upon further investigation, we've determined that the start time of impact is different than initially understood. We've 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.

We've applied manual fixes across all known affected environments to restore credit card transaction processing on Commerce Cloud checkouts. We're now testing an automated, fleet-wide fix to fully resolve this issue. We'll provide an update in 30 minutes or sooner if additional information becomes available.

The Salesforce Technology team has identified a potential trigger for the issue and is working on implementing a fix to resolve it. Thank you for your patience as we work to resolve this issue.

We isolated the issue and are working to find the best path to resolution. We will provide an update in 30 minutes or sooner if we have more information.

We continue to explore multiple paths to quickly resolve the issue, which include analysing the error logs and additional diagnostic information to determine the potential trigger. We’ll provide an update upon a significant update, or sooner if additional information becomes available.