Salesforce's Data 360 feature started degrading at 19:21 UTC on 16 November, and the fault sits on Salesforce's side, not yours. Batch transforms and AI functions that rely on tenant-based models are affected across 18 listed components, and the incident is still open with no resolution time given.
Started
19:21 UTC
Duration
Ongoing
Source
IsDown
Next time
Salesforce is degraded. Is your product working?
An incident on their side does not always mean an outage on yours, and the only way to know is to already be watching your own endpoints. Set that up in under a minute and you will be notified the next time your own site is affected. And there will be a next time.
50 monitors for your own websites. No card.
Affected components: CS253, NA238, USA870S, AUS24S, AUS88, CS272, FRA88, ITA18, NA236, SWE24, USA1, USA250S, USA252S, USA254S, USA542S, USA566, USA872S, USA874S
What happened?
Salesforce logged this as a Feature Degradation starting 19:21 UTC, and its only update so far came three minutes later at 19:24 UTC. That update says customers may be unable to run batch transforms, and that AI functions relying on tenant-based models may fail because model downloads are not completing, including Extension Functions. Salesforce named a recent release as a possible trigger and said it was preparing a roll back, with a further update promised in 60 minutes or sooner. No second update has landed since, so the incident is still open at 19:24 UTC, the last point Salesforce published anything. The affected component list spans 18 entries, including CS253, NA238, USA870S, AUS24S, AUS88, CS272, FRA88, ITA18, NA236, SWE24, USA1 and USA250S, a spread that covers instances across several regions rather than a single data centre. No user reports came in during this window, which for a feature-level degradation like this is not unusual, since the symptoms (failed batch jobs, missing AI model downloads) often surface quietly inside a workflow rather than as an obvious front-end break. No post-mortem has been published, and Salesforce has not confirmed the release rollback actually happened. Over the past 90 days Salesforce has logged 121 incidents with a median duration of 2 hours 3 minutes, and across its history on record it has had 2,522 incidents. One update from Salesforce, naming a possible cause and a plan, is what the record holds right now, nothing more.
Learning
This incident sits in the layer of Salesforce's platform that handles batch processing and AI model downloads for Data 360, not in the basic request and response path that most external monitoring checks. That is why a simple uptime check against your own Salesforce-connected endpoints could stay green through this whole thing: the front door is open, but a specific feature behind it is failing to complete its job. Watching your own endpoints tells you whether your product can reach Salesforce, not whether Salesforce's internal processing for a given feature is actually finishing. The shape of this failure, background jobs silently not completing, is the kind of fault that often goes unnoticed until a customer asks why a transform never ran. Over the last 30 days, UptimeRobot recorded 51 incidents tied to this provider with a median duration of 1h 40m, so stoppages like this are a recurring part of using the platform, not a one-off. Your monitoring covers your half, UptimeRobot now watches the provider's half too, so next time you know which side broke without guessing.