Salesforce favicon

Salesforce outage on 2026-10-21

salesforce.com
October 2026 21 Minor incident

Salesforce's Agentforce Coworker feature broke for 36 minutes on the morning of 21 October 2026. Salesforce traced it to a third-party service provider on their side, not to anything in your setup.

Started

02:57 UTC

Duration

Lasted 36m

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: JPN10S, JPN12S, JPN132, JPN134, JPN136, JPN138, JPN140, JPN142, JPN144, JPN146, JPN18S, JPN20S, JPN22S, JPN24S, JPN28S, JPN2S, JPN4S, JPN6S, JPN8S

What happened?

Salesforce logged one incident on 21 October 2026, starting at 02:57 UTC and lasting 36 minutes. The title given was "Feature Degradation", affecting a set of Japan-region instances: JPN2S, JPN4S, JPN6S, JPN8S, JPN10S, JPN12S, JPN18S, JPN20S, JPN22S, JPN24S, JPN28S and JPN132, twelve of the 19 components Salesforce tracks. The symptom was specific: Agentforce Coworker conversations could stop partway through an answer with no error message shown, and a subset of customers saw intermittent request failures. Salesforce's first update, at 03:11 UTC, said they had engaged a third-party service provider and were seeing error rates decrease, with a further update promised within 30 minutes. A second update at 03:25 UTC was a start time revision, pushing back when Salesforce believes impact actually began, with no further explanation offered. No further updates were filed after that, and no user reports came in from your side during this window. Salesforce's own incident history for the past 90 days shows 110 incidents with a median duration of 2 hours 25 minutes, so this one resolved faster than typical. No post-mortem has been published, and the cause beyond "third-party service provider" is not on the record.

Learning

This incident sat inside Agentforce Coworker, a feature layer that depends on a provider Salesforce doesn't name, and the failure mode, answers silently cutting off mid-conversation, is the kind of thing an uptime check on your own endpoints will never catch because your systems were never down. If you were only watching your own product surface for errors, you'd have seen nothing wrong, because nothing in your stack was broken. The fault lived entirely on Salesforce's side, in a dependency they engaged directly to bring error rates down. Over the last 30 days UptimeRobot recorded 37 incidents against this provider with a median duration of 1h 53m, so short, unexplained gaps like this one are a recurring part of the pattern, 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.

4.7
stars out of 5
284+ reviews on

Start monitoring in 30 seconds.

There's nothing to install. No credit card required. 50 monitors for free.