- Sources: primary, active incidents API, discussion
- Summary: Salesforce opened a major service disruption, externalId 95390343, with impact starting 2026-09-16 07:50 UTC, described in the 08:45 UTC update as affecting multiple instances across all regions, and customers reported severe delays, intermittent errors and an inability to open support cases. The 14:00 UTC update states the impact radius is narrower than initially understood and that instances which have not experienced impact will be removed from the incident, and the active-incidents record carries
affectsAll false with an enumerated instance list. A fleetwide fix began rolling out region by region from 10:56 UTC, with manual restarts where it did not fully apply, and the incident was still active at 14:27 UTC with GovCloud stated out of impact, the rootCause field null and no published root cause. Every figure here was read from the active-incidents endpoint while the incident was open. That endpoint lists only open incidents and has since dropped externalId 95390343, and the per-incident path for that id returns HTTP 404, so the timeline and the instance list cannot be re-checked there. - Why it matters: The only mechanism Salesforce has named is a shared login path, given at 09:10 UTC as requests stalling while waiting on a response from an internal login service, at 09:57 UTC as a suspected external dependency failure affecting the legacy login server, and at 10:18 UTC as a core system component under increased load, so what a reader can act on is a vendor-named dependency rather than a cause.
- Follow-up: Track the resolution time, the final instance list and the root cause Salesforce publishes.
send feedback on this story