The status page that cried 'resolved': premature all-clears during outages

Status pages exist so users and administrators can trust an authoritative source about whether a service is up. During Dropbox's worldwide outage on 30 August 2015, that trust was undermined: the company's status page indicated that service had returned to normal even as numerous customers reported that they remained unable to access the cloud-storage service and their files. The Register, covering the outage as it unfolded, put the gap plainly: "a quick scan of 'Dropbox down' tweets told a different story, with many customers moaning that they remained locked out" even after the company had said it was working the issue.

The premature 'all clear' meant that people trying to diagnose why their files would not load were told by Dropbox's own authoritative dashboard that everything was fine. For IT staff triaging the incident, an inaccurate status page is worse than no status page, because it actively misleads them into looking for problems on their own end. The discrepancy was noted in contemporaneous reporting on the outage.

Only one confirmed instance of this pattern is documented here: the August 2015 episode, sourced to a single piece of contemporaneous reporting. No other premature all-clear is asserted. A related but distinct pattern turns up again in this archive's 2026-outage-cluster-status-page-incidents entry, which documents eleven separate 2026 status-page incidents typically closed with a single generic line — "This incident has been resolved" — and no further public account of what had actually failed or when it was truly fixed for affected users. That is not the same claim as a false all-clear, but it is the same underlying problem: a status page's own language can be a starting point rather than a complete account of what users experienced.