Okta's login page went unavailable in its OK6 cell for 1h 31m on 9 November 2026. The Core Platform component was affected, and the fault sat on Okta's side, not yours.
Started
17:30 UTC
Duration
Lasted 1h 31m
Source
IsDown
Next time
Okta 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: Core Platform
What happened?
Okta logged one incident on 9 November 2026, titled "Login Page Unavailable in OK6", starting at 17:30 UTC and lasting 1h 31m. The affected component is listed as Core Platform, and no other components appear on the record. Okta filed no updates at all during the incident, so there is no note on what caused the login page to go down or what was done to bring it back. No user reports were filed during the window, so there is no outside account of what people saw when they tried to sign in. For a company the size of Okta, OK6 is one of several cells it runs, and a login page failure there would have blocked anyone routed to that cell from reaching their dashboard. Okta's status page is the only record here, and with zero updates attached, it carries less weight than a fuller filing would. Over the past 90 days Okta has logged 14 incidents with a median duration of 2 hours 35 minutes, so this one resolved faster than Okta's typical incident. Across all time, Okta averages 2.4 incidents a month, putting this one in line with a pattern rather than being unusual. There is no indication this incident is still open: Okta marked it closed, with nothing further filed since.
Learning
A login page outage sits in the identity layer, not in your own application code, and that is exactly the kind of failure your own endpoint checks will never catch because your product can be running perfectly while the door to it is locked. In the 30 days around this incident, UptimeRobot's own checks against Okta recorded 7 incidents, all resolved, with a median duration of 2h 48m, which is a separate measurement from Okta's own incident log and does not need to match it exactly to both be correct. Your uptime monitoring tells you whether your own servers answer, but it says nothing about whether Okta's login flow is reachable upstream of you. When sign-in breaks at the provider, your side can stay green the whole time and still leave people locked out. That gap is the reason to watch the provider directly rather than assume your own green dashboard covers it. Your monitoring covers your half, UptimeRobot now watches Okta's half too, so next time something breaks you know which side it was without guessing.