Meta favicon

Meta outage on 2026-10-20

metastatus.com
October 2026 20 Major incident

Meta's Data Transparency component, specifically the Meta Secure Research Environment, went down for 24 minutes on 20 October 2026. The fault sat on Meta's side, affecting the Jupyter Hub researchers use inside that environment.

Started

13:28 UTC

Duration

Lasted 24m

Source

IsDown

Next time

Meta 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: Data Transparency

What happened?

Meta's status page logged one incident on this component, affecting the Meta Secure Research Environment. The first update, timestamped 02:14, says Meta's team was aware of an outage that may be impacting Jupyter Hub and was actively looking to resolve it. The resolution update came at 13:28, stating services had been recovered and restored, with an apology for any inconvenience. Meta's own duration figure for the incident is 24 minutes. No cause is given in either update, so the reason Jupyter Hub stopped working is not on the record. This is a narrow component. Data Transparency covers the Secure Research Environment, a tool used by outside researchers to access Meta data under controlled conditions, not the consumer apps most people associate with Meta. No user reports were filed during the window, which is unsurprising given how small the affected population is. Meta's history for this component shows 11 incidents in the last 90 days with a median duration of 2 hours 4 minutes, so a 24 minute outage sits well below their recent typical length. Across all time Meta has logged 573 incidents on record, averaging 14.3 per month, which gives a sense of how often something somewhere in their stack goes down. This one filing from Meta is the only account of what happened, there is no independent confirmation beyond it.

Learning

UptimeRobot's own 30 day stats for this component show 3 incidents, all resolved, with a median duration of 24 minutes, which lines up closely with what Meta itself reported for this specific event. That agreement matters because the Secure Research Environment is not a public-facing endpoint you can simply ping from outside, it is an internal research tool gated behind Meta's own access controls, so external checks against it depend entirely on what Meta chooses to expose and report. When a tool like Jupyter Hub fails inside that environment, the failure happens behind Meta's authentication and infrastructure layer, somewhere your own monitoring of your own product would never see it, because the breakage isn't in your code or your endpoints, it's upstream in Meta's environment. If you build anything that depends on this research access, your dashboards staying green tells you your half is fine, it says nothing about whether the environment itself is reachable. Meta's status page is the only record of this outage, no post-mortem has been published, and no cause is stated beyond "an outage." 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.