NPM had two separate outages on 22 October 2026, hitting package installation, publishing and the www.npmjs.com website. The longer of the two ran 5h 18m, and both were on NPM's side, not yours.
Started
02:33 UTC
Duration
Lasted 1h 51m
Source
IsDown
Next time
npm 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: www.npmjs.com website, Package installation, Package publishing
What happened?
NPM's status page logged two incidents on this day. The first started at 02:33 UTC, titled "Degraded Package Install and Publishing", and ran for 1h 51m. NPM's team posted an investigating update at 02:33, another at 02:37, identified the issue at 03:07, implemented a fix by 04:10, and marked it resolved at 04:24. The second incident started at 06:43 UTC, titled "Degraded Publish, Install, and View Package", and lasted 5h 18m, the longer of the two. That one got only three updates: investigating at 06:43, a fix implemented and monitoring at 09:00, and resolved at 11:59, so the three-hour gap between those last two updates isn't explained beyond "monitoring the results". Together, three components were affected across the day: www.npmjs.com, package installation, and package publishing. No user reports were filed in this window, which is unusual for an outage spanning registry installs and publishes, but it is what the record shows. NPM's own filings are the only account of these incidents available, and they don't go into root cause for either one. In the 90 days before this, NPM logged 9 incidents with a median duration of 2 hours 19 minutes, so today's second incident ran well past that median. Over all time, NPM has logged 189 incidents, averaging 2.5 a month.
Learning
UptimeRobot's own probes against NPM's registry, checked every 60 seconds from both EU and NA regions, registered this incident, with 24-hour uptime dropping to 99.664% while 7-day and 30-day figures stayed closer to full. That dip lines up with NPM's own filings, which is the useful case: your monitoring and the vendor's status page agree that something broke. Over the last 30 days, UptimeRobot recorded 4 incidents on NPM, all resolved, with a median duration of 3h 48m, longer than NPM's own 90-day median, which tells you NPM's status page updates can lag behind what external checks actually see. The shape of this failure sits at the registry layer: installs and publishes degrade or fail even though nothing in your own application code or deployment pipeline has changed. Watching your own endpoints wouldn't catch this, because your build and deploy steps depend on NPM responding correctly, not on anything you control. Your monitoring covers your half, UptimeRobot now watches the provider's half too, so next time you know which side broke without guessing.
