CloudStatus

Npmjs Status Updates

Increased 503 rates for EU users requesting scoped packages

Jun 29, 16:41 UTC Resolved - At around 14:10 UTC our monitoring alerted us of increased registry 503 rates. We've isolated the issue to our EU region and identified the cause to be a misconfiguration of our deployment stack on one of the servers in that region. The issue was resolved around 14:55 UTC . We've since taken additional steps in order to ensure consistency of our deployment configuration in other regions and will be taking steps to maintain that consistency.

About 381 days ago - Thu 29 Jun 2017 16:41 (GMT)

Source: Npmjs - npm

Npmjs History

elevated level of 503s in Europe

May 2, 15:19 UTC Resolved - This incident has been resolved.May 2, 15:08 UTC Monitoring - A fix has been implemented and we are monitoring the results.May 2, 15:00 UTC Investigating - We are currently seeing an elevated level of 503s for our Points of Presence in Europe, affecting both the registry and the website.

About 227 days ago - Fri 1 Dec 2017 10:58 (GMT)

More info

503s for scoped packages

May 4, 13:52 UTC Resolved - This incident has been resolved.May 4, 12:38 UTC Monitoring - A fix has been implemented and we are monitoring the results.May 4, 12:27 UTC Investigating - We are currently experiencing 503s for scoped packages, the team is investigating.

About 227 days ago - Fri 1 Dec 2017 10:58 (GMT)

More info

Tarball and package pages outage

May 8, 21:18 UTC Resolved - We have identified the outage to be caused by one of our package servers' TLS terminator shutting down due to a misconfiguration. In addition to this, that service did not restart properly, causing need for human intervention. We are taking steps to fix the configuration that caused the incident and add more detailed monitoring.May 8, 20:50 UTC Investigating - We are currently investigating an outage causing unavailability of registry's tarballs and website's package pages.

About 227 days ago - Fri 1 Dec 2017 10:58 (GMT)

More info

elevated level of 503s on /search endpoint

May 9, 21:10 UTC Resolved - This incident has been resolved.May 9, 19:38 UTC Monitoring - A fix has been implemented and we are monitoring the results.May 9, 19:24 UTC Investigating - We are currently seeing an elevated level of 503s on all requests to our /search endpoint. The team is currently investigating.

About 227 days ago - Fri 1 Dec 2017 10:58 (GMT)

More info

Increased 503 rates for public packages

Jun 9, 14:30 UTC Resolved - We've identified the issue as a case of a service's process getting stuck and resolved it by rolling the affected service.Jun 9, 14:25 UTC Investigating - We're currently investigating issues related to accessing scoped packages.

About 227 days ago - Fri 1 Dec 2017 10:58 (GMT)

More info

Experiencing problems with scoped package downloads

Jun 13, 21:46 UTC Resolved - A service had fallen over and wasn't coming back up. We brought it back up.Jun 13, 21:40 UTC Investigating - We're experiencing an increased 503 rate for scoped package JSON documents as well as tarballs.

About 227 days ago - Fri 1 Dec 2017 10:58 (GMT)

More info

Increased error rates due to a CDN provider outage

Jun 28, 14:41 UTC Resolved - Starting at around 13:50 UTC our monitoring started reporting increased error rates, which we shortly after identified as related to an outage of our CDN provider. The situation was resolved by that provider at around 14:03 UTC . We are continuing to pay close attention to our metrics.

About 227 days ago - Fri 1 Dec 2017 10:58 (GMT)

More info

package tarball issues in us-east.

Apr 21, 02:36 UTC Resolved - Appears to have been a network flicker. Apologies for the trouble!Apr 21, 02:35 UTC Investigating - We are investigating an issue with a tarball server in us-east.

About 451 days ago - Fri 21 Apr 2017 02:37 (GMT)

More info

package tarball issues in us-east.

Apr 21, 02:35 UTC Investigating - We are investigating an issue with a tarball server in us-east.

About 451 days ago - Fri 21 Apr 2017 02:36 (GMT)

More info

forced unpublishes currently failing

Mar 30, 02:00 UTC Resolved - we have verified that "unpublish -f" is back to working normally.Mar 30, 01:14 UTC Monitoring - We have pushed and tested a fix for forced package unpublishes and are watching logs to verify.Mar 29, 19:22 UTC Identified - A bug has been identified related to "npm unpublish -f", resulting in packages only partially being removed from the registry. This results in the packages continuing to be installable, but in users being unable to publish updates. We are currently working on a fix, and will keep this issue updated.

About 468 days ago - Mon 3 Apr 2017 22:42 (GMT)

More info