Resolved -
We have completed the database maintenance. If you notice problems, please contact @npm_support.
Aug 28, 04:57 UTC
Investigating -
We are conducting database maintenance over the next hour. During this time, all npm services (the website, registry) will be read-only. We'll keep you updated as this maintenance is being performed.
Aug 28, 04:08 UTC
Resolved -
We have resolved this issue (404s observed for scoped package registry reads).
Aug 23, 03:54 UTC
Monitoring -
We have applied the fix for the issue (404s noted for scoped registry reads). We are monitoring the result of the applied fix.
Aug 23, 03:32 UTC
Update -
We are continuing to work on a fix for this issue.
Aug 23, 03:31 UTC
Identified -
We have identified the cause of the scoped package registry 404 issue. We are now implementing a fix.
Aug 23, 03:18 UTC
Investigating -
Currently we are observing 404s on scoped package registry reads. We are currently investigating this issue, and will update when more information is available.
Aug 23, 02:15 UTC
Resolved -
Newly-published package visibility should be back to normal.
Aug 21, 18:18 UTC
Monitoring -
The upstream service is now green and package-version availability should be returning to normal. Continuing to monitor.
Aug 21, 18:00 UTC
Resolved -
The issue with failures on requests to unpublish packages within the initially permitted 72-hour window is now resolved.
Aug 17, 00:25 UTC
Monitoring -
We've applied a fix for unpublish errors during the unpublish window.
Aug 16, 21:04 UTC
Update -
We have narrowed this down even further to a failing elasticsearch query and are working on a fix.
Aug 16, 14:43 UTC
Identified -
We are aware that a large proportion of requests to unpublish packages within the initially permitted 72 hour window are erroring with a generic HTTP 500 error.
This is due to an internal service request timing out when authenticating the unpublish requests, and is not affecting any parts of the package publishing workflow, and we are working on a fix.
Aug 16, 00:47 UTC
Resolved -
This incident has been resolved.
Aug 16, 20:04 UTC
Monitoring -
A fix has been implemented and users should be able to unpublish packages successfully (subject to the package unpublish policies: https://www.npmjs.com/policies/unpublish )
Aug 16, 18:47 UTC
Identified -
This has been identified as a failing elasticsearch query and we are working on a fix.
Aug 16, 14:44 UTC
Investigating -
We are currently investigating delays in updates to the "Recently updated packages" section of the homepage.
We suspect the delays started around 7:20 UTC and are caused by an earlier production outage.
Aug 14, 14:31 UTC
Resolved -
Replication has gone back to normal on replicate.npmjs.com. This incident is resolved.
Aug 15, 18:37 UTC
Monitoring -
We have implemented a fix and are seeing lower replication delays. We're continuing to monitor the incident.
Aug 15, 16:23 UTC
Investigating -
Our public packages replication host, replicate.npmjs.com is currently experiencing increased error rates and problems with replicating from upstream. We are investigating these issue.
Aug 15, 16:03 UTC