Gradle outage history

While trying to deploy an upgrade, the application was down for 20 minutes from 11:19 GMT+2. We will investigate the root cause and attempt the upgrade at a later date.

We are following up news from Cloudflare

Listing more impacted services, although recovery might be in progress

Cloudflare implemented a fix and services seem to be back https://www.cloudflarestatus.com/incidents/lfrm31y6sw9q

Services are back to normal

The plugin portal suffered connection issues possibly related to incidents with the different providers used: * Heroku: https://status.heroku.com/incidents/2867 * AWS: https://health.aws.amazon.com/health/status?eventID=arn:aws:health:us-east-1::event/INTERNETCONNECTIVITY/AWS_INTERNETCONNECTIVITY_OPERATIONAL_ISSUE/AWS_INTERNETCONNECTIVITY_OPERATIONAL_ISSUE_49824_16DD211EFC1 When restarted, the application started behaving properly.

We are currently investigating this issue.

The issue has been identified and a fix is being implemented.

A fix has been implemented and we are monitoring the results.

services.gradle.org is fully operational again. The outage was caused by a configuration change on services.gradle.org that was not intended to affect this service.

We are currently investigating this issue.

This incident has been resolved.

We are currently investigating this issue.

This incident has been resolved. Between 19:14 and 19:36 UTC the Plugin Portal UI experienced degraded performance and had a small portion of requests fail with errors.

On May 23, from 7:42 AM UTC to 8 AM, the Plugin Portal was unavailable following a failed maintenance operation. Service has been restored successfully.

We are currently investigating the issue. Traffic against the plugin portal has increased heavily this week, with a serious uptake in traffic from user agents identifying as Renovate and some specific source IPs having increased traffic from nearly nothing to multiple millions requests an hour. We have enforced rate limiting for the worst offender.

A fix has been implemented and we are monitoring the results.

This incident has been resolved. We will continue to monitor the system in the coming days.

We are currently investigating this issue.

A fix has been implemented and we are monitoring the results.

This incident has been resolved. We will continue close monitoring of the system in the coming days.

Without any prior announcement, it looks like JCenter is now redirecting to Maven Central instead of serving artifacts. Builds that have build logic that relies on artifacts available only on JCenter and not Maven Central will fail. We are investigating what we can do to mitigate the issue. We also contacted JFrog to understand this change.

JCenter is now serving artifacts again. We're monitoring if this has resolved the situation and we'll provide more information later.

This incident has been resolved.

We are currently investigating this issue.

We continue looking into the issue. We disabled uploading of new plugins to the plugin portal for the time being.

We are continuing to investigate slow response times and errors on https://plugins.gradle.org. New account creation and plugin uploads are still disabled.

A fix has been implemented and we're monitoring the impact on the Plugin Portal. We are still seeing above average response times, but overall timeouts are down. New account creation and plugin uploads have been re-enabled.

We are seeing increased load and timeouts again. We're investigating the cause.

We've deployed another fix and are monitoring the impact.

This incident has been resolved.

The plugin portal is encountering issues with JCenter right now. We are seeing "Remote host terminated the handshake". Investigating and reaching out to JFrog.

JCenter is now accepting connections, we are monitoring the situation.

We are continuing to monitor for any further issues.

This incident has been resolved.

JCenter has expired SSL certificates, the Plugin Portal now uses Maven Central has its backing repository. This will cause issue for libraries that are available on JCenter but not Maven Central. This can include older versions of plugins.

JCenter's certificate has been renewed.

We are currently investigating this issue.

The issue appears to be upstream. We're monitoring for failures and will take action if failures increase.

We received more reports that some dependency downloads are failing from jcenter.bintray.com. For now, we've changed the backend to redirect to Maven Central. You may notice some redirects to JCenter for the next few hours until caches expire. Unfortunately, this means artifacts that were only available on JCenter/Bintray and were never migrated to Maven Central will not be available from the Plugin Portal. If you are using dependency verification, you may also notice different artifact checksums for libraries that were re-uploaded to Maven Central. We'll continue to monitor the situation.

We've restored the redirect to JCenter. You may notice some redirects to Maven Central for the next few hours until caches expire. We will continue to monitor for increased failures.

This incident has been resolved.

While the mirroring put in place alleviates some of the JCenter being down impact, the mirror is a bit out of date and currently cannot sync recent artifacts. This continues to impact builds.

The new settings have resolved most of the issues. Note that you may still experience problems if the package only exists on JCenter and was never mirrored until now.

JCenter has had issues a good part of Sunday October 30th and continues to have issues this morning. A temporary solution has been put in place to reduce the impact on users of the Gradle Plugin Portal.

Impacts from the JCenter issue should be resolved. We will keep monitoring the situation.

The plugins.gradle.org service depends on the Heroku hosting platform, which is currently suffering a widespread outage affecting DNS resolution. We are currently investigating options and tracking the root cause Heroku outage.

Temporary mitigation has been put in place to restore service. As the upstream Heroku outage evolves, further downtime may be experienced.

A fix has been implemented and we are monitoring the results.

This incident has been resolved.

We are currently investigating this issue.

We have identified that our communication with S3 in the backend was showing an increased error rate. This had an effect on response time for the plugin portal. Our logging shows that the error rate is back to normal. We will continue to monitor the service closely for a bit longer.

The root cause was the Cloudflare outage - see https://www.cloudflarestatus.com/incidents/xvs51y9qs9dj - which is now resolved.

We are currently investigating this issue.

A change is being deployed with expectations that it will improve the situation.

This incident has been resolved.

Currently no Build Scan activation emails are sent when activating a Build Scan. We are currently investigating together with the mail provider what the issue is that emails are not sent. Build Scan ingestion works and all build scans can be viewed once the activation email service is functional again.

This incident has been resolved.