Docker proxy repo continually times out waiting for pool connection

We’re seeing an odd issue with one of our proxied docker repositories on Nexus 3.38.0. We have a number of docker proxies in a docker group, and all but the quay.io proxy seem to work fine. All requests for images from the quay proxy time out with the following error in the logs:

WARN [qtp2111565389-23149] *UNKNOWN org.sonatype.nexus.repository.docker.internal.orient.DockerProxyFacetImpl - Exception org.apache.http.conn.ConnectionPoolTimeoutException: Timeout waiting for connection from pool checking remote for update, proxy repo quay-io-proxy failed to fetch v2/prometheus-operator/prometheus-operator/manifests/v0.47.0, content not in cache.

At a glance, it looks like an internal connection pool is maxed out but I don’t see any other obviously related errors in the logs. The issue persists though service restarts, and even after deleting and recreating the quay proxy. When pulling a packet capture, there’s no traffic between nexus and quay.io which leads me to think this issue is completely internal to nexus. I’m not sure where to go from here to continue troubleshooting this issue.

I believe we’ve fixed this and it will be in the next release - [NEXUS-31322] docker tags/list outbound paged request can consume all available outbound connections for the proxy repository - Sonatype JIRA

1 Like