goharbor/harbor

The nightly pipeline db_offline always failed at the testcase system_cve testcase. failed screenshot failed log.html log.html It seems that the date time picker failed to find the previous/next day of current day.

Is your feature request related to a problem? Our organization is deploying Harbor using the official Helm chart with an external PostgreSQL database. While the Harbor documentation mentions support for external PostgreSQL (12 or later), it does not explicitly validate or document support for PostgreSQL 16, 17, and 18. Our organization requires using the latest PostgreSQL major versions to meet security, compliance, and infrastructure standardization requirements. Describe the solution you'd like Please add official support and validation for PostgreSQL 16, 17, and 18 when deploying Harbor using the official Helm chart with an external PostgreSQL database. Why is this needed? Organizations are adopting newer PostgreSQL major versions as part of their standard platform. Using the latest supported database versions provides access to security updates, performance improvements, and long-term support. Official compatibility documentation would help users confidently deploy Harbor with new

I have a namespace named harbor in my k8s cluster. trivy returns an error and does not find vulnerabilities in the images: {"error":"running trivy wrapper: running trivy: exit status 1: 2026-07-08T13:10:44.337Z\t\u001b[34mINFO\u001b[0m\tNeed to update DB\n2026-07-08T13:10:44.337Z\t\u001b[34mINFO\u001b[0m\tDownloading DB...\n2026-07-08T13:11:15.867Z\t\u001b[31mFATAL\u001b[0m\tfailed to download vulnerability DB: failed to download vulnerability DB: DB file not found\n","level":"error","msg":"Scan job failed","scan_job_id":"7ac89af31d68e0b3cb4265d7","time":"2026-07-08T13:11:16Z"}. If I go to the node where the scanner is located and checked the availability of remote resources: [root@node7 ~]# curl -k https://ghcr.io/aquasecurity/trivy-db:2 [root@node7 ~]# curl -k https://ghcr.io/v2/ [root@node7 ~]# curl -k https://github.com [root@node7 ~]# curl -k https://api.github.com [root@node7 ~]# curl -k github-releases.githubusercontent.com and # curl -k https://140.82.121.33/ <!DOCTYPE htm

Is your feature request related to a problem? Please describe. When replicating images from Harbor to AWS ECR, Harbor checks if a repository exists and automatically creates it if it does not. This breaks AWS ECR repository creation templates, because ECR templates are only applied when the repository is created automatically by ECR itself (create-on-push). As a result, replicated repositories are created with default settings instead of the expected configuration (immutability, lifecycle policies, encryption, tags, etc.). Describe the solution you'd like Add an optional replication setting like “Create repository if not exists” (enabled by default for backward compatibility). When disabled, Harbor should not call CreateRepository on the target registry. Harbor should attempt to push images directly and let the target registry auto-create the repository. If creation templates are not enabled the replication will fail with "repository not found". Describe the main design/architectur

I just installed harbor and opened the ui on my mobile Phone. The interface can not be used inside a Chrome Browser on an Android Phone. Are there any Plans for a responsive interface?