Description of problem: When the cluster is installed in a new region that is unknown in the AWS SDK version being used, the image-registry pod crashes. How reproducible: 100% Steps to Reproduce: 1. Install a cluster in ap-northeast-3. Use custom endpoints for ec2 and elb so that the machine-api-operator can create the worker machines. 2. Note that the image-registry pods crashloop. Actual results: time="2021-04-22T16:34:48.852602706Z" level=info msg="start registry" distribution_version=v2.7.1+unknown go.version=go1.16.1 openshift_version=4.8.0-202104162013.p0-1b129a6 time="2021-04-22T16:34:48.853154564Z" level=info msg="caching project quota objects with TTL 1m0s" go.version=go1.16.1 panic: Invalid region provided: ap-northeast-3 goroutine 1 [running]: github.com/docker/distribution/registry/handlers.NewApp(0x1ef27e8, 0xc000128010, 0xc0002c7880, 0xc0003934d0) /go/src/github.com/openshift/image-registry/vendor/github.com/docker/distribution/registry/handlers/app.go:127 +0x3133 github.com/openshift/image-registry/pkg/dockerregistry/server/supermiddleware.NewApp(0x1ef27e8, 0xc000128010, 0xc0002c7880, 0x1efab98, 0xc000403200, 0x1efe100) /go/src/github.com/openshift/image-registry/pkg/dockerregistry/server/supermiddleware/app.go:96 +0x87 github.com/openshift/image-registry/pkg/dockerregistry/server.NewApp(0x1ef27e8, 0xc000128010, 0x1ec6518, 0xc000550bc8, 0xc0002c7880, 0xc0000b20a0, 0x0, 0x0, 0xc0003920f0, 0xc00053f7c0) /go/src/github.com/openshift/image-registry/pkg/dockerregistry/server/app.go:138 +0x2c5 github.com/openshift/image-registry/pkg/cmd/dockerregistry.NewServer(0x1ef27e8, 0xc000128010, 0xc0002c7880, 0xc0000b20a0, 0x0, 0x0, 0x1f142b8) /go/src/github.com/openshift/image-registry/pkg/cmd/dockerregistry/dockerregistry.go:210 +0x1ce github.com/openshift/image-registry/pkg/cmd/dockerregistry.Execute(0x1eb9c40, 0xc0005509b8) /go/src/github.com/openshift/image-registry/pkg/cmd/dockerregistry/dockerregistry.go:164 +0xa65 main.main() /go/src/github.com/openshift/image-registry/cmd/dockerregistry/main.go:93 +0x573 Expected results: The image-registry will accept the new region as it matches the regular expression for a region in the default AWS partition.
Still met the error when testing with above two prs.
Are you sure? These PRs bump aws-sdk-go to v1.38.35, the region ap-northeast-3 was introduced in v1.37.24. Which region do you use? What message do you see?
(In reply to Oleg Bulatov from comment #3) > Are you sure? These PRs bump aws-sdk-go to v1.38.35, the region > ap-northeast-3 was introduced in v1.37.24. > > Which region do you use? What message do you see? Sorry, after double confirm, I did a misoperation， will try again.
Installation succeeds with the prs: openshift/image-registry#272,openshift/cluster-image-registry-operator#685
Verified on 4.8.0-0.nightly-2021-06-01-002052. storage: managementState: Managed s3: bucket: wzheng-48-gr7js-image-registry-ap-northeast-3-erpntneleqsylxkm encrypt: true region: ap-northeast-3 virtualHostedStyle: false unsupportedConfigOverrides: null
Since the problem described in this bug report should be resolved in a recent advisory, it has been closed with a resolution of ERRATA. For information on the advisory (Moderate: OpenShift Container Platform 4.8.2 bug fix and security update), and where to find the updated files, follow the link below. If the solution does not work for you, open a new bug report. https://access.redhat.com/errata/RHSA-2021:2438