Note: This bug is displayed in read-only format because the product is no longer active in Red Hat Bugzilla.

Bug 1761677

Summary: Cannot pull images from dockerhub in proxy testing environment
Product: OpenShift Container Platform Reporter: xiyuan
Component: NetworkingAssignee: Ben Bennett <bbennett>
Networking sub component: openshift-sdn QA Contact: zhaozhanqi <zzhao>
Status: CLOSED WORKSFORME Docs Contact:
Severity: low    
Priority: low CC: aos-bugs, bbennett, bparees, dmace, dmoessne, mfojtik, slaznick, xtian, yinzhou, zzhao
Version: 4.2.0   
Target Milestone: ---   
Target Release: 4.4.0   
Hardware: Unspecified   
OS: Unspecified   
Whiteboard:
Fixed In Version: Doc Type: If docs needed, set a value
Doc Text:
Story Points: ---
Clone Of: Environment:
Last Closed: 2020-04-28 15:46:50 UTC Type: Bug
Regression: --- Mount Type: ---
Documentation: --- CRM:
Verified Versions: Category: ---
oVirt Team: --- RHEL 7.3 requirements from Atomic Host:
Cloudforms Team: --- Target Upstream Version:
Embargoed:

Description xiyuan 2019-10-15 05:05:14 UTC
Description of problem:
https://prow.svc.ci.openshift.org/view/gcs/origin-ci-test/logs/release-openshift-ocp-installer-e2e-aws-proxy-4.2/77

Actual results:
fail [github.com/openshift/origin/test/extended/oauth/oauth_ldap.go:53]: Unexpected error:
    <*errors.errorString | 0xc003d9d8a0>: {
        s: "replica for openldap-server not avaiable: timed out waiting for the condition",
    }
    replica for openldap-server not avaiable: timed out waiting for the condition
occurred

Expected results:


Additional info:

Comment 1 Standa Laznicka 2019-10-15 08:53:57 UTC
"Oct 14 13:43:09.242: INFO: At 2019-10-14 13:39:58 +0000 UTC - event for openldap-server-df855dfcb-qpk8p: {kubelet ip-10-0-132-1.ec2.internal} Failed: Failed to pull image "docker.io/mrogers950/origin-openldap-test:fedora29": rpc error: code = Unknown desc = Get https://auth.docker.io/token?scope=repository%3Amrogers950%2Forigin-openldap-test%3Apull&service=registry.docker.io: proxyconnect tcp: dial tcp 35.196.128.173:3128: connect: no route to host"

The above means that the test couldn't pull an image from DockerHub in proxy environment. I have no idea who to assign it to, so moving to routing who lead this feature.

Comment 2 Daneyon Hansen 2019-10-15 16:27:41 UTC
I am able to pull and run the image using payload 4.3.0-0.okd-2019-10-14-190618:

$ oc run -it openldap --image=mrogers950/origin-openldap-test:fedora29
kubectl run --generator=deploymentconfig/v1 is DEPRECATED and will be removed in a future version. Use kubectl run --generator=run-pod/v1 or kubectl create instead.
If you don't see a command prompt, try pressing enter.
5da5ef11 conn=1000 fd=14 ACCEPT from IP=[::1]:48344 (IP=[::]:389)
5da5ef11 conn=1000 op=0 SRCH base="" scope=0 deref=0 filter="(objectClass=*)"
5da5ef11 conn=1000 op=0 SRCH attr=supportedSASLMechanisms
5da5ef11 conn=1000 op=0 SEARCH RESULT tag=101 err=0 nentries=1 text=
5da5ef11 conn=1000 op=1 UNBIND
<SNIP>

The "proxyconnect tcp: dial tcp 35.196.128.173:3128: connect: no route to host" error indicates that kubelet does not have a route to proxy 35.196.128.173:3128. Check the routing table on node ip-10-0-132-1.ec2.internal and make sure the default route is associated to the node's correct network interface and that the node can connect to the proxy.

Comment 3 zhou ying 2019-10-16 02:53:23 UTC
Also reproduced in: https://prow.svc.ci.openshift.org/view/gcs/origin-ci-test/logs/release-openshift-ocp-installer-e2e-aws-proxy-4.2/83

Caching blobs under "/var/cache/blobs".

Pulling image busybox:latest ...
Warning: Pull failed, retrying in 5s ...
Warning: Pull failed, retrying in 5s ...
Warning: Pull failed, retrying in 5s ...
error: build error: failed to pull image: After retrying 2 times, Pull image still failed due to error: while pulling "docker://busybox:latest" as "busybox:latest": Error initializing source docker://busybox:latest: pinging docker registry returned: Get https://registry-1.docker.io/v2/: proxyconnect tcp: dial tcp 35.196.128.173:3128: connect: no route to host
<----end of log for "multi-stage-build"/"docker-build"

Comment 4 Standa Laznicka 2019-10-16 07:26:45 UTC
*** Bug 1762164 has been marked as a duplicate of this bug. ***

Comment 5 Daneyon Hansen 2019-10-21 21:48:40 UTC
Eric Wolinetz increased the resource capacity of the proxy that gets created by the e2e-aws-proxy job. Please retest and update the bug based on your findings.

Comment 6 Dan Mace 2019-10-31 20:15:42 UTC
The dns-operator manages /etc/resolv.conf on the nodes, which is the only way I can see that dns-operator could be relevant in this context. I see the IP is resolved, but the client can't open a socket to the registry IP. And, this connection doesn't involve ingress. I am thus reassigning to SDN.

Comment 7 Ben Bennett 2019-11-08 16:20:54 UTC
Please retest based on https://bugzilla.redhat.com/show_bug.cgi?id=1761677#c5

Comment 8 Casey Callendrello 2019-11-15 15:26:39 UTC
The proxy CI job is completely broken; I can't see if this is still an issue.

Comment 9 Ben Parees 2020-04-28 15:46:50 UTC
not seeing lad test failures in current runs:
https://testgrid.k8s.io/redhat-openshift-ocp-release-4.2-informing#release-openshift-ocp-installer-e2e-aws-proxy-4.2&sort-by-failures=

I do see the error message:
https://search.svc.ci.openshift.org/?search=replica+for+openldap-server+not+avaiable&maxAge=336h&context=1&type=bug%2Bjunit&name=&maxMatches=5&maxBytes=20971520&groupBy=job

but it only occurs in runs where many many things were going wrong.

Closing this as working now.

Comment 10 Red Hat Bugzilla 2023-09-14 05:44:32 UTC
The needinfo request[s] on this closed bug have been removed as they have been unresolved for 1000 days