Wenjing, Can you verify the same proxy settings exist on the node running the source pod? This node should also be running a coredns pod from the cluster-dns daemonset.
Imagestream import happens on the openshift apiserver - re-assigning (I thought proxy was applied to the entire apiserver).
Can we get must-gather data for a cluster exhibiting the "read udp [...]->172.30.0.10:53: i/o timeout" errors? In particular, I would like to see the "dns" clusteroperator and the logs for the dns-operator deployment (in the "openshift-dns-operator" namespace) and the dns-default daemonset (in the "openshift-dns" namespace).
Still no available proxy cluster till now for bug #1737683(verification fails on QA)
Imagestream can be imported successfully on 4.2.0-0.nightly-2019-08-15-205330. So moving this bug to be verified.
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, 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/RHBA-2019:2922