From a80df921a2f4490a5925b2b0208c5190f44fe2a7 Mon Sep 17 00:00:00 2001 From: Rolfe Dlugy-Hegwer Date: Thu, 21 Oct 2021 09:12:03 -0400 Subject: [PATCH] RHDEVDOCS-3418 Create known issue for 2011293 --- release_notes/ocp-4-6-release-notes.adoc | 19 ++++++++++++++++++- 1 file changed, 18 insertions(+), 1 deletion(-) diff --git a/release_notes/ocp-4-6-release-notes.adoc b/release_notes/ocp-4-6-release-notes.adoc index bd8ce0432559..d8f9071a27d9 100644 --- a/release_notes/ocp-4-6-release-notes.adoc +++ b/release_notes/ocp-4-6-release-notes.adoc @@ -71,7 +71,7 @@ If you are creating machine configs for day 1 or day 2 operations that use Ignit [id="ocp-4-6-additional-steps-to-add-nodes-to-clusters"] ==== Additional steps to add nodes to existing clusters -For clusters that have been upgraded to {product-title} 4.6, you can add more nodes to your {product-title} cluster. These instructions are only applicable if you originally installed a cluster prior to {product-title} 4.6 and have since upgraded to 4.6. +For clusters that have been upgraded to {product-title} 4.6, you can add more nodes to your {product-title} cluster. These instructions are only applicable if you originally installed a cluster prior to {product-title} 4.6 and have since upgraded to 4.6. If you installed a user-provisioned cluster on bare metal or vSphere, you must ensure that your boot media or OVA image matches the version that your cluster was upgraded to. Additionally, your Ignition configuration file must be modified to be spec v3 compatible. For more detailed instructions and an example Ignition config file, see the link:https://access.redhat.com/solutions/5514051[Adding new nodes to UPI cluster fails after upgrading to OpenShift 4.6+] Knowledgebase Solution article. @@ -2421,6 +2421,23 @@ This caused image pulls to fail against the inaccessible private registry, norma * Currently, a Kubernetes port collision issue can cause a breakdown in pod-to-pod communication, even after pods are redeployed. For detailed information and a workaround, see the Red Hat Knowledge Base solution link:https://access.redhat.com/solutions/5940711[Port collisions between pod and cluster IPs on OpenShift 4 with OVN-Kubernetes]. (link:https://bugzilla.redhat.com/show_bug.cgi?id=1939676[*BZ#1939676*], link:https://bugzilla.redhat.com/show_bug.cgi?id=1939045[*BZ#1939045*]) +[id="ocp-4-6-devdocs-3418"] +* There is currently a known issue. By default, build pods do not pull images from the following Red Hat registries: `registry.redhat.io`, `registry.access.redhat.com`, and `quay.io` registries. This happens because the link:https://github.com/openshift/builder/commit/32f5b57382cedf4329039752f0e15a14b7f98366[pull request] for link:https://bugzilla.redhat.com/show_bug.cgi?id=1826183[BZ#1826183] introduced a default `registries.conf` file for buildah that does not include those Red Hat registries in the default search list. As a result, the Red Hat registries are not included in searches for image references. ++ +Workaround: If you have `` privileges, update `registries.conf` with the following lines: ++ +---- +tbd +---- ++ +Otherwise, if you cannot update `registries.conf`, include the `registry.redhat.io`, `registry.access.redhat.com`, or `quay.io` registry names when you specify an image. For example, in the `` file specify an image like this: ++ +---- + +---- ++ +(link:https://bugzilla.redhat.com/show_bug.cgi?id=2011293[*BZ#2011293*]) + [id="ocp-4-6-asynchronous-errata-updates"] == Asynchronous errata updates