Description of the problem: When installing an 4.11 IPv6 DHCP cluster, after masters get to rebooting stage they never join back to the cluster due to errors during execution of the resolv-prepender script. The nodes journal displays "NM resolv-prepender: NM resolv.conf still empty of nameserver" multiple times before attempting to pull container images and ultimately failing. Release version: 4.11.0-0.nightly-2022-07-26-041421 Operator snapshot version: 2.1.0-DOWNANDBACK-2022-07-25-16-01-39 OCP version: 4.11 Steps to reproduce: 1. Install a 4.11 multi-node cluster with IPv6 DHCP networking Actual results: Installation fails due to masters not rebooting within timeout Expected results: Installation completes successfully Additional info: Jul 26 07:34:10 mdhcp-master-0-0 nm-dispatcher[1972]: + [[ OVNKubernetes == \O\V\N\K\u\b\e\r\n\e\t\e\s ]] Jul 26 07:34:10 mdhcp-master-0-0 nm-dispatcher[1972]: + [[ '' == \W\i\r\e\d\ \C\o\n\n\e\c\t\i\o\n ]] Jul 26 07:34:10 mdhcp-master-0-0 nm-dispatcher[1972]: + '[' -z ']' Jul 26 07:34:10 mdhcp-master-0-0 nm-dispatcher[1972]: + echo 'Not a DHCP4 address. Ignoring.' Jul 26 07:34:10 mdhcp-master-0-0 nm-dispatcher[1972]: Not a DHCP4 address. Ignoring. Jul 26 07:34:10 mdhcp-master-0-0 nm-dispatcher[1972]: + exit 0 Jul 26 07:34:10 mdhcp-master-0-0 nm-dispatcher[1973]: + '[' -z ']' Jul 26 07:34:10 mdhcp-master-0-0 nm-dispatcher[1973]: + echo 'Not a DHCP6 address. Ignoring.' Jul 26 07:34:10 mdhcp-master-0-0 nm-dispatcher[1973]: Not a DHCP6 address. Ignoring. Jul 26 07:34:10 mdhcp-master-0-0 nm-dispatcher[1973]: + exit 0 Jul 26 07:34:10 mdhcp-master-0-0 nm-dispatcher[1976]: Error: Device '' not found. Jul 26 07:34:10 mdhcp-master-0-0 nm-dispatcher[1986]: + [[ OVNKubernetes == \O\V\N\K\u\b\e\r\n\e\t\e\s ]] Jul 26 07:34:10 mdhcp-master-0-0 nm-dispatcher[1986]: + [[ '' == \W\i\r\e\d\ \C\o\n\n\e\c\t\i\o\n ]] Jul 26 07:34:10 mdhcp-master-0-0 nm-dispatcher[1986]: + '[' -z ']' Jul 26 07:34:10 mdhcp-master-0-0 nm-dispatcher[1986]: + echo 'Not a DHCP4 address. Ignoring.' Jul 26 07:34:10 mdhcp-master-0-0 nm-dispatcher[1986]: Not a DHCP4 address. Ignoring. Jul 26 07:34:10 mdhcp-master-0-0 nm-dispatcher[1986]: + exit 0 Jul 26 07:34:10 mdhcp-master-0-0 nm-dispatcher[1987]: + '[' -z ']' Jul 26 07:34:10 mdhcp-master-0-0 nm-dispatcher[1987]: + echo 'Not a DHCP6 address. Ignoring.' Jul 26 07:34:10 mdhcp-master-0-0 nm-dispatcher[1987]: Not a DHCP6 address. Ignoring. Jul 26 07:34:10 mdhcp-master-0-0 nm-dispatcher[1987]: + exit 0 Jul 26 07:34:10 mdhcp-master-0-0 nm-dispatcher[1990]: Error: Device '' not found.
*** This bug has been marked as a duplicate of bug 2105069 ***
I'm going to reopen this bug as a blocker and use it to track the following blocking bugs: - BZ2100456 - BZ2105069 - BZ2111632 Once these bugs are fixed we can validate with the infrastructure operator and close this out.
Per latest update, fix was merged, waiting for the next OCP release (weekly)。 May we know when exactly the OCP release with the PR merged is available? Does it require a new AI build at ACM side? Thanks!
Hi Yuanyuan, This is the PR we are depending on for a fix in 4.11: https://github.com/openshift/machine-config-operator/pull/3287. So far it has only been merged for 4.12
This was fixed and verified by QE