Bug 2196452

Summary: DNS name resolution failing in server
Product: Red Hat OpenStack Reporter: Saumik Paul <sapaul>
Component: documentationAssignee: Greg Rakauskas <gregraka>
Status: ASSIGNED --- QA Contact: RHOS Documentation Team <rhos-docs>
Severity: high Docs Contact:
Priority: unspecified    
Version: 16.2 (Train)CC: apevec, gregraka, ihrachys, jamsmith, lhh, majopela, mlavalle, njohnston, rhos-docs, scohen, tpushkar, ykarel
Target Milestone: ---Keywords: Triaged
Target Release: ---   
Hardware: Unspecified   
OS: Linux   
Whiteboard:
Fixed In Version: Doc Type: If docs needed, set a value
Doc Text:
Story Points: ---
Clone Of: Environment:
Last Closed: 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:

Comment 30 Ihar Hrachyshka 2023-05-22 16:44:14 UTC 
@Saumik,

I believe this is the code that disseminates DNS server address to DHCP clients: https://github.com/openstack/neutron/blob/d4cc45131979f1d297f39e2e5aad37a45917e358/neutron/common/ovn/utils.py#L578C52-L580

Maybe this can be worked around by setting subnet's dns_nameservers field to point to the (implied) gateway_ip for DNS native resolution. E.g. for subnet 10.1.2.0/24 you would point it to 10.1.2.1. I am going to test it in next weeks but if you can beat me to it, I would appreciate.

Also something to check: if DHCP agent with ml2/ovs driver behaves differently from ml2/ovn native implementation.

In general, I think this is something we could change to make DNS resolution work for isolated networks. I will check the current behavior with upstream bits and report back.