Bug 1599789

Summary: py-amqp will try to resolve hostnames through nameservers unless they have both IPv4 and IPv6 entries in /etc/hosts leading to timeouts
Product: Red Hat OpenStack Reporter: Daniel Alvarez Sanchez <dalvarez>
Component: python-amqpAssignee: John Eckersberg <jeckersb>
Status: CLOSED WONTFIX QA Contact: Shai Revivo <srevivo>
Severity: high Docs Contact:
Priority: high    
Version: 10.0 (Newton)CC: apevec, jschluet, knoha, lhh, lmiccini, rhos-maint, srevivo
Target Milestone: zstreamKeywords: Triaged, ZStream
Target Release: ---   
Hardware: Unspecified   
OS: Unspecified   
Whiteboard:
Fixed In Version: Doc Type: If docs needed, set a value
Doc Text:
Story Points: ---
Clone Of: 1599787 Environment:
Last Closed: 2020-03-03 15:28:15 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:
Bug Depends On: 1599766, 1599787, 1607962    
Bug Blocks: 1599784    

Description Daniel Alvarez Sanchez 2018-07-10 15:19:03 UTC 
+++ This bug was initially created as a clone of Bug #1599766 +++

Description of problem:
OSPD normally configures the controllers hostnames in /etc/hosts (IPv4) but py-amqp will try to query nameservers because there's no IPv6 entry there.

When nameservers are non-existent or unreachable, this leads to 60 seconds timeout when connecting to Rabbit.

The fix got recently merged in py-amqp:

https://github.com/celery/py-amqp/commit/1ad97fb14c0c3c57395ca525932f95a830e51a88

We would need this fix in OSP10.
Comment 1 Luca Miccini 2020-03-03 15:28:15 UTC 
At this time we do not plan to backport these patches to osp10. Please reopen if this needs to be re-evaluated.
Thanks.