error: Failed to connect to 2001:4800:7813:516:3bc3:d7f6:ff04:aacb: Network is unreachable while accessing https://git.openstack.org/openstack/tempest/info/refs

[Clint Byrum]

I believe this was a transient failure:

http://logs.openstack.org/05/59905/5/gate/gate-tempest-dsvm-large-ops/be1ad16/

https://review.openstack.org/#/c/59905/

[Florent Flament (florentflament)]

Same issue there:
2014-01-23 14:20:21.164 | fatal: unable to connect to git.openstack.org:
2014-01-23 14:20:21.164 | git.openstack.org[0: 2001:4800:7813:516:3bc3:d7f6:ff04:aacb]: errno=Network is unreachable

http://logs.openstack.org/06/68606/1/check/check-tempest-dsvm-postgres-full/19f813a/console.html
http://logs.openstack.org/06/68606/1/check/gate-tempest-dsvm-large-ops/51ebbce/console.html

[Jeremy Stanley (fungi)]

I've added https://review.openstack.org/67911 to help us collect some more details the next time this happens. My best guess is some other tenant in that same flat network has a rogue IPv6 router advertisement daemon running.

[OpenStack Infra (hudson-openstack)]

Related fix proposed to branch: master
Review: https://review.openstack.org/69670

[Jeremy Stanley (fungi)]

Grabbing this one since I want to see if we're still getting any similar incidents now that we've added more diagnostics so as to hopefully be able to identify the source.

[Clark Boylan (cboylan)]

Is this possibly a git bug? The host didn't appear to have an ipv6 interface but git decided to talk ipv6 anyways. Should we consider a git.ipv4.openstack.org A record that slaves can talk to instead of the git.o.o name which has a AAAA record?

[Jeremy Stanley (fungi)]

This hasn't happened in two weeks according to elastic-recheck (and much longer than that from what I know), so I think it's safe to not care about tracking this for now.

[Besmir Zanaj (besmirzanaj-gmail)]

I recreated the by downloading the devstack repository in another folder other than "/opt/stack".
When I ran ./stack.sh from "devstack" folder in /opt/stack then the error was not present anymore