Created attachment 989522 [details]
ovirt datastore glusterfs log showing the ipv6 lookup issue

Description of problem:
Using gluster with ovirt as a hyperconverged platform based on rhel7. by default rhel7 has ipv6 enabled. during configuration I see address resolution failures preventing ovirt hosted-engine deployment from working.


Version-Release number of selected component (if applicable):

glusterfs 3.6.2
ovirt 3.5
ovirt-hosted-engine-setup (from Sandro Bonazzola - 1.2.2)

How reproducible:
When ipv6 is enabled this happens every time hosted-engine --deploy is attempted

Steps to Reproduce:
1. have dns resolution working for ipv4 only
2. configure gluster (peers and volume)
3. attempt to run hosted-engine --deploy (with ipv6 enabled on the host)


Actual results:
[ INFO  ] Waiting for VDSM hardware info
[ INFO  ] Waiting for VDSM hardware info
[ INFO  ] Configuring the management bridge
[ ERROR ] Failed to execute stage 'Misc configuration': Connection to storage server failed
[ INFO  ] Stage: Clean up
[ INFO  ] Generating answer file '/var/lib/ovirt-hosted-engine-setup/answers/answers-20150203193700.conf'
[ INFO  ] Stage: Pre-termination
[ INFO  ] Stage: Termination

snip from glusterd log file
[2015-02-04 00:36:59.996195] E [common-utils.c:227:gf_resolve_ip6] 0-resolver: getaddrinfo failed (Name or service not known)
[2015-02-04 00:36:59.996292] E [name.c:242:af_inet_client_get_remote_sockaddr] 0-glusterfs: DNS resolution failed on host gprfc085-glfs.rhev-rhss.lab


Expected results:
If ipv6 is enabled, but DNS does not return an ipv6 entry, retry with ipv6 - only fail the request if lookups to both ipv6 and ipv4 fail.

Additional info:
Workaround was to disable ipv6 on the host - subsequent lookups are ipv4 only and name resoltion works as expected.

log for the storage domain that was attempted to be located is atteched - timestamp - 2015-02-04 00:36:39.959656