On Tue, Jan 10, 2017 at 8:52 AM, Nathan Cutler <ncutler@xxxxxxx> wrote: >>> Sorry, I shouldn't have been so cryptic. I meant: if I prepare an >>> integration branch in ceph-ci and Shaman builds it, can you do a manual >>> test? >> >> >> Sure thing! > > > wip-hammer-backports now includes your PR and is ready for testing: > > http://tracker.ceph.com/issues/17151#note-45 > https://shaman.ceph.com/builds/ceph/wip-hammer-backports/ To replicate, I install the latest from the backports branch onto a single CentOS 7 machine: $ ceph-deploy install --dev=wip-hammer-backports 1.node.a ... [1.node.a][DEBUG ] Complete! [1.node.a][INFO ] Running command: sudo ceph --version [1.node.a][DEBUG ] ceph version 0.94.9-4522-g953992f (953992f4b8d3c1de632cc7182412f8997052d18c) I edit the ceph.conf to have a different hostname (but still resolvable) than the machine: $ cat ceph.conf [global] fsid = 99070b44-f70f-4379-9a6e-1be066a37bb5 mon_initial_members = node1 mon_host = 192.168.111.100 auth_cluster_required = cephx auth_service_required = cephx auth_client_required = cephx I call create-initial that will fail: $ ceph-deploy mon create-initial ... [node1][WARNIN] ******************************************************************************** [node1][WARNIN] provided hostname must match remote hostname [node1][WARNIN] provided hostname: node1 [node1][WARNIN] remote hostname: 1 [node1][WARNIN] monitors may not reach quorum and create-keys will not complete [node1][WARNIN] ******************************************************************************** ... [ceph_deploy.mon][ERROR ] Some monitors have still not reached quorum: [ceph_deploy.mon][ERROR ] node1 Now I log in and check the ceph-create-keys process which is still around: [vagrant@1 ~] ps aux | grep create-keys root 12491 0.1 1.4 53292 7352 ? S 14:41 0:00 python /usr/sbin/ceph-create-keys --cluster ceph -i 1 vagrant 13343 0.0 0.1 112648 956 pts/0 S+ 14:44 0:00 grep --color=auto create I call date to check how long it will be until it dies: [vagrant@1 ~]$ date Tue Jan 10 14:47:53 UTC 2017 Then check again after 10 minutes (the maximum the code waits before exiting): [vagrant@1 ~]$ date Tue Jan 10 14:55:28 UTC 2017 [vagrant@1 ~]$ ps aux | grep create vagrant 15690 0.0 0.1 112648 956 pts/0 S+ 14:55 0:00 grep --color=auto create And ceph-create-keys is no more. To double check this, I call ceph-create-keys manually with increased verbosity: [vagrant@1 ~]$ sudo python /usr/sbin/ceph-create-keys -v --cluster ceph -i 1 INFO:ceph-create-keys:ceph-mon is not in quorum: u'probing' INFO:ceph-create-keys:ceph-mon is not in quorum: u'probing' INFO:ceph-create-keys:ceph-mon is not in quorum: u'probing' INFO:ceph-create-keys:ceph-mon is not in quorum: u'probing' ... ceph-mon was not able to join quorum within 10 minutes [vagrant@1 ~]$ ceph --version ceph version 0.94.9-4522-g953992f (953992f4b8d3c1de632cc7182412f8997052d18c) > > Thanks, > Nathan -- To unsubscribe from this list: send the line "unsubscribe ceph-devel" in the body of a message to majordomo@xxxxxxxxxxxxxxx More majordomo info at http://vger.kernel.org/majordomo-info.html
