On Fri, Oct 25, 2013 at 5:10 AM, Laurent Durnez <laurent.durnez@xxxxxxx> wrote: > Hi everyone, > > I'm trying to set up a second cluster specifying its name but I can't bring > up the first mon nodes. The second cluster is on different hardware that my > first one. I'm following the quick start : > http://ceph.com/docs/master/start/quick-ceph-deploy > > I've reproduced it with 1 node : > 1) ceph-deploy --cluster cluster02 new cl02-cephnode01 > 2) ceph-deploy --cluster cluster02 install cl02-cephnode01 > 3) ceph-deploy --cluster cluster02 mon create cl02-cephnode01 > > And I've got the following log : > > [ceph_deploy.mon][INFO ] distro info: Debian 7.2 wheezy > [cl02-cephnode01][DEBUG ] determining if provided host has same hostname in > remote > [cl02-cephnode01][DEBUG ] deploying mon to cl02-cephnode01 > [cl02-cephnode01][DEBUG ] remote hostname: cl02-cephnode01 > [cl02-cephnode01][INFO ] write cluster configuration to > /etc/ceph/{cluster}.conf > [cl02-cephnode01][INFO ] creating path: > /var/lib/ceph/mon/cluster02-cl02-cephnode01 > [cl02-cephnode01][DEBUG ] checking for done path: > /var/lib/ceph/mon/cluster02-cl02-cephnode01/done > [cl02-cephnode01][DEBUG ] done path does not exist: > /var/lib/ceph/mon/cluster02-cl02-cephnode01/done > [cl02-cephnode01][INFO ] creating tmp path: /var/lib/ceph/tmp > [cl02-cephnode01][INFO ] creating keyring file: > /var/lib/ceph/tmp/cluster02-cl02-cephnode01.mon.keyring > [cl02-cephnode01][INFO ] create the monitor keyring file > [cl02-cephnode01][INFO ] Running command: ceph-mon --cluster cluster02 > --mkfs -i cl02-cephnode01 --keyring > /var/lib/ceph/tmp/cluster02-cl02-cephnode01.mon.keyring > [cl02-cephnode01][INFO ] ceph-mon: mon.noname-a 192.168.1.111:6789/0 is > local, renaming to mon.cl02-cephnode01 > [cl02-cephnode01][INFO ] ceph-mon: set fsid to > 6636a4f5-a839-4b9f-aff4-e8788fcd3434 > [cl02-cephnode01][INFO ] ceph-mon: created monfs at > /var/lib/ceph/mon/cluster02-cl02-cephnode01 for mon.cl02-cephnode01 > [cl02-cephnode01][INFO ] unlinking keyring file > /var/lib/ceph/tmp/cluster02-cl02-cephnode01.mon.keyring > [cl02-cephnode01][INFO ] create a done file to avoid re-doing the mon > deployment > [cl02-cephnode01][INFO ] create the init path if it does not exist > [cl02-cephnode01][INFO ] locating `service` executable... > [cl02-cephnode01][INFO ] found `service` executable: /usr/sbin/service > [cl02-cephnode01][INFO ] Running command: sudo /usr/sbin/service ceph -c > /etc/ceph/cluster02.conf start mon.cl02-cephnode01 > [cl02-cephnode01][INFO ] Running command: sudo ceph --admin-daemon > /var/run/ceph/ceph-mon.cl02-cephnode01.asok mon_status > [cl02-cephnode01][ERROR ] admin_socket: exception getting command > descriptions: [Errno 2] No such file or directory > [cl02-cephnode01][WARNIN] monitor: mon.cl02-cephnode01, might not be running > yet > [cl02-cephnode01][INFO ] Running command: sudo ceph --admin-daemon > /var/run/ceph/ceph-mon.cl02-cephnode01.asok mon_status > [cl02-cephnode01][ERROR ] admin_socket: exception getting command > descriptions: [Errno 2] No such file or directory > [cl02-cephnode01][WARNIN] monitor cl02-cephnode01 does not exist in monmap > [cl02-cephnode01][WARNIN] neither `public_addr` nor `public_network` keys > are defined for monitors > [cl02-cephnode01][WARNIN] monitors may not be able to form quorum > > On the mon node, I've got this line in the log : > 2013-10-25 10:22:26.427836 7f0d6d957780 0 ceph version 0.67.4 > (ad85b8bfafea6232d64cb7ba76a8b6e8252fa0c7), process ceph-mon, pid 2505 > > my cluster02.conf is the one that has been created by default : > [global] > fsid = 6636a4f5-a839-4b9f-aff4-e8788fcd3434 > mon_initial_members = cl02-cephnode01 > mon_host = 192.168.1.111 > auth_supported = cephx > osd_journal_size = 1024 > filestore_xattr_use_omap = true > > > However If I repeat the same steps without specifying the cluster name, my > 1st node comes up without problem : > [cl02-cephnode01][INFO ] monitor: mon.cl02-cephnode01 is running > > Any clue on what I'm missing? This is a problem with the recent addition of error checking and an issue was opened for it (but not yet completed) http://tracker.ceph.com/issues/6587 You should ignore (for now) those warnings and rely on specific checks for the mon in that host. This should be fixed soon though, stay tuned for that :) > > > Regards, > Laurent Durnez > > > _______________________________________________ > ceph-users mailing list > ceph-users@xxxxxxxxxxxxxx > http://lists.ceph.com/listinfo.cgi/ceph-users-ceph.com _______________________________________________ ceph-users mailing list ceph-users@xxxxxxxxxxxxxx http://lists.ceph.com/listinfo.cgi/ceph-users-ceph.com
 |