Created attachment 1395278 [details] FIle contains all.yml,osds.yml,hosts,ansible-playbook log Description of problem: The admin keyring is not copied onto the collocated nodes (rgw+osd node ,mds+osd node) even though copy_admin_key is set to true in rgws.yml and mdss.yml config file. Version-Release number of selected component (if applicable): ceph-ansible-3.0.23-1.el7cp.noarch.rpm How reproducible: Always Steps to Reproduce: 1. collocate RGW and MDS daemons with OSD( OSD+RGW on node & OSD+MDS on another) 2. In yaml conf file osds.yml, keep copy_admin_key to false(defaults) and in rgws.yml & mdss.yml set copy_admin_key to true 3. Run ansible playbook site-docker.yml Actual results: admin keyring is not copied onto the collocated nodes Expected results: admin keyring should be copied Additional info: Different values for a single parameter in mutliple files( here mdss/rgw yml and osds yml file) should have a conflict ..only one can override the other config,apparently in this case osds.yml file parameter value takes the precedence.Not sure if it is supposed to work that way and also which yml config file should take precedence in collocated scenarios either mdss/rgws.yml or osd.yml.
User needs to copy admin keyring to RGW/MDS using scp - From monitor node - #scp /etc/ceph/<cluster-name>.client.admin.keyring <rgw-or-mds>:/etc/ceph/
will be in v3.0.28
Erin, doc is fine thanks.
The issue has been fixed for a RGW node but we are facing same issue for MDS, clients and NFS nodes. Moving BZ back to ASSIGNED state.
Since the problem described in this bug report should be resolved in a recent advisory, it has been closed with a resolution of ERRATA. For information on the advisory, and where to find the updated files, follow the link below. If the solution does not work for you, open a new bug report. https://access.redhat.com/errata/RHBA-2018:1259