Description of problem: On commenting the "brick_dirs" section in conf file and running gdeploy for multiple volume creation, gdeploy fails with bad error message. It is not pointing to exact issue. Version-Release number of selected component (if applicable): gdeploy-2.0.2-14.el6rhs.noarch ansible-2.3.1.0-1.el6.noarch How reproducible: Always Steps to Reproduce: 1.Comment the brick_dirs section in the gdeploy conf file 2. run gdeploy 3. Actual results: Fails with bad error message Expected results: Should fail with exact error message Additional info: Traceback (most recent call last): File "/usr/bin/gdeploy", line 219, in <module> main(sys.argv[1:]) File "/usr/bin/gdeploy", line 208, in main call_features() File "/usr/lib/python2.6/site-packages/gdeploylib/call_features.py", line 39, in call_features map(get_feature_dir, Global.sections) File "/usr/lib/python2.6/site-packages/gdeploylib/call_features.py", line 94, in get_feature_dir section_dict, yml = feature_call(section_dict) File "/usr/lib/python2.6/site-packages/gdeployfeatures/volume/volume.py", line 23, in volume_create (section_dict['volname'], section_dict['brick_dirs'])) KeyError: 'brick_dirs'
Vivek, do you by chance have the config file for this? This works for me. The config file I am using: ===================== [hosts] 10.70.42.10 10.70.42.122 #[backend-setup] #devices=vdb #mountpoints=/gluster/brick1/ #brick_dirs=/gluster/brick1/f [volume] action=create volname=vol1 transport=tcp replica_count=2 #brick_dirs=/gluster/brick1/foo [volume2] action=create volname=vol2 transport=tcp replica_count=2 #brick_dirs=/gluster/brick1/bar ===================================== [sac@dhcp35-44 gdeploy_config]$ gdeploy -c vol-create.conf Error: 'brick_dirs' not provided for all the hosts.
Hi I am unable to reproduce the error messages again now with the below version gdeploy-2.0.2-14 ansible-2.3.2.0-1