Description of problem: If condor_configure_pool is used to add features and set the schedd/qmf information, the param values entered by the user will be lost. Version-Release number of selected component (if applicable): condor-wallaby-4.0-6 How reproducible: 100% Steps to Reproduce: 1. condor_configure_pool -n <node> -a -f ExecuteNode,Master --schedds 2. Set schedd name 3. Apply config and provide CONDOR_HOST value 4. activate fails because CONDOR_HOST isn't set Actual results: Expected results: Additional info:
The issue only occurs if you attempt to set schedd or qmf info with the same command where you will be setting params (so adding params or features that have must_change params). There are a few ways to work around this issue: 1) Issue commands to add/remove schedd (--schedds) or qmf info (--qmfbroker) on separate command invocations from other activities. 2) Manually add the params/values with a separate command invocation.
Fixed on BZ700863-lose-param-settings
Technical note added. If any revisions are required, please edit the "Technical Notes" field accordingly. All revisions will be proofread by the Engineering Content Services team. New Contents: C: Using condor_configure_pool to set a feature with must_change parameters or parameters on their own at the same time as setting scheduler/qmfbroker information. C: Any param changes from the condor_configure_pool invocation besides the scheduler/qmfbroker information would be lost. F: Update the qmf group object before applying any scheudler/qmfbroker params. R: Using condor_configure_pool to set a feature with must_change parameters or parameters on their own at the same time as setting scheduler/qmfbroker information now works as expected
Reproduced on: condor-wallaby-client-4.0-6.el5 condor-wallaby-tools-4.0-6.el5 ruby-wallaby-0.10.5-5.el5 wallaby-0.10.5-5.el5 python-wallabyclient-4.0-6.el5 condor-wallaby-base-db-1.13-1.el5 wallaby-utils-0.10.5-5.el5 # condor_configure_pool -n `hostname` -a -f ExecuteNode,Master --schedds Enter the name of the default scheduler: schedd Is this a High Available Scheduler [y/N] ? Apply these changes [Y/n] ? y The following parameters need to be set for this configuration to be valid. CONDOR_HOST Set these parameters now ? [y/N] y CONDOR_HOST: localhost Configuration applied Create a named snapshot of this configuration [y/N] ? Activate the changes [y/N] ? y Activating configuration. This may take a while, please be patient The configuration is not valid Node: localhost Unsatisfied feature dependencies: NodeAccess Unset necessary parameters: CONDOR_HOST Configuration not activated
Retested over all supported platforms x86,x86_64/RHEL5,RHEL6 with: condor-wallaby-base-db-1.14-1.el6.noarch wallaby-utils-0.10.5-6.el6.noarch python-wallabyclient-4.1-4.el6.noarch wallaby-0.10.5-6.el6.noarch ruby-wallaby-0.10.5-6.el6.noarch condor-wallaby-tools-4.1-4.el6.noarch # condor_configure_pool -n `hostname` -a -f ExecuteNode,Master --schedds Enter the name of the default scheduler: schedd Is this a High Available Scheduler [y/N] ? Apply these changes [Y/n] ? y The following parameters need to be set for this configuration to be valid. CONDOR_HOST Set these parameters now ? [y/N] y CONDOR_HOST: localhost Configuration applied Create a named snapshot of this configuration [y/N] ? Activate the changes [y/N] ? y Activating configuration. This may take a while, please be patient Configuration activated The condor_configure_pool now takes both - feature parameter and schedd/qmf. >>> VERIFIED
An advisory has been issued which should help the problem described in this bug report. This report is therefore being closed with a resolution of ERRATA. For more information on therefore solution and/or where to find the updated files, please follow the link below. You may reopen this bug report if the solution does not work for you. http://rhn.redhat.com/errata/RHSA-2011-1249.html