Created attachment 672810 [details] development.log Description of problem: The application status is changed after move inlcuding moving stopped application and idle application. All stopped and idle applications are changed to started after move part of development.log: 2013-01-05 01:23:42.705 [DEBUG] DEBUG: Getting existing app 'dancer' status before moving (pid:4024) 2013-01-05 01:23:42.715 [DEBUG] DEBUG: rpc_exec_direct: rpc_client=#<MCollective::RPC::Client:0x000000064920e0> (pid:4024) 2013-01-05 01:23:42.716 [DEBUG] DEBUG: rpc_client.custom_request('cartridge_do', {:cartridge=>"perl-5.10", :action=>"status", :args=>"'dancer' 'qgong8' '50e7a11c776856a80800019f'"}, ip-10-100-222-253, {'identity' => ip-10-100-222-253}) (pid:4024) 2013-01-05 01:23:43.455 [DEBUG] DEBUG: [#<MCollective::RPC::Result:0x00000006719250 @agent="openshift", @action="cartridge_do", @results={:sender=>"ip-10-100-222-253", :statuscode=>0, :statusmsg=>"OK", :data=>{:time=>nil, :output=>"CLIENT_RESULT: Application 'dancer' is either stopped or inaccessible\nATTR: status=ALREADY_STOPPED\nATTR: quota_blocks=1048576\nATTR: quota_files=40000\n", :exitcode=>0}}>] (pid:4024) 2013-01-05 01:23:43.457 [DEBUG] DEBUG: Gear component 'perl-5.10' was running (pid:4024) 2013-01-05 01:23:43.461 [DEBUG] DEBUG: Stopping existing app cartridge 'perl-5.10' before moving (pid:4024) Version-Release number of selected component (if applicable): fork_ami_refctr1_381 How reproducible: always Steps to Reproduce: 1.Setup multi noed env. 2.Create 2 apps, one is stopped, another is idled, move them. Actual results: After move, Stopped and Idle app status is changed to started [root@ip-10-38-21-76 openshift]# oo-admin-move --gear_uuid 50e7a11c776856a80800019f -i ip-10-38-21-76 URL: http://dancer-qgong8.dev.rhcloud.com Login: qgong App UUID: 50e7a11c776856a80800019f Gear UUID: 50e7a11c776856a80800019f DEBUG: Source district uuid: 50e79b17776856de32000001 DEBUG: Destination district uuid: 50e79b17776856de32000001 DEBUG: District unchanged keeping uid DEBUG: Getting existing app 'dancer' status before moving DEBUG: Gear component 'perl-5.10' was running DEBUG: Stopping existing app cartridge 'perl-5.10' before moving DEBUG: Force stopping existing app cartridge 'perl-5.10' before moving DEBUG: Creating new account for gear 'dancer' on ip-10-38-21-76 DEBUG: Moving content for app 'dancer', gear 'dancer' to ip-10-38-21-76 Identity added: /var/www/openshift/broker/config/keys/rsync_id_rsa (/var/www/openshift/broker/config/keys/rsync_id_rsa) Agent pid 4415 unset SSH_AUTH_SOCK; unset SSH_AGENT_PID; echo Agent pid 4415 killed; DEBUG: Moving system components for app 'dancer', gear 'dancer' to ip-10-38-21-76 Identity added: /var/www/openshift/broker/config/keys/rsync_id_rsa (/var/www/openshift/broker/config/keys/rsync_id_rsa) Agent pid 4485 unset SSH_AUTH_SOCK; unset SSH_AGENT_PID; echo Agent pid 4485 killed; DEBUG: Restarting httpd proxy for 'perl-5.10' on ip-10-38-21-76 DEBUG: Starting cartridge 'perl-5.10' in 'dancer' after move on ip-10-38-21-76 DEBUG: Fixing DNS and mongo for gear 'dancer' after move DEBUG: Changing server identity of 'dancer' from 'ip-10-100-222-253' to 'ip-10-38-21-76' DEBUG: Deconfiguring old app 'dancer' on ip-10-100-222-253 after move Successfully moved 'dancer' with gear uuid '50e7a11c776856a80800019f' from 'ip-10-100-222-253' to 'ip-10-38-21-76' Expected results: After move, app status should same as before move. Additional info:
fixed with rev#ff65b9907f1b6fa32f2f468ed4f2f3d6eaec7a7b ps - this did not make it into fork_ami_refctr1_395
Verified on fork_ami_refctr1_404 1. stop app [qgong@localhost dev]$ rhc app stop -a qsperl RESULT: qsperl stopped 2. move app 3. Check app status [qgong@localhost dev]$ rhc app status -a qsperl RESULT: Gear group perl-5.10 is stopped. The app's status keep same after move.