[ https://issues.apache.org/jira/browse/CLOUDSTACK-5779?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13902310#comment-13902310 ]
ASF subversion and git services commented on CLOUDSTACK-5779: ------------------------------------------------------------- Commit 2b1b4352bc6af037430cff76b38d3eca061de0f7 in cloudstack's branch refs/heads/master from [~yasker] [ https://git-wip-us.apache.org/repos/asf?p=cloudstack.git;h=2b1b435 ] Revert "CLOUDSTACK-5779: Fix missing clean up period for VR" This reverts commit 161e7d93ca9d03e9aa7e7be9e12d1ad337de9b14. Duplicate code, bad memory... > Consolidate all the VR commands execution to one resource > --------------------------------------------------------- > > Key: CLOUDSTACK-5779 > URL: https://issues.apache.org/jira/browse/CLOUDSTACK-5779 > Project: CloudStack > Issue Type: Improvement > Security Level: Public(Anyone can view this level - this is the > default.) > Reporter: Sheng Yang > Assignee: Sheng Yang > Priority: Critical > Fix For: 4.4.0 > > > Currently when anyone add one command to VR(which happened very frequently > these days), they need to modify every hypervisor resources to accommodate > the new command, which is really inconvenient as well as bug prone, since > e.g. the parameters formation would be duplicate across the hypervisors. > This improvement would make all VR commands to be executed by only one > resource, and in the future, any new command/improvement or bug fixes to VR > resource would only need to be done once in the VR resource. > This would be done through: > 1. Use routeProxy script for all VR commands, no more standalone script for > each VR scripts in the host. All VR command scripts would be moved to > /opt/cloud/bin for this purpose as well. > 2. Every hypervisor would provide an executor to VR resource, to execute the > command in giving hypervisor. -- This message was sent by Atlassian JIRA (v6.1.5#6160)