Description of problem: When installer fails for whatever reason, in my case the iscsi gave some issues, /var/tmp/installvm* is not cleaned. This means that with a couple reruns, /var/tmp will be full: [root@rhv1 tmp]# ls -als | grep local 4 drwxrwxr-x. 2 vdsm kvm 4096 Mar 24 15:53 localvm_25EXd 4 drwxrwxr-x. 4 vdsm kvm 4096 Mar 24 15:22 localvmVvspJr [root@rhv1 tmp]# du -sh local* 4.0K localvm_25EXd 4.5G localvmVvspJr Version-Release number of selected component (if applicable): 4.3.8 How reproducible: Easy Steps to Reproduce: 1. Setup rhvh node 2. Run installer from cockpit. 3. Stop at storage. Actual results: Disk full. Expected results: Reuse of resources. Now, omnipotency does not exist; also meaning redeployment is actually a full redeploy. Additional info:
let's check why /var/tmp got filled. If bootstrap VM content was created there the code should be smart enough to know what to do with the content there.
ovirt-hosted-engine-cleanup tool cleans /var/tmp/localvm*. Can we run the cleanup tool from cockpit when clicking "Yes" in "Exit Wizard" only after "Prepare VM" stage?