Description of problem: While creating a new FC storage domain, the FC LUN of Hosted_storage can also be elected to create storage domain on it. Version-Release number of selected component (if applicable): Red Hat Enterprise Virtualization 3.6.3 (hosted_engine setup with FC) How reproducible: Always. Steps to Reproduce: 1. Install Hosted_Engine using FC LUNs. 2. Try to create new domain, when the scan completes, Hosted_storage LUN can also be selected from the checklist. This would destroy HE env. 3. Actual results: Hosted_storage LUN can be selected. Expected results: Hosted Storage LUN should be grayed out/ Should not be available to select. Additional info: Attaching screenshots.
Roy, iirc you blocked this from happening, isn't it?
The auto import in 3.6 should register the domain's LUNs. Moving to SLA to take lead on this issue.
(In reply to Allon Mureinik from comment #4) > The auto import in 3.6 should register the domain's LUNs. Moving to SLA to > take lead on this issue. If a LUN is being used by a domain, why is it available for other domain creation and won't that fail in validation stage anyway?
(In reply to Roy Golan from comment #5) > (In reply to Allon Mureinik from comment #4) > > The auto import in 3.6 should register the domain's LUNs. Moving to SLA to > > take lead on this issue. > > If a LUN is being used by a domain, why is it available for other domain > creation and won't that fail in validation stage anyway? If it's properly registered in the DB, which seems the registration isn't doing.
The current plan is to take the special status from hosted engine storage domain and allow its usage for other VMs as well. All components are going to be updated to allow that and behave properly.
And Node 0 deployment will remove the "before import" window as the engine will know about hosted engine objects (VM, storage domain) right from the start.
Assuming Node 0 got merged in time for 4.2.0/4.2.1, please just verify that this is not possible in hosted engine deployed using the Node 0 method. There should no longer be any "before import" phase.
The Node 0 deployment was made the default method by https://gerrit.ovirt.org/#/c/83900/ and this issue will not longer happen there. Please test using ansible based deployment.
WARN: Bug status wasn't changed from MODIFIED to ON_QA due to the following reason: [Found non-acked flags: '{'rhevm-4.2-ga': '?'}', ] For more info please contact: rhv-devops: Bug status wasn't changed from MODIFIED to ON_QA due to the following reason: [Found non-acked flags: '{'rhevm-4.2-ga': '?'}', ] For more info please contact: rhv-devops
Created attachment 1385537 [details] Screenshot from 2018-01-24 13-59-09.png
Works for me on these conponents on FC deployment using ansible. ovirt-hosted-engine-ha-2.2.5-0.0.master.20180117160214.20180117160210.gitd5b58a8.el7.centos.noarch ovirt-hosted-engine-setup-2.2.8-0.0.master.20180123120416.gitfa69a78.el7.centos.noarch ovirt-engine-appliance-4.2-20180121.1.el7.centos.noarch Linux 3.10.0-693.17.1.el7.x86_64 #1 SMP Sun Jan 14 10:36:03 EST 2018 x86_64 x86_64 x86_64 GNU/Linux SHE LUN greyed out and not selectable, right after deployment is finished and there is no need of adding additional data storage domain for auto-import, for Node Zero any more. Moving to verified.
Since the problem described in this bug report should be resolved in a recent advisory, it has been closed with a resolution of ERRATA. For information on the advisory, and where to find the updated files, follow the link below. If the solution does not work for you, open a new bug report. https://access.redhat.com/errata/RHEA-2018:1488
BZ<2>Jira Resync