Description of problem: After deploying the self-hosted engine on an NFS storage from the RHEV-H configuration menu, once accessing the manager through the portal, the default datacenter isn't initialized, no storage and VM are detected. Version-Release number of selected component (if applicable): rhev-hypervisor7-7.2-20160413.0 rhevm-appliance-20160413.0-1.x86_64.rhevm 3.6.5.3-0.1.el6 How reproducible: N/A, didn't try yet to re-deploy the appliance. Steps to Reproduce: 1. Deploy the self-hosted engine using the Manager appliance using a NFS storage 2. Login to the portal as admin 3. Navigate to datacenter, storage and VM menu Actual results: Default datacenter has unintialized state Storage menu is empty VM menu is empty Expected results: The datacenter should initialize The manager should detect the storage where it's itself installed. The VM of the manager should appear in the VM list. Additional info: Restarting the hypervisor and restarting the manager (using hosted-engine vm-stop)
Simone, does this sound familiar to you?
It's not a bug, it's just the initial status: the hosted-engine-storage domain is hosted-engine specific, it's up to the user to add the first regular storage domain for regular VMs. Only at that point the datacenter will become active and this will also trigger the auto-import procedure for the hosted-engine storage domain and for the engine VM. Alexis, could you please try adding a storage domain from the webadmin interface?
Thank you, I guess this bug should be marked as closed : after attaching a storage domain, I can see the VM and then access to it from "console" (via SPICE) to be able to enable ssh..(which seems a very strange way to make things to me, by the way, is SPICE more secured than ssh, which would explain disabling ssh root authenthication but enabling SPICE ?) There is no message explaining this behavior is normal and a the fact a datastore domain must be added. Should I open a bug for it ? Should I call it "request of functionality" or do you have specific terms for that ? Also, in the documentation, there is no link to the storage configuration (https://access.redhat.com/documentation/en-US/Red_Hat_Enterprise_Virtualization/3.6/html/Installation_Guide/chap-Configuring_Storage.html) from the self-hosted engine guide, should I report a bug for that too, about documentation ?
(In reply to ahauser from comment #3) > Thank you, I guess this bug should be marked as closed : after attaching a > storage domain, I can see the VM and then access to it from "console" (via > SPICE) to be able to enable ssh..(which seems a very strange way to make > things to me, by the way, is SPICE more secured than ssh, which would > explain disabling ssh root authenthication but enabling SPICE ?) But console requires still access, right? > > > There is no message explaining this behavior is normal and a the fact a > datastore domain must be added. > Should I open a bug for it ? Should I call it "request of functionality" or > do you have specific terms for that ? Missing documentation. > > Also, in the documentation, there is no link to the storage configuration > (https://access.redhat.com/documentation/en-US/ > Red_Hat_Enterprise_Virtualization/3.6/html/Installation_Guide/chap- > Configuring_Storage.html) from the self-hosted engine guide, should I report > a bug for that too, about documentation ? Yes, please.