[ovirt-users] Re: Connection issues when using gluster + infiniband + RDMA

My gluster volume is mounted with *.rdma. I cannot make changes to the switch as it doesn’t have any management to it as it is unmanaged. I can transfer files and run VMs from the gluster volume. I just receive these messages. I am using IBoIP. I have no problems doing a NFS over RDMA share

[ovirt-users] Re: oVirt 4.2.5 : VM snapshot creation does not work : command HSMGetAllTasksStatusesVDS failed: Could not acquire resource

On Mon, Aug 13, 2018 at 1:45 PM Aleksey Maksimov < aleksey.i.maksi...@yandex.ru> wrote: > We use oVirt 4.2.5.2-1.el7 (Hosted engine / 4 hosts in cluster / about > twenty virtual machines) > Virtual machine disks are located on the Data Domain from FC SAN. > Snapshots of all virtual machines are

[ovirt-users] Re: Connection issues when using gluster + infiniband + RDMA

try to leave datagramm mode, do not change mtu. it looks like gluster is connected over the tcp/ipoib. you might get packets drops with mtu> 2k. as i remember you should tune your ib switch for the mtu size,at least on the mellanox managed switch. if the gluster connected over

[ovirt-users] Connection issues when using gluster + infiniband + RDMA

I am trying to setup gluster using Infiniband and mounting it in RDMA mode. I have created a network hook so that it configures the interfaces as MTU=65520 and CONNECTED_MODE=yes. I have a second server doing NFS with RDMA over Infiniband with some VMs on it. When I try and transfer files to

[ovirt-users] Connection issues when using gluster + infiniband + RDMA

I am trying to setup gluster using Infiniband and mounting it in RDMA mode. I have created a network hook so that it configures the interfaces as MTU=65520 and CONNECTED_MODE=yes. I have a second server doing NFS with RDMA over Infiniband with some VMs on it. When I try and transfer files to

[ovirt-users] Passing through Quadro cards to guest

Has anyone experimented with this? I have had really good success with AMD cards, but consumer end graphics cards fail to load due to NVIDIA blocking them. I happened to have an NVIDIA Quadro P2000 lying around and I thought I would give it a shot. I really like these cards as they only require

[ovirt-users] "ISCSI multipathing" tab isn't appear in datacenter settings

Hello. We have 6 servers in our cluster, which use 2 storage through iSCSI connections. Each storage has 2 nodes. Each node has 2 IP addresses in two different VLANs. Each host has 2 networks in this VLANs, so, the iSCSI traffic is separated from other types of traffic. I want to turn on iSCSI

[ovirt-users] Re: Isolated POC

On Mon, Aug 13, 2018 at 4:22 PM Martin Schutte wrote: > Hi there Dan, > > This is another very good suggestion. I am familiar with the RHEV repos, > but not so much the upstream oVirt ones. As the POC is not based on RHEV, I > am going to have to do this for oVirt. > I doubt there is much of a

[ovirt-users] Re: Isolated POC

Hi there Dan, This is another very good suggestion. I am familiar with the RHEV repos, but not so much the upstream oVirt ones. As the POC is not based on RHEV, I am going to have to do this for oVirt. Regards, Martin. ___ Users mailing list --

[ovirt-users] Re: Isolated POC

I am using bare metal systems with all SSD drives. Going to do a vanilla install for oVirt. I have given consideration to the dd idea, as it is a good one. I will have to take one of the systems out of the rack to install it offsite and return it. It might be worth it though given the time

[ovirt-users] Re: Isolated POC

This has been done for RHEV more times than I care to remember. As simple as creating a local repo and copying it locally with reposync. You might drag in some unneeded extras, but disk space is cheap and time is not. So in short, clone the repos onto a USB drive, bring them to your environment,

[ovirt-users] Re: Isolated POC

Il Lun 13 Ago 2018, 13:02 Martin Schutte ha scritto: > Good day to all, > > I am in the process of setting up an oVirt proof of concept for a client. > There are some strict requirements for this POC. > > Firstly the entire POC must be isolated from the network, which means no > online repos. >

[ovirt-users] Re: Isolated POC

Hi there, Yes I can confirm that having a yum proxy host is extremely useful. I have used it myself on a few occasions. My problem is that the POC is completely isolated. No jump host to use as a proxy server. Separate switches and systems. So the only way I can see this working is by

[ovirt-users] Re: Isolated POC

Do not know if this is acceptable in your environment, but what I do is to use an http proxy for yum access. On the oVirt engine / compute nodes (CentOS 7) I set on /etc/yum.conf http_proxy=http://your-proxy-server:your-proxy-port http_caching=packages The proxy side in my case is an apache

[ovirt-users] Re: Isolated POC

Hi Martin, I had the same issue with the offline repo one year ago and for me it was quite difficult to setup. The repos, they need are changing from time to time and I always have to test on a dev system, where I first install the new ovirt repo rpm to see which changes have been made and

[ovirt-users] Isolated POC

Good day to all, I am in the process of setting up an oVirt proof of concept for a client. There are some strict requirements for this POC. Firstly the entire POC must be isolated from the network, which means no online repos. Secondly there is a time constraint for the completion of the

[ovirt-users] Re: oVirt node 4.2.x - custom multipath.conf and external storage availability

Hello Roberto I solved a similar problem by adding an additional line "# VDSM PRIVATE" to the beginning of the file /etc/multipath.conf For example, the beginning of the file looks like this: # VDSM REVISION 1.3 # VDSM PRIVATE … This prevents configuration overwriting.

[ovirt-users] oVirt 4.2.5 : VM snapshot creation does not work : command HSMGetAllTasksStatusesVDS failed: Could not acquire resource

Hello We use oVirt 4.2.5.2-1.el7 (Hosted engine / 4 hosts in cluster / about twenty virtual machines) Virtual machine disks are located on the Data Domain from FC SAN. Snapshots of all virtual machines are created normally. But for one virtual machine, we can not create a snapshot. When we try

[ovirt-users] Re: oVirt node 4.2.x - custom multipath.conf and external storage availability

Hi all any feedback about ? Thanks Il giorno mar 31 lug 2018 alle ore 13:45 Roberto Nunin ha scritto: > HI all > > How can I provide to my oVirt hosts a custom multipath.conf, considering > that we are using shared/mirrored FC storage from HPE 3PAR with high > availability of provided LUNs,

[ovirt-users] Re: ovirt engine service needs restart after network connectivity is lost and reestablished

On Fri, Aug 10, 2018 at 10:08 PM, wrote: > today our network administration did some upgrades on the networking > equipment, so the engine vlan went down for a while. Afterwards when it > came back up 3 hosts was found as non responding, I couldn't see anything > suspicious on the hosts, the