Trunk VLAN Ovirt 4.2
by Sebastian Antunez N.
Hello Guys
Have a lab with 6 nodes Ovirt 4.2. All nodes have 4 nic 1GB and two nic are
for Management and and I will assign two nic in bonding for VM traffic
My problem is the following.
The switch Cisco have 2 VLAN (56 and 57) and I need to be able to create a
new network with the two vlan but I am not clear how to make it so that the
vlan go through this created network. I have read that I must create a
network with vlan 56 for example and assign it an IP, and then create the
network with vlan 57 and assign it an IP and later assign it to bonding.
Is what I indicate or should another process be correct?
Thanks for the help.
Sebastian
5 years, 11 months
Re: [ovirt-devel] Hard disk requirement
by Nir Soffer
On Sun, Jan 13, 2019 at 6:18 PM Hetz Ben Hamo <hetz(a)hetz.biz> wrote:
> Hi,
>
> The old oVirt (3.x?) ISO image wasn't requiring a big hard disk in order
> to install the node part on a physical machine, since the HE and other
> parts were running using NFS/iSCSI etc.
>
> oVirt 4.X if I understand correctly - does require hard disk on the node.
>
I think you can setup temporary storage on a diskless host using NFS or
by connecting to temporary LUN and setting up a file system on it.
Once the bootstrap engine is ready on the "local" storage, you we move
engine disk to shared storage, and you can remove the local storage.
Adding Simone to add more info.
> Can this requirement be avoided and just use an SD Card? Whats the minimum
> storage for the local node?
>
Another option is to use /dev/shm, if you have a server with lot of memory.
Note that this list is for ovirt developers. This question is more about
using ovirt, so the users
mailing list is better. Others users may already solved this issue and can
help more than
developers which have much less experience with actual deployment.
Nir
5 years, 11 months
hosted-engine deploy Restore from backup failure
by Callum Smith
Restoring from backup appears to work perfectly, until it decides it hasn't worked and throws out. I can't see any relevant error here, does anyone have any insight?
Regards,
Callum
--
Callum Smith
Research Computing Core
Wellcome Trust Centre for Human Genetics
University of Oxford
e. callum(a)well.ox.ac.uk<mailto:callum@well.ox.ac.uk>
5 years, 11 months
Ovirt VDI Solution network issue
by mdcarr＠meditech.com
Hello, first time poster looking for help on setting up the network for a VDI deployment. A little background, I have Ovirt engine running on a Vmware virtual machine and have added a physical host to serve up virtual machines running windows 7 to our developers. The host has a static IP address and when I create a virtual machine I can see the default network is attached and up but does not receive an IP address. It does have a MAC so I'm wondering if I need to have our network team assign an IP for that MAC? I will be creating around 20 vm's that would get wiped after use so assigning static ip's might cause issues. Am I missing something or is there a better way to setup the LAN for this host? Thanks for any help!
5 years, 11 months
Hypervisor host CPU utilization metrics
by Sigbjorn Lie-Soland
Hi,
Linux see twice the amount of CPUs for a host with SMT/hyperthreading enabled. Inspecting the individual cores presented to the operating system using mpstat shows that the guests are running primarily on the “physical cores”, which is expected as the system is SMT aware. The “hyperthreaded cores” are showing guest activity when the host cpu is heavily loaded. Using the traditional tools such as top and vmstat does not take hyperthreading into consideration when calculating the total cpu% usage, again, as expected.
How is the hypervisor host CPU utilization metrics in Ovirt calculated when a hypervisor host has SMT/hyperthreading enabled? Does Ovirt calculate in the same manner as the traditional vmstat/top/etc Linux tools, or does it only count CPU usage on the “physical cores” on the hypervisor? Does it read CPU metrics from the Intel PCM (https://github.com/opcm/pcm <https://github.com/opcm/pcm>) ?
Does the “Count Threads as Cores” setting affect how the host hypervisor CPU utilization metrics are calculated?
The reason for asking is when using tools for capacity and utilization reporting to attempt to establish cluster growth and to anticipate when and how much hypervisor capacity to add, we need the input utilization data to be a correct representation of what is happening at the hypervisor host.
Thanks.
Regards,
Siggi
5 years, 11 months
exporting VMs using vdsm-client
by Callum Smith
Is it possible to export VMs using vdsm-client to import into a new oVirt engine in a case of catastrophic failure such that they can be imported into a new setup?
Regards,
Callum
--
Callum Smith
Research Computing Core
Wellcome Trust Centre for Human Genetics
University of Oxford
e. callum(a)well.ox.ac.uk<mailto:callum@well.ox.ac.uk>
5 years, 11 months
[ANN] oVirt 4.3.0 First Release Candidate is now available for testing
by Sandro Bonazzola
The oVirt Project is pleased to announce the availability of the First
Release Candidate of oVirt 4.3.0, as of January 10th, 2018
This is pre-release software. This pre-release should not to be used in
production.
Please take a look at our community page[1] to learn how to ask questions
and interact with developers and users.
All issues or bugs should be reported via oVirt Bugzilla[2].
This update is the first release candidate of the 4.3.0 version.
This release brings more than 120 enhancements and more than 420 bug fixes
on top of oVirt 4.2 series.
What's new in oVirt 4.3.0?
* Q35 chipset, support booting using UEFI and Secure Boot
* Skylake-server and AMD EPYC support
* New smbus driver in windows guest tools
* Improved support for v2v
* Hundreds of bug fixes on top of oVirt 4.2 series
* New VM portal details page (see a preview here:
https://imgur.com/a/ExINpci)
* New Cluster upgrade UI
* OVN security groups
* IPv6 (static host addresses)
* LLDP Labeler
* Support of Neutron from RH OSP 13 as external network provider
* Support of using Skydive from RH OSP 14 as Tech Preview
* Support for 3.6 and 4.0 data centers, clusters and hosts were removed
* Now using PostgreSQL 10
This release is available now on x86_64 architecture for:
* Red Hat Enterprise Linux 7.6 or later
* CentOS Linux (or similar) 7.6 or later
This release supports Hypervisor Hosts on x86_64 and ppc64le architectures
for:
* Red Hat Enterprise Linux 7.6 or later
* CentOS Linux (or similar) 7.6 or later
* oVirt Node 4.3 (available for x86_64 only)
Experimental tech preview for x86_64 and s390x architectures for Fedora 28
is also included.
See the release notes draft [3] for installation / upgrade instructions and
a list of new features and bugs fixed.
Notes:
- oVirt Appliance is already available for both CentOS 7 and Fedora 28
(tech preview).
- oVirt Node NG is already available for both CentOS 7 and Fedora 28 (tech
preview).
Additional Resources:
* Read more about the oVirt 4.3.0 release highlights:
http://www.ovirt.org/release/4.3.0/
* Get more oVirt project updates on Twitter: https://twitter.com/ovirt
* Check out the latest project news on the oVirt blog:
http://www.ovirt.org/blog/
[1] https://www.ovirt.org/community/
[2] https://bugzilla.redhat.com/enter_bug.cgi?classification=oVirt
[3] http://www.ovirt.org/release/4.3.0/
[4] http://resources.ovirt.org/pub/ovirt-4.3-pre/iso/
--
SANDRO BONAZZOLA
MANAGER, SOFTWARE ENGINEERING, EMEA R&D RHV
Red Hat EMEA <https://www.redhat.com/>
sbonazzo(a)redhat.com
<https://red.ht/sig>
5 years, 11 months
quota_name field length missmatch in db schema
by Pavel Šipoš
Hi!
We have a problem with Quotas whose name has a length grater than 60
characters.
Such quotas can be created (or renamed) through oVirt-shell but cannot
be managed through the oVirt Engine interface. E.g. VMs in that quota
can't be started, created, quota comment can't be added...
It turns out that the problem is with the engine database definition of
"audit_log" table, that declares "quota_name" column as "character
varying(60)".
However "quota.quota_name" is defined as "character varying(65)".
I think this is a bug and should be fixed in the future.
Error example:
PL/pgSQL function insertauditlog(integer,timestamp with time
zone,integer,character varying,integer,text,uuid,character
varying,uuid,character varying,uuid,character varying,uuid,character
varying,uuid,character varying,uuid,character varying,uuid,character
varying,uuid,character varying,character varying,uuid,uuid,character
varying,text,boolean,uuid,text) line 9 at SQL statement; nested
exception is org.postgresql.util.PSQLException: ERROR: value too long
for type character varying(60)
Regards,
Pavel
5 years, 11 months
Datacenter Down Storage error
by ahuser＠web.de
Hi @all,
I have a big problem with my datacenter.
After pressing optimization for vm-storage in a active gluster volume are the vms paused. And now, i cant activate the storage in the Data-Center. I have try reboot all Servers.
It runs the latest Stable ovirt Version:
ovirt-engine-4.2.7.5-1.el7.noarch
vdsm-gluster-4.20.43-1.el7.x86_64
glusterfs-3.12.15-1.el7.x86_64
All Servers Up-to Date. Centos 7.6
Gluster Volume Replica 3 with Arbiter
/var/log/sanlock.log
set scheduler RR|RESET_ON_FORK priority 99 failed: Operation not permitted
VDSM ovnode3 command SpmStatusVDS failed: (22, 'Sanlock resource read failure', 'Invalid argument')
can anybody help me?
thx
5 years, 11 months
JSON internal error setting bonding using vdsm-client
by Callum Smith
Dear All,
I’m getting the error: “(code=32603, message=Internal JSON-RPC error: {‘reason’: “‘unicode’ object has no attribute ’sort’”})
when trying to apply networking configuration with vdsm-client -f bond.json Host setupNetworks
bond.json contains:
{
“networks”: {},
“bondings”: {
“bond0”: {
“nics”: “eno1+eno2”,
“options”: “mode=4”
}
},
“options”: {}
}
Of course python is handling all the arguments as unicode entities rather than strings. Any idea what might be wrong?
Regards,
Callum
--
Callum Smith
Research Computing Core
Wellcome Trust Centre for Human Genetics
University of Oxford
e. callum(a)well.ox.ac.uk<mailto:callum@well.ox.ac.uk>
5 years, 11 months
