Re: [ovirt-users] Adding existing hosts to a new ovirt engine.

[Roy Golan]

On 05/06/2015 04:55 PM, Logan McNaughton wrote:
I'm also curious about what to do in this situation, perhaps someone 
from Red Hat has some insight?




- you can disable the automatic install when adding a host using a 
config value
On Mon, May 4, 2015 at 12:48 AM, Jan Stomphorst 
mailto:jstompho...@promedico.nl>> wrote:


Hi guys,

My engine died and I have hosts with virtual machines running. Now
I want to add the running host to a new engine. Without disturbing
the virtual machines on the host.

If I add an host it tries to install the ovirt packages, but there
already running.



problem is that once the host is up and the 1st master domain is up all 
your running VMs would be
discovered and auto added as external VMs and you will not be able to 
manage them.


I hope you can get a your db backups and have a new engine installation 
import the db.


it its a new engine installation then the certificate on the hosts is 
invalid. new installation creates new certs. if you have a copy, great, 
if no, you'll have to add and install (deploy) you're hosts.


if you can import you db dump then go to [1] to add a host without 
installing it


assuming your running 3.5, if you don't have a backup you can at least 
import the storage domains and import the VMs from the storage domains.



[1]  to add a host without auto installing

echo InstallVds > /tmp/props ; engine-config -s InstallVds=false -p 
/tmp/props


restart engine for the configuration to take effect





How do I resolve this.. J

Regards,

Jan Stomphorst


___
Users mailing list
Users@ovirt.org 
http://lists.ovirt.org/mailman/listinfo/users




___
Users mailing list
Users@ovirt.org
http://lists.ovirt.org/mailman/listinfo/users


___
Users mailing list
Users@ovirt.org
http://lists.ovirt.org/mailman/listinfo/users

Re: [ovirt-users] R: PXE boot of a VM on vdsm don't read DHCP offer

[Rik Theys]

Hi,

Is your DHCP server a VM running on the same host? I've seem some
strange issues where a VM could not obtain a DHCP lease if it was
running on the same physical machine as the client. If this is the case,
I can look up what I had to change, otherwise ignore it.

Regards,

Rik


Further info about this case:

An already installed and running VM (Centos7) with static IPv4 assignment, if  
changed to DHCP mode, do not acquire the
IP address.

 

In this case, tcpdump taken on the VM, do not show DHCP offer packets that are 
instead seen on the host bond interface.

Seems that something is filtering DHCP offers between host physical eth 
interfaces and VM virtio eth interface.

Physical servers on the same VLAN keep DHCP offers and boot from PXE correctly.

 

Roberto

 

 

 

>Hi all

 

>We are using oVirt engine 3.5.1-0.0 on Centos 6.6

>We are deploying two hosts with vdsm-4.16.10-8.gitc937927.el7.x86_64

>No hosted-engine, it run on a dedicates VM, outside oVirt.

 

>Behavior: PXE boot of a VM, ends in timeout (0x4c106035), instead to accept 
the DHCP offer coming from DHCP server.

>Tcpdump capture started on the vdsm host, bond0 interface shows clearly that 
DHCP offers reach the vdsm interfaces
>three times before PXE client ends in timeout.

>Incoming DHCP offer is correctly tagged when it comes to the bond0 interface 
and forwarded to the bond0.bridge
>interface.

>PXE simply ignore it. PXE version is gPXE 0.9.7.

>bond0.bridge interface is already setup with STP=off and DELAY=0.

 

>If we install a VM using command  line boot parameters, VM install & run fine. 
The issue is only related to PXE
process, >when it is expected to use the DHCP offer.

 

>I can provide tcpdump capture, but I’ve not attached to the email because I’m 
quite new of the community and don’t
>know if it is allowed/correct.

 

>On another host, under the same engine, running 
vdsm-4.16.12-7.gita30da75.el6.x86_64 on Centos6.6, this behavior is
>not happening, everything works fine.

 

>Any idea/suggestion/further investigation ?

>Thanks for attention

>Best regards

 

 

Roberto Nunin

Infrastructure Manager

Italy

 

 

Here are interfaces configs:

eno1:

DEVICE="eno1"

HWADDR="38:63:bb:4a:47:b0"

MASTER="bond0"

NM_CONTROLLED="no"

ONBOOT="yes"

SLAVE="yes"

eno2:

DEVICE="eno2"

HWADDR="38:63:bb:4a:47:b4"

MASTER="bond0"

NM_CONTROLLED="no"

ONBOOT="yes"

SLAVE="yes"

bond0:

BONDING_OPTS="mode=4 miimon=100"

DEVICE="bond0"

NM_CONTROLLED="no"

ONBOOT="yes"

TYPE="Bond"

bond0.3500:

DEVICE=bond0.3500

VLAN=yes

BRIDGE=DMZ3_DEV

ONBOOT=no

MTU=1500

NM_CONTROLLED=no

HOTPLUG=no

DMZ3_DEV:

DEVICE=DMZ3_DEV

TYPE=Bridge

DELAY=0

STP=off

ONBOOT=no

MTU=1500

DEFROUTE=no

NM_CONTROLLED=no

HOTPLUG=no

 

 

 


___
Questo messaggio e' indirizzato esclusivamente al destinatario indicato e 
potrebbe contenere informazioni
confidenziali, riservate o proprietarie. Qualora la presente venisse ricevuta 
per errore, si prega di segnalarlo
immediatamente al mittente, cancellando l'originale e ogni sua copia e 
distruggendo eventuali copie cartacee. Ogni
altro uso e' strettamente proibito e potrebbe essere fonte di violazione di 
legge.

This message is for the designated recipient only and may contain privileged, 
proprietary, or otherwise private
information. If you have received it in error, please notify the sender 
immediately, deleting the original and all
copies and destroying any hard copies. Any other use is strictly prohibited and 
may be unlawful.

___
Users mailing list
Users@ovirt.org
http://lists.ovirt.org/mailman/listinfo/users

Re: [ovirt-users] Local Storage

[Scott Worthington]

On 5/7/2015 12:11 PM, Andrea Ghelardi wrote:
> It is true that you cannot mix local and shared storage types.
> 
> However, it is actually possible tricking local storage to be presented as a
> shared iscsi volume and then passed to ovirt
> 
> There are drawbacks but we are actively using this solution.
> 
> Cheers
> 
> AG

Hello,

The major point that needs to be stressed again and again
on this mailing list is:

At this time, it is best practice to _NEVER_ combine the
shared storage onto the cluster hosts.

There are MANY many MANY posts from people with down oVirt
environments because they combined their shared storage
on their hosted cluster (especially gluster storage).

They do not fully understand the horrible consequences that will
happen when 1) a host cluster node gets fenced, 2) their gluster
cluster becomes split-brained because they failed to understand
gluster and quorums, or 3) a network connectivity issue of some
sort breaks the storage cluster on one (or many) node(s) and then
the issue cascades to the other systems bringing the entire shared
storage + clustered hosts to a grinding halt.

As a consequence, this mailing list gets flooded with pleas for help
from a crashed oVirt setup that did not follow best practices.

In my opinion, those who combine shared-storage onto the host-nodes
need to pull the plug on one of their hosts, and then work through
the numerous issues that will result in their hyper-converged
configuration.

If the SysAdmin does not have that experience of working through
all of the problems that result from hyper-converging shared
storage on the host nodes, then they certainly need to avoid
combing shared storage on the host nodes.

Good Luck!
--ScottW

___
Users mailing list
Users@ovirt.org
http://lists.ovirt.org/mailman/listinfo/users

Re: [ovirt-users] adding LINKDELAY to ovirtmgmt when vdsm configures it

[Jason Keltz]

On 05/06/2015 10:48 AM, Jason Keltz wrote:

Hi.

Is it possible through engine to specify configuration parameters for 
network interfaces that must be passed through on initialization?  For 
example, I need em1 and em2 configured with "LINKDELAY=10".  I can add 
it after the fact, but I don't want to do that.  I want vdsm to 
configure the interface with that setup already the first time.


Jason.


Argh x 10.

I have 2 x Intel 10G ports on an Intel X540 adapter.  If I kickstart a 
brand new host, and configure em1 with "LINKDELAY=10", the system boots, 
and gets an IP via DHCP.  If I then add the host through engine, vdsm 
reconfigures em1 and removes LINKDELAY.  Now, since the network adapter 
takes a little extra time to initialize, install fails.  If I add 
"LINKDELAY=10" back manually to 
/etc/sysconfig/network-scripts/ifcfg-em1, VDSM will happily remove it on 
the next reboot.  If I modify 
/var/lib/vdsm/persistence/netconf/nets/ovirtmgmt, and insert: 
"linkdelay":"10", and reboot, vdsm fills it in for me as I'd expect, but 
it's too late - engine thinks the install already failed!  I can delete 
the "failed install" host, and re-do the install.  This time it will 
work, but then my instructions are messy... "Install the host, let it 
fail, modify a file manually, re-install"... clearly, not clear!!   
Finally, as a last attempt, I've tried to copy 
/var/lib/vdsm/persistence/netconf/nets and bonds to the host during 
kickstart, hoping that when vdsm initialized, it would see those files 
there, and I'd have all 4 networks configured already for me. The result 
is weird.  ovirtmgmt configured fine including LINKDELAY=10.The 
other ones did not.I suspect it's a bad idea to write directly to 
/var/lib/vdsm/persistence/netconf ... What I want to do doesn't seem 
like it should be too complicated...


Ideally, when you add a host through engine, you should be able to 
specify custom network parameters.  Likewise, when you add the other 
interfaces, you should be able to do the same.  Any ideas how to solve 
this problem?


Jason.


___
Users mailing list
Users@ovirt.org
http://lists.ovirt.org/mailman/listinfo/users

Re: [ovirt-users] Local Storage

[Andrea Ghelardi]

It is true that you cannot mix local and shared storage types.

However, it is actually possible tricking local storage to be presented as
a shared iscsi volume and then passed to ovirt

There are drawbacks but we are actively using this solution.

Cheers

AG



*From:* users-boun...@ovirt.org [mailto:users-boun...@ovirt.org] *On Behalf
Of *Aharon Canan
*Sent:* Sunday, May 03, 2015 10:54 AM
*To:* Marcin Fabiańczyk
*Cc:* users@ovirt.org
*Subject:* Re: [ovirt-users] Local Storage



you can't



Host is part of cluster which is part of DC.

DC can be shared or local only, can't be both.







Regards,
__
*Aharon Canan*
int phone - 8272036
ext phone - +97297692036
email - aca...@redhat.com


--

*From: *"Marcin Fabiańczyk" 
*To: *users@ovirt.org
*Sent: *Friday, May 1, 2015 1:48:54 PM
*Subject: *[ovirt-users] Local Storage



Hello



I have create DC with a shared storage but I would like also add my
servers local
resources. It is possible in ovirt-hypervisior?



Regards.

MF




___
Users mailing list
Users@ovirt.org
http://lists.ovirt.org/mailman/listinfo/users
___
Users mailing list
Users@ovirt.org
http://lists.ovirt.org/mailman/listinfo/users

[ovirt-users] missing disk after storage domain expansion

[Andrea Ghelardi]

Hi gentlemen,



I recently found an error on 1 of my storages: it was complaining about no
free space but the VM was running and disk operational.

Since I needed to perform some maintenance on the VM, I shut it down and at
restart VM couldn’t boot up properly.

Checked VM via console and a disk was missing. Edited fstab (luckily this
disk was not root but heck! It had a Sybase DB on it!) and restarted VM
this time ok.



Since the disk resides on the dstore with no space, I expanded the iSCSI
LUN, then refreshed multipath on hosts, then resized PVs and now ovirt is
showing the correct size (logs do not complain anymore on no free space).



BUUUT



Now disk is missing. It is not shown anymore on Disks tab nor anywhere else.

Problem is that storage shows 214GB occupancy (size of the missing disk) so
data is there but cannot find it anymore.



Logs show original disk creation, errors from the lack of space, refresh of
the storage size and then no more references on the disk.



What can I do to find those missing ~210GBs?



Cheers

Andrea Ghelardi
___
Users mailing list
Users@ovirt.org
http://lists.ovirt.org/mailman/listinfo/users

[ovirt-users] R: PXE boot of a VM on vdsm don't read DHCP offer

[NUNIN Roberto]

Further info about this case:

An already installed and running VM (Centos7) with static IPv4 assignment, if  
changed to DHCP mode, do not acquire the IP address.

In this case, tcpdump taken on the VM, do not show DHCP offer packets that are 
instead seen on the host bond interface.
Seems that something is filtering DHCP offers between host physical eth 
interfaces and VM virtio eth interface.
Physical servers on the same VLAN keep DHCP offers and boot from PXE correctly.

Roberto



>Hi all

>We are using oVirt engine 3.5.1-0.0 on Centos 6.6
>We are deploying two hosts with vdsm-4.16.10-8.gitc937927.el7.x86_64
>No hosted-engine, it run on a dedicates VM, outside oVirt.

>Behavior: PXE boot of a VM, ends in timeout (0x4c106035), instead to accept 
>the DHCP offer coming from DHCP server.
>Tcpdump capture started on the vdsm host, bond0 interface shows clearly that 
>DHCP offers reach the vdsm interfaces >three times before PXE client ends in 
>timeout.
>Incoming DHCP offer is correctly tagged when it comes to the bond0 interface 
>and forwarded to the bond0.bridge >interface.
>PXE simply ignore it. PXE version is gPXE 0.9.7.
>bond0.bridge interface is already setup with STP=off and DELAY=0.

>If we install a VM using command  line boot parameters, VM install & run fine. 
>The issue is only related to PXE process, >when it is expected to use the DHCP 
>offer.

>I can provide tcpdump capture, but I've not attached to the email because I'm 
>quite new of the community and don't >know if it is allowed/correct.

>On another host, under the same engine, running 
>vdsm-4.16.12-7.gita30da75.el6.x86_64 on Centos6.6, this behavior is >not 
>happening, everything works fine.

>Any idea/suggestion/further investigation ?
>Thanks for attention
>Best regards


Roberto Nunin
Infrastructure Manager
Italy


Here are interfaces configs:
eno1:
DEVICE="eno1"
HWADDR="38:63:bb:4a:47:b0"
MASTER="bond0"
NM_CONTROLLED="no"
ONBOOT="yes"
SLAVE="yes"
eno2:
DEVICE="eno2"
HWADDR="38:63:bb:4a:47:b4"
MASTER="bond0"
NM_CONTROLLED="no"
ONBOOT="yes"
SLAVE="yes"
bond0:
BONDING_OPTS="mode=4 miimon=100"
DEVICE="bond0"
NM_CONTROLLED="no"
ONBOOT="yes"
TYPE="Bond"
bond0.3500:
DEVICE=bond0.3500
VLAN=yes
BRIDGE=DMZ3_DEV
ONBOOT=no
MTU=1500
NM_CONTROLLED=no
HOTPLUG=no
DMZ3_DEV:
DEVICE=DMZ3_DEV
TYPE=Bridge
DELAY=0
STP=off
ONBOOT=no
MTU=1500
DEFROUTE=no
NM_CONTROLLED=no
HOTPLUG=no





Questo messaggio e' indirizzato esclusivamente al destinatario indicato e 
potrebbe contenere informazioni confidenziali, riservate o proprietarie. 
Qualora la presente venisse ricevuta per errore, si prega di segnalarlo 
immediatamente al mittente, cancellando l'originale e ogni sua copia e 
distruggendo eventuali copie cartacee. Ogni altro uso e' strettamente proibito 
e potrebbe essere fonte di violazione di legge.

This message is for the designated recipient only and may contain privileged, 
proprietary, or otherwise private information. If you have received it in 
error, please notify the sender immediately, deleting the original and all 
copies and destroying any hard copies. Any other use is strictly prohibited and 
may be unlawful.
___
Users mailing list
Users@ovirt.org
http://lists.ovirt.org/mailman/listinfo/users

Re: [ovirt-users] selectively disabling IPv6 on bridges

[Dan Kenigsberg]

On Thu, May 07, 2015 at 01:06:32PM +0200, Lukáš Nykrýn wrote:
> Dan Kenigsberg píše v Čt 07. 05. 2015 v 11:46 +0100:
> > On Wed, May 06, 2015 at 01:53:35PM +0100, Dan Kenigsberg wrote:
> > > On Wed, May 06, 2015 at 01:28:30PM +0200, Rik Theys wrote:
> > > > Hi,
> > > > 
> > > > I'm looking for a way to selectively disable IPv6 on the bridge 
> > > > interfaces
> > > > on the oVirt hosts.
> > > > 
> > > > When oVirt creates the bridges for all logical networks on the host, it
> > > > keeps the default settings for IPv6 which means all bridges get a 
> > > > link-local
> > > > address and accept router advertisements.
> > > > 
> > > > When a VM is created on the logical network, it can now reach the host 
> > > > over
> > > > IPv6 (but not over IPv4 if no IP address has been assigned on the 
> > > > host). If
> > > > it sends out a router advertisement it can even create a global IPv6 
> > > > address
> > > > (haven't tested this).
> > > > 
> > > > How can I prevent this?
> > > > 
> > > > I would like to prevent the guest from IPv6 access to the host but the 
> > > > guest
> > > > itself still needs IPv6 access (global IPv6 addresses).
> > > > 
> > > > Is it sufficient to create a sysctl config file that says:
> > > > 
> > > > net.ipv6.conf.default.disable_ipv6 = 1
> > > 
> > > Yes, I believe that this would do the trick. For any newly-created
> > > device on the system, regardless of ovirt bridges.
> > > 
> > > I now see that el7 has changed the default for IPV6INIT to "yes". We
> > > should be more prudent and set IPV6INIT=no on all our devices.
> > 
> > Lukáš, it seems that setting IPV6INIT=no is not enough:
> > 
> > IPV6INIT=yes|no
> >   Enable or disable IPv6 static, DHCP, or autoconf configuration for 
> > this interface
> >   Default: yes
> > 
> > The bridge still gets a link-local ipv6 address anyway. Is there an 
> > initscript
> > means to disable this completely, or should we resort to
> > /proc/sys/net/ipv6/conf//disable_ipv6 ?
> > 
> > Dan.
> 
> You should disable this in kernel.  IPV6INIT=no basically means that
> network-scripts will not touch it. But kernel will setup the link-local
> address.

Thanks.

On Thu, May 07, 2015 at 01:09:15PM +0200, Rik Theys wrote:
> 
> I think you also have to disable this on the physical interface that's part
> of the bridge to fully disable this?

yes, we should disable ipv6 for all devices that have Layer-2
accessibility from the VMs.

Dan.
___
Users mailing list
Users@ovirt.org
http://lists.ovirt.org/mailman/listinfo/users

[ovirt-users] Invitation: Office Hours for oVirt Community - Integration @ Wed May 13, 2015 4pm - 5pm (sbona...@redhat.com)

[Sandro Bonazzola]

BEGIN:VCALENDAR
PRODID:-//Google Inc//Google Calendar 70.9054//EN
VERSION:2.0
CALSCALE:GREGORIAN
METHOD:REQUEST
BEGIN:VEVENT
DTSTART:20150513T14Z
DTEND:20150513T15Z
DTSTAMP:20150507T124544Z
ORGANIZER;CN=Sandro Bonazzola:mailto:sbona...@redhat.com
UID:7as3vig9m3o2l2ku55v2rlj...@google.com
ATTENDEE;CUTYPE=INDIVIDUAL;ROLE=REQ-PARTICIPANT;PARTSTAT=NEEDS-ACTION;RSVP=
 TRUE;CN=de...@ovirt.org;X-NUM-GUESTS=0:mailto:de...@ovirt.org
ATTENDEE;CUTYPE=INDIVIDUAL;ROLE=REQ-PARTICIPANT;PARTSTAT=NEEDS-ACTION;RSVP=
 TRUE;CN=users@ovirt.org;X-NUM-GUESTS=0:mailto:users@ovirt.org
ATTENDEE;CUTYPE=INDIVIDUAL;ROLE=REQ-PARTICIPANT;PARTSTAT=ACCEPTED;RSVP=TRUE
 ;CN=Sandro Bonazzola;X-NUM-GUESTS=0:mailto:sbona...@redhat.com
CREATED:20150507T124544Z
DESCRIPTION:Following past oVirt Sync Meeting decisions\, I'll be around in
  #ovirt channel trying to answer questions and discussing any cross-team/co
 mmunity-wide issues as needed.\nView your event at https://www.google.com/c
 alendar/event?action=VIEW&eid=N2FzM3ZpZzltM28ybDJrdTU1djJybGoxZW8gdXNlcnNAb
 3ZpcnQub3Jn&tok=MTkjc2JvbmF6em9AcmVkaGF0LmNvbTdiMDViYmUxYTFjZjFhNjVhYTk2MTl
 kZjY5NGZlNTM1NzFjYzU4ZDc&ctz=Europe/Rome&hl=en.
LAST-MODIFIED:20150507T124544Z
LOCATION:
SEQUENCE:0
STATUS:CONFIRMED
SUMMARY:Office Hours for oVirt Community - Integration
TRANSP:OPAQUE
END:VEVENT
END:VCALENDAR


invite.ics
Description: application/ics
___
Users mailing list
Users@ovirt.org
http://lists.ovirt.org/mailman/listinfo/users

Re: [ovirt-users] [ovirt-devel] vdsClient getVGList returned Empty

[Nir Soffer]

> Hi, guys
> I have a vdsm host, when i use "vgs" command, it will display 2 volume group
> which already exist. But when i use "vdsClient -s 0 getVGList", it displays
> nothing. That means vdsm cannot see the existed vg. Why? Does anybody have
> any idea?
> 
> P.S. In fact, i am trying to resolve the problem of "Cannot find master
> domain"

Can you shared the output of:

pvscan --cache && lvs -o all
pvscan --cache && vgs -o all

Nir
___
Users mailing list
Users@ovirt.org
http://lists.ovirt.org/mailman/listinfo/users

[ovirt-users] Get involved in oVirt project! May edition

[Sandro Bonazzola]

Hi,
have you got some free time and do you want to get involved in oVirt 
project?

Do you like the idea of having fresh disk images of recent distribution in 
oVirt Gluster repository?
You can help us by testing existing online images (like 
https://getfedora.org/en/cloud/download/) ensuring they works with cloud-init
or creating one yourself and report your success to de...@ovirt.org.
We'll be happy to upload the images once these are ready.


Do you like Debian and do you have some programming skill?
Help us getting VDSM running on it! Debian Jessie has been released a couple of 
weeks ago and it's a good time for giving it a try.
You can follow the progress here: http://www.ovirt.org/VDSM_on_Debian

Here are some bugs you can hopefully fix in less that one day or you can just 
try to reproduce providing info:

Bug ID  Status  Whiteboard  Summary
1065350 POSTintegration hosted-engine should prompt a question 
at the user when the host was already a host in the engine
1059952 NEW integration hosted-engine --deploy (additional 
host) will fail if the engine is not using the default self-signed CA
1083104 ASSIGNEDintegration engine-setup --offline does not update 
versionlock


Do you want something easier?

Bug ID  Whiteboard  Status  Summary
1174285 i18nNEW [de-DE] "Live Snapshot Support" reads "Live 
Snapsnot Support"
772931  infra   NEW [RFE] Reports should include the name of the 
oVirt engine
1143817 integration NEW [TEXT ONLY] - Hosted Engine - Instructions for 
FQDN are not clear enough
1156060 integration NEW [text] engine admin password prompt consistency
1115059 network NEW Incomplete error message when adding VNIC 
profile to running VM
734120  storage NEW [RFE] VDSM: use virt-sparsify/zerofree to 
reduce image size


Do you love "DevOps?", you count stable builds in jenkins ci while trying to 
fall a sleep?
Then oVirt infra team is looking for you!, join the infra team and dive in to 
do the newest and coolest devops tools today!
Here are some of our open tasks you can help with: 
https://fedorahosted.org/ovirt/report/1

Don't miss next week oVirt infra hackathon[8]!


You don't have programming skills, not enough time for DevOps but you want 
still to contribute?
Here are some bugs you can take care of, also without writing a line of code:
https://bugzilla.redhat.com/buglist.cgi?quicksearch=product%3Aovirt%20whiteboard%3Adocs

Do you prefer to test things? We have some test cases[5] you can try using 
nightly snapshots[6]

Do you want to contribute test cases? Most of the features[7] included in oVirt 
are missing a test case, you're welcome to contribute one!

Is this the first time you try to contribute to oVirt project?
You can start from here [1][2]!

You don't know gerrit very well? You can find some more docs here [3].

Any other question about development? Feel free to ask on de...@ovirt.org or on 
irc channel[4].

Let us know you're getting involved, present yourself and tell us what you're 
going to do, you'll be welcome!

[1] http://www.ovirt.org/Develop
[2] http://www.ovirt.org/Working_with_oVirt_Gerrit
[3] https://gerrit-review.googlesource.com/Documentation
[4] http://www.ovirt.org/Community
[5] http://www.ovirt.org/Category:TestCase
[6] http://www.ovirt.org/Install_nightly_snapshot
[7] http://www.ovirt.org/Category:Feature
[8] http://lists.ovirt.org/pipermail/devel/2015-May/010426.html

-- 
Sandro Bonazzola
Better technology. Faster innovation. Powered by community collaboration.
See how it works at redhat.com
___
Users mailing list
Users@ovirt.org
http://lists.ovirt.org/mailman/listinfo/users

Re: [ovirt-users] selectively disabling IPv6 on bridges

[Rik Theys]

Hi,

On 05/07/2015 12:46 PM, Dan Kenigsberg wrote:

On Wed, May 06, 2015 at 01:53:35PM +0100, Dan Kenigsberg wrote:

On Wed, May 06, 2015 at 01:28:30PM +0200, Rik Theys wrote:

Hi,

I'm looking for a way to selectively disable IPv6 on the bridge interfaces
on the oVirt hosts.

When oVirt creates the bridges for all logical networks on the host, it
keeps the default settings for IPv6 which means all bridges get a link-local
address and accept router advertisements.

When a VM is created on the logical network, it can now reach the host over
IPv6 (but not over IPv4 if no IP address has been assigned on the host). If
it sends out a router advertisement it can even create a global IPv6 address
(haven't tested this).

How can I prevent this?

I would like to prevent the guest from IPv6 access to the host but the guest
itself still needs IPv6 access (global IPv6 addresses).

Is it sufficient to create a sysctl config file that says:

net.ipv6.conf.default.disable_ipv6 = 1


Yes, I believe that this would do the trick. For any newly-created
device on the system, regardless of ovirt bridges.

I now see that el7 has changed the default for IPV6INIT to "yes". We
should be more prudent and set IPV6INIT=no on all our devices.


Lukáš, it seems that setting IPV6INIT=no is not enough:

 IPV6INIT=yes|no
   Enable or disable IPv6 static, DHCP, or autoconf configuration for this 
interface
   Default: yes

The bridge still gets a link-local ipv6 address anyway. Is there an initscript
means to disable this completely, or should we resort to
/proc/sys/net/ipv6/conf//disable_ipv6 ?


I think you also have to disable this on the physical interface that's 
part of the bridge to fully disable this?


Regards,

Rik

--
Rik Theys
System Engineer
KU Leuven - Dept. Elektrotechniek (ESAT)
Kasteelpark Arenberg 10 bus 2440  - B-3001 Leuven-Heverlee
+32(0)16/32.11.07

<>
___
Users mailing list
Users@ovirt.org
http://lists.ovirt.org/mailman/listinfo/users

Re: [ovirt-users] About installing ovirt - ovirt-engine not found

[Roy Golan]

On 04/30/2015 04:06 PM, Simone Tiraboschi wrote:


- Original Message -

From: "Lars Nielsen" 
To: "Simone Tiraboschi" 
Cc: "users" 
Sent: Thursday, April 30, 2015 2:29:05 PM
Subject: Re: [ovirt-users] About installing ovirt - ovirt-engine not found



On 30/04/15 14:21, Simone Tiraboschi wrote:

- Original Message -

From: "Lars Nielsen" 
To: "Simone Tiraboschi" 
Cc: "users" 
Sent: Thursday, April 30, 2015 2:08:10 PM
Subject: Re: [ovirt-users] About installing ovirt - ovirt-engine not found



On 30/04/15 13:55, Simone Tiraboschi wrote:

- Original Message -

From: "Lars Nielsen" 
To: "users" 
Sent: Thursday, April 30, 2015 1:45:29 PM
Subject: [ovirt-users] About installing ovirt - ovirt-engine not found

Hey I need help, with the installation of oVirt (again).
After installing gluster and getting this up and running, I am trying to
install ovirt, by first running this command:

yum install http://resources.ovirt.org/pub/yum-repo/ovirt-release35.rpm

and then

yum install -y ovirt-engine

However this does not work as I get: Error packages does not exist.

Sorry, which distribution and which architecture are you trying to deploy
on?

x86_64 and fedora 21

We are still not supporting fedora 21 and we will probably jump to fedora
22 skipping 21 at all.
Please see:
https://bugzilla.redhat.com/show_bug.cgi?id=1163062

But should I not still be able to at least see the package?

No, cause it is using fc$releasever variable in the yum repo specification and 
it doesn't exist for fc21.


Lars, will a developer setup (namely no RPM) is something which can get 
you started? maybe be worth giving a try if you don't plan to use some 
supported platform

Can some please tell me how to fix it? I have tried to run yum update
-y, which did not help at all

Thanks in advance.

- Lars
___
Users mailing list
Users@ovirt.org
http://lists.ovirt.org/mailman/listinfo/users


--
Med venlig hilsen / Best Regards
Lars Nielsen
Student developer at Steinwurf
l...@steinwurf.com



--
Med venlig hilsen / Best Regards
Lars Nielsen
Student developer at Steinwurf
l...@steinwurf.com



___
Users mailing list
Users@ovirt.org
http://lists.ovirt.org/mailman/listinfo/users


___
Users mailing list
Users@ovirt.org
http://lists.ovirt.org/mailman/listinfo/users

Re: [ovirt-users] selectively disabling IPv6 on bridges

[Dan Kenigsberg]

On Wed, May 06, 2015 at 01:53:35PM +0100, Dan Kenigsberg wrote:
> On Wed, May 06, 2015 at 01:28:30PM +0200, Rik Theys wrote:
> > Hi,
> > 
> > I'm looking for a way to selectively disable IPv6 on the bridge interfaces
> > on the oVirt hosts.
> > 
> > When oVirt creates the bridges for all logical networks on the host, it
> > keeps the default settings for IPv6 which means all bridges get a link-local
> > address and accept router advertisements.
> > 
> > When a VM is created on the logical network, it can now reach the host over
> > IPv6 (but not over IPv4 if no IP address has been assigned on the host). If
> > it sends out a router advertisement it can even create a global IPv6 address
> > (haven't tested this).
> > 
> > How can I prevent this?
> > 
> > I would like to prevent the guest from IPv6 access to the host but the guest
> > itself still needs IPv6 access (global IPv6 addresses).
> > 
> > Is it sufficient to create a sysctl config file that says:
> > 
> > net.ipv6.conf.default.disable_ipv6 = 1
> 
> Yes, I believe that this would do the trick. For any newly-created
> device on the system, regardless of ovirt bridges.
> 
> I now see that el7 has changed the default for IPV6INIT to "yes". We
> should be more prudent and set IPV6INIT=no on all our devices.

Lukáš, it seems that setting IPV6INIT=no is not enough:

IPV6INIT=yes|no
  Enable or disable IPv6 static, DHCP, or autoconf configuration for this 
interface
  Default: yes

The bridge still gets a link-local ipv6 address anyway. Is there an initscript
means to disable this completely, or should we resort to
/proc/sys/net/ipv6/conf//disable_ipv6 ?

Dan.
___
Users mailing list
Users@ovirt.org
http://lists.ovirt.org/mailman/listinfo/users

Re: [ovirt-users] Upgrade Node through oVirt Manager

[Fabian Deutsch]

- Original Message -
> Hi all,
> 
> Hi installed two nodes with the 3.5.1 ISO. Now I want to upgrade them to
> 3.5.2
> new ISO. I uploaded the ISO file to my ISO domain with engine-iso-uploader
> command. But I can't choose my ISO file in the Manager to upgrade my nodes.
> How can I select the good Iso file ?

Hey Kevin,

the update of the Node on the host is independent of the ISO domain.

You need to install the Node iso through the also provided rpm on the Engine 
host.

I.e the job below contains an iso and an rpm. The rpm needs to be installed on 
the Engine host,
then the new ISO should appear in the upgrade dialog.

http://jenkins.ovirt.org/user/fabiand/my-views/view/Node/job/ovirt-node_ovirt-3.5_create-iso-el7_merged/

Greetings
fabian
___
Users mailing list
Users@ovirt.org
http://lists.ovirt.org/mailman/listinfo/users

Re: [ovirt-users] autologin into vm desktop

[Adolfo]

Thank you very much.

The problem is that the broker supports a wide variety of 
authenticators, and the users that logins UDS are not oVirt users... :(. 
(I.e. we support SAML as auth, o eDirectory users, and UDS users are not 
registered inside oVirt, i mean, they are not oVirt users)


The easier way is that API exposes "vdsClient", but it did not.. (maybe 
i need a post on devel list)?


Right now, i'm going to try to connect using or vdsClient or, if i can, 
the XMLRPC interface... I have to do some tests...


Thank you very much for your help ;-)

Regards,

Adolfo Gómez


El 07/05/2015 a las 11:29, Juan Hernández escribió:

On 05/07/2015 05:25 AM, Adolfo wrote:

:`(... after looking at this, this is not what i was looking for (but
almost).

This API call seems to make logon on desktop using "the credentials of
the client. I mean, we can't pass the user credentials (as with
vdsClient), ovirt takes them from their own. (I think that from user
that makes the API call).

I need to pass the "username" "password" "domain" data in a way similar
to this, but be able to do it without registering the user inside ovirt,
making a call with the users credentials, etc... so back to beginning i
guess... :(

Anyway, thank you very much for taking your time for helping me ;-)

Regards,

Adolfo gómez


If your broker has the credentials of the user then you can connect to
the API using those credentials, and invoke the "logon" operation,
something like this:

---8<---
#!/bin/sh -ex

curl \
--verbose \
--cacert /the/path/to/the/ca/cert \
--request POST \
--user "myuser@mydomain:mypassword" \
--header "Content-Type: application/xml" \
--header "Accept: application/xml" \
--data '

' \
"https://engine.example.com/ovirt-engine/api/vms/myvm/login";
--->8---


El 07/05/2015 a las 5:10, Adolfo escribió:

O.o that is what i was looking for!!!... Has to take a BIG look at
it... ;)

Thank you very much again, i was looking for this, but after
"googling" a lot didn't found it!!! ;)

Adolfo Gómez


El 07/05/2015 a las 5:05, Dan Yasny escribió:

...and it looks like in the latest versions the REST API
has //vms/{vmid}/logon/
/
/
/https://access.redhat.com/documentation/en-US/Red_Hat_Enterprise_Virtualization/3.5/html/Technical_Guide/sect-Actions5.html#Enable_user_logon_to_access_a_virtual_machine_from_an_external_console
/

On Wed, May 6, 2015 at 10:55 PM, Dan Yasny mailto:dya...@gmail.com>> wrote:

 You can pass the credentials directly to the guest agent using
 vdsClient on the host, among other things:

 desktopLock
 
 Logoff current user
 desktopLogin

 Login to vmId desktop using the supplied credentials
 desktopLogoff
  
 Lock user session. force should be set to true/false

 Will probably require key based remote ssh execution, and API
 calls to the engine, to determine the host and VM UUID

 On Wed, May 6, 2015 at 10:45 PM, Adolfo mailto:ago...@virtualcable.es>> wrote:

 I know, but this all is used from "ovirt portal", and we are
 only using ovirt api, the portal is provided by own broker :)

 I'm currently looking at the code of vdsClient, to see if i
 can replicate the "desktopLogin" feature.

 This was why i was wondering if this is the place to post
 this, because it's more related to "development", but not to
 de development of ovirt itself (or yes, don't know right
 now... :) )

 Thank you

 Adolfo Gómez



 El 07/05/2015 a las 4:40, Dan Yasny escribió:

 This is exactly what the SSO feature is for.

 http://www.ovirt.org/Features/SSO
 http://www.ovirt.org/OVirt_Guest_Agent/Single_Sign_On/Windows
 
https://access.redhat.com/documentation/en-US/Red_Hat_Enterprise_Virtualization/3.1/html/Power_User_Portal_Guide/Single_Sign_On-Windows.html



 On Wed, May 6, 2015 at 10:24 PM, Adolfo
 mailto:ago...@virtualcable.es>> wrote:

 Hello,

 My name is Adolfo. I'm in charge of the development of
 UDS, an open source connection broker (with commercial
 support if requested)
 (http://www.ovirt.org/Universidad_de_Sevilla_Case_Study
 is done with it for example).

 I don't know if this is the place to post this "request
 for help", if not, please forgive me :)

 The case is that we are including Spice as an accepted
 protocol for connecting to VMs (currently we allow rdp,
 rgs, nx, ...) provided by oVirt, and we have found the
 following "issue".

 It's ease to get the connection parameters for the VM
 using REST api, even get the ticket for allowing
 connection, but i have been looking for a way "logging
 user" directly into desktop, not only connect to
 "display" but also "log in" into rem

[ovirt-users] vdsClient getVGList returned Empty

[like...@cs2c.com.cn]

Hi, guys
I have a vdsm host, when i use "vgs" command, it will display 2 volume group 
which already exist. But when i use "vdsClient -s 0 getVGList", it displays 
nothing. That means vdsm cannot see the existed vg. Why? Does anybody have any 
idea?

P.S. In fact, i am trying to resolve the problem of "Cannot find master domain"



like...@cs2c.com.cn
___
Users mailing list
Users@ovirt.org
http://lists.ovirt.org/mailman/listinfo/users

Re: [ovirt-users] GetGlusterVolumeAdvancedDetailsQuery & GetGlusterVolumeProfileInfoQuery when using seperate storage network

[Sahina Bose]

On 05/07/2015 01:34 PM, Jorick Astrego wrote:



On 05/06/2015 08:15 PM, knarra wrote:

On 05/06/2015 11:22 PM, Jorick Astrego wrote:



On 05/06/2015 06:24 PM, knarra wrote:

On 05/06/2015 06:59 PM, Jorick Astrego wrote:



On 05/06/2015 02:49 PM, knarra wrote:

On 05/06/2015 05:33 PM, Jorick Astrego wrote:

Hi,

Looking forward to bz 1049994Allow choosing network 
interface for gluster domain traffic.


Currently I have the bricks on a different storage network and 
can't get the volume details or profile it.


Will this be handled in 3.6 properly? I don't see any changes in 
gerrit regarding this, but I can be ovelooking it.


The errors I get currently:

Could not fetch brick profile stats

2015-05-06 10:34:22,430 ERROR
[org.ovirt.engine.core.bll.gluster.GetGlusterVolumeProfileInfoQuery]
(ajp--127.0.0.1-8702-27) Query
GetGlusterVolumeProfileInfoQuery failed. Exception message
is null : java.lang.NullPointerException:
java.lang.NullPointerException

and

Error in fetching the brick details, please try again.

2015-05-06 10:36:14,205 ERROR
[org.ovirt.engine.core.bll.gluster.GetGlusterVolumeAdvancedDetailsQuery]
(ajp--127.0.0.1-8702-55) Query
GetGlusterVolumeAdvancedDetailsQuery failed. Exception
message is VdcBLLException: Volume status failed
error: Staging failed on *.*.*.*. Error: No brick
glustertest1.netbulae.test/gluster/brick1 in volume data
Staging failed on *.*.*.*. Error: No brick
glustertest1.netbulae.test:/gluster/brick1 in volume data
return code: -1 (Failed with error GlusterVolumeStatusFailed
and code 4157) :
org.ovirt.engine.core.common.errors.VdcBLLException:
VdcBLLException: Volume status failed
error: Staging failed on *.*.*.*. Error: No brick
glustertest1.netbulae.test:/gluster/brick1 in volume data
Staging failed on *.*.*.*. Error: No brick
glustertest1.netbulae.test:/gluster/brick1 in volume data
return code: -1 (Failed with error GlusterVolumeStatusFailed
and code 4157):
org.ovirt.engine.core.common.errors.VdcBLLException:
VdcBLLException: Volume status failed
error: Staging failed on *.*.*.*. Error: No brick
glustertest1.netbulae.test:/gluster/brick1 in volume data
Staging failed on *.*.*.*. Error: No brick
glustertest1.netbulae.test:/gluster/brick1 in volume data
return code: -1 (Failed with error GlusterVolumeStatusFailed
and code 4157)





Met vriendelijke groet, With kind regards,

Jorick Astrego*

Netbulae Virtualization Experts *

Tel: 053 20 30 270 	i...@netbulae.eu 	Staalsteden 4-3A 	KvK 
08198180
Fax: 053 20 30 271 	www.netbulae.eu 	7547 TA Enschede 	BTW 
NL821234584B01







___
Users mailing list
Users@ovirt.org
http://lists.ovirt.org/mailman/listinfo/users

Hi,

Can you please check what does your gluster peer status on 
each of your nodes return?  I guess they are in disconnected 
state and that is the reason you are not able to view these details.


Thanks
kasturi


On the nodes it gives me the following:

gluster peer status
Connection failed. Please check if gluster daemon is operational.

This means that glusterd is not running on this node. you could 
check the status of glusterd by running the command 'service 
glusterd status'.


please start glusterd by running the command 'service glusterd 
start' on both of your nodes.


Ideally when glusterd goes down node in ovirt should move to 
non-operational. Because of this BZ 1207150 as of now it is not 
changing the state to non operational.


There is no glusterd on the compute nodes in our setup, we have two 
clusters. One for virt hosts only and one for GlusterFS only.



Like I said, everything is Up and running fine. It's just that I 
can't get the stats because the hostname != GlusterFS NIC ip






But everyting is up and ovirt found the manually configured volume 
perfectly. But the hostname it lists as glustertest1.netbulae.test 
is not what my volume uses for communication as I created the 
volume using the ip's of the storage network.



gluster peer status
Number of Peers: 2

Hostname: 10.1.1.3
Uuid: 1cc0875e-1699-42ae-aed2-9152667ed5af
State: Peer in Cluster (Connected)

Hostname: 10.1.1.2
Uuid: a0b3ac13-7388-441a-a238-1deb023cab6c
State: Peer in Cluster (Connected)


Did you import already existing cluster ?


No, I provisioned the nodes. Added them to our glusterfs cluster 
(with virt service disabled) and created the volume manually.


oVirt auto-discovered the manual created volume after that.

Error: No brick *glustertest1.netbulae.test*:/gluster/brick1 in
volume data

Hostname: *10.1.1.3*

Things should work better in 3.6 (bz1049994), but I don't see any 
code changes to "GetGlusterVolumeProfileInfoQuery" linked 

[ovirt-users] Presentation

[NUNIN Roberto]

Hello

I'm Roberto Nunin, responsible for Infrastructure in an Italian Company, part 
of multi-national group.
I'm really interested in oVirt technology and it's future developments.

Currently we are running a PoC of three clusters, 6 hosts, with an issue I've 
already submitted to the community.

Hope to find inside community answers, solutions, suggestions related to the 
product.
Will try to enhance my knowledge about it.

Have a nice day

Roberto Nunin
RHCE#110-006-970



Questo messaggio e' indirizzato esclusivamente al destinatario indicato e 
potrebbe contenere informazioni confidenziali, riservate o proprietarie. 
Qualora la presente venisse ricevuta per errore, si prega di segnalarlo 
immediatamente al mittente, cancellando l'originale e ogni sua copia e 
distruggendo eventuali copie cartacee. Ogni altro uso e' strettamente proibito 
e potrebbe essere fonte di violazione di legge.

This message is for the designated recipient only and may contain privileged, 
proprietary, or otherwise private information. If you have received it in 
error, please notify the sender immediately, deleting the original and all 
copies and destroying any hard copies. Any other use is strictly prohibited and 
may be unlawful.
___
Users mailing list
Users@ovirt.org
http://lists.ovirt.org/mailman/listinfo/users

Re: [ovirt-users] autologin into vm desktop

[Juan Hernández]

On 05/07/2015 05:25 AM, Adolfo wrote:
> :`(... after looking at this, this is not what i was looking for (but
> almost).
> 
> This API call seems to make logon on desktop using "the credentials of
> the client. I mean, we can't pass the user credentials (as with
> vdsClient), ovirt takes them from their own. (I think that from user
> that makes the API call).
> 
> I need to pass the "username" "password" "domain" data in a way similar
> to this, but be able to do it without registering the user inside ovirt,
> making a call with the users credentials, etc... so back to beginning i
> guess... :(
> 
> Anyway, thank you very much for taking your time for helping me ;-)
> 
> Regards,
> 
> Adolfo gómez
> 

If your broker has the credentials of the user then you can connect to
the API using those credentials, and invoke the "logon" operation,
something like this:

---8<---
#!/bin/sh -ex

curl \
--verbose \
--cacert /the/path/to/the/ca/cert \
--request POST \
--user "myuser@mydomain:mypassword" \
--header "Content-Type: application/xml" \
--header "Accept: application/xml" \
--data '

' \
"https://engine.example.com/ovirt-engine/api/vms/myvm/login";
--->8---

> 
> El 07/05/2015 a las 5:10, Adolfo escribió:
>> O.o that is what i was looking for!!!... Has to take a BIG look at
>> it... ;)
>>
>> Thank you very much again, i was looking for this, but after
>> "googling" a lot didn't found it!!! ;)
>>
>> Adolfo Gómez
>>
>>
>> El 07/05/2015 a las 5:05, Dan Yasny escribió:
>>> ...and it looks like in the latest versions the REST API
>>> has //vms/{vmid}/logon/
>>> /
>>> /
>>> /https://access.redhat.com/documentation/en-US/Red_Hat_Enterprise_Virtualization/3.5/html/Technical_Guide/sect-Actions5.html#Enable_user_logon_to_access_a_virtual_machine_from_an_external_console
>>> /
>>>
>>> On Wed, May 6, 2015 at 10:55 PM, Dan Yasny >> > wrote:
>>>
>>> You can pass the credentials directly to the guest agent using
>>> vdsClient on the host, among other things:
>>>
>>> desktopLock
>>> 
>>> Logoff current user
>>> desktopLogin
>>>
>>> Login to vmId desktop using the supplied credentials
>>> desktopLogoff
>>>  
>>> Lock user session. force should be set to true/false
>>>
>>> Will probably require key based remote ssh execution, and API
>>> calls to the engine, to determine the host and VM UUID
>>>
>>> On Wed, May 6, 2015 at 10:45 PM, Adolfo >> > wrote:
>>>
>>> I know, but this all is used from "ovirt portal", and we are
>>> only using ovirt api, the portal is provided by own broker :)
>>>
>>> I'm currently looking at the code of vdsClient, to see if i
>>> can replicate the "desktopLogin" feature.
>>>
>>> This was why i was wondering if this is the place to post
>>> this, because it's more related to "development", but not to
>>> de development of ovirt itself (or yes, don't know right
>>> now... :) )
>>>
>>> Thank you
>>>
>>> Adolfo Gómez
>>>
>>>
>>>
>>> El 07/05/2015 a las 4:40, Dan Yasny escribió:
 This is exactly what the SSO feature is for. 

 http://www.ovirt.org/Features/SSO
 http://www.ovirt.org/OVirt_Guest_Agent/Single_Sign_On/Windows
 
 https://access.redhat.com/documentation/en-US/Red_Hat_Enterprise_Virtualization/3.1/html/Power_User_Portal_Guide/Single_Sign_On-Windows.html



 On Wed, May 6, 2015 at 10:24 PM, Adolfo
 mailto:ago...@virtualcable.es>> wrote:

 Hello,

 My name is Adolfo. I'm in charge of the development of
 UDS, an open source connection broker (with commercial
 support if requested)
 (http://www.ovirt.org/Universidad_de_Sevilla_Case_Study
 is done with it for example).

 I don't know if this is the place to post this "request
 for help", if not, please forgive me :)

 The case is that we are including Spice as an accepted
 protocol for connecting to VMs (currently we allow rdp,
 rgs, nx, ...) provided by oVirt, and we have found the
 following "issue".

 It's ease to get the connection parameters for the VM
 using REST api, even get the ticket for allowing
 connection, but i have been looking for a way "logging
 user" directly into desktop, not only connect to
 "display" but also "log in" into remote without needed
 to use a second authentication.

 I have seen that oVirt Portal currently allows this, and
 i have found also that vsdClient can do login using
 "vdsClient -s  desktopLogin  
  ", and

Re: [ovirt-users] GetGlusterVolumeAdvancedDetailsQuery & GetGlusterVolumeProfileInfoQuery when using seperate storage network

[Jorick Astrego]

On 05/06/2015 08:15 PM, knarra wrote:
> On 05/06/2015 11:22 PM, Jorick Astrego wrote:
>>
>>
>> On 05/06/2015 06:24 PM, knarra wrote:
>>> On 05/06/2015 06:59 PM, Jorick Astrego wrote:


 On 05/06/2015 02:49 PM, knarra wrote:
> On 05/06/2015 05:33 PM, Jorick Astrego wrote:
>> Hi,
>>
>> Looking forward to bz 1049994Allow choosing network interface
>> for gluster domain traffic.
>>
>> Currently I have the bricks on a different storage network and
>> can't get the volume details or profile it.
>>
>> Will this be handled in 3.6 properly? I don't see any changes in
>> gerrit regarding this, but I can be ovelooking it.
>>
>> The errors I get currently:
>>
>> Could not fetch brick profile stats
>>
>> 2015-05-06 10:34:22,430 ERROR
>> [org.ovirt.engine.core.bll.gluster.GetGlusterVolumeProfileInfoQuery]
>> (ajp--127.0.0.1-8702-27) Query
>> GetGlusterVolumeProfileInfoQuery failed. Exception message is
>> null : java.lang.NullPointerException:
>> java.lang.NullPointerException
>>
>> and
>>
>> Error in fetching the brick details, please try again.
>>
>> 2015-05-06 10:36:14,205 ERROR
>> 
>> [org.ovirt.engine.core.bll.gluster.GetGlusterVolumeAdvancedDetailsQuery]
>> (ajp--127.0.0.1-8702-55) Query
>> GetGlusterVolumeAdvancedDetailsQuery failed. Exception
>> message is VdcBLLException: Volume status failed
>> error: Staging failed on *.*.*.*. Error: No brick
>> glustertest1.netbulae.test/gluster/brick1 in volume data
>> Staging failed on *.*.*.*. Error: No brick 
>> glustertest1.netbulae.test:/gluster/brick1 in volume data
>> return code: -1 (Failed with error GlusterVolumeStatusFailed
>> and code 4157) :
>> org.ovirt.engine.core.common.errors.VdcBLLException:
>> VdcBLLException: Volume status failed
>> error: Staging failed on *.*.*.*. Error: No brick 
>> glustertest1.netbulae.test:/gluster/brick1 in volume data
>> Staging failed on *.*.*.*. Error: No brick 
>> glustertest1.netbulae.test:/gluster/brick1 in volume data
>> return code: -1 (Failed with error GlusterVolumeStatusFailed
>> and code 4157):
>> org.ovirt.engine.core.common.errors.VdcBLLException:
>> VdcBLLException: Volume status failed
>> error: Staging failed on *.*.*.*. Error: No brick 
>> glustertest1.netbulae.test:/gluster/brick1 in volume data
>> Staging failed on *.*.*.*. Error: No brick 
>> glustertest1.netbulae.test:/gluster/brick1 in volume data
>> return code: -1 (Failed with error GlusterVolumeStatusFailed
>> and code 4157)
>>
>>
>>
>>
>>
>> Met vriendelijke groet, With kind regards,
>>
>> Jorick Astrego*
>>
>> Netbulae Virtualization Experts *
>> 
>> Tel: 053 20 30 270   i...@netbulae.euStaalsteden 4-3A
>> KvK 08198180
>> Fax: 053 20 30 271   www.netbulae.eu 7547 TA Enschede
>> BTW
>> NL821234584B01
>>
>>
>> 
>>
>>
>>
>> ___
>> Users mailing list
>> Users@ovirt.org
>> http://lists.ovirt.org/mailman/listinfo/users
> Hi,
>
> Can you please check what does your gluster peer status on
> each of your nodes return?  I guess they are in disconnected state
> and that is the reason you are not able to view these details.
>
> Thanks
> kasturi
>
 On the nodes it gives me the following:

 gluster peer status
 Connection failed. Please check if gluster daemon is operational.

>>> This means that glusterd is not running on this node. you could
>>> check the status of glusterd by running the command 'service
>>> glusterd status'.
>>>
>>> please start glusterd by running the command 'service glusterd
>>> start' on both of your nodes.
>>>
>>> Ideally when glusterd goes down node in ovirt should move to
>>> non-operational. Because of this BZ 1207150 as of now it is not
>>> changing the state to non operational.
>>
>> There is no glusterd on the compute nodes in our setup, we have two
>> clusters. One for virt hosts only and one for GlusterFS only.
>>
>>
>> Like I said, everything is Up and running fine. It's just that I
>> can't get the stats because the hostname != GlusterFS NIC ip
>>
>>


 But everyting is up and ovirt found the manually configured volume
 perfectly. But the hostname it lists as glustertest1.netbulae.test
 is not what my volume uses for communication as I created the
 volume using the ip's of the storage network.

  
 gluster peer status
   

Re: [ovirt-users] selectively disabling IPv6 on bridges

[Rik Theys]

Hi,

On 05/06/2015 02:53 PM, Dan Kenigsberg wrote:

On Wed, May 06, 2015 at 01:28:30PM +0200, Rik Theys wrote:

I'm looking for a way to selectively disable IPv6 on the bridge interfaces
on the oVirt hosts.

When oVirt creates the bridges for all logical networks on the host, it
keeps the default settings for IPv6 which means all bridges get a link-local
address and accept router advertisements.

When a VM is created on the logical network, it can now reach the host over
IPv6 (but not over IPv4 if no IP address has been assigned on the host). If
it sends out a router advertisement it can even create a global IPv6 address
(haven't tested this).

How can I prevent this?

I would like to prevent the guest from IPv6 access to the host but the guest
itself still needs IPv6 access (global IPv6 addresses).

Is it sufficient to create a sysctl config file that says:

net.ipv6.conf.default.disable_ipv6 = 1


Yes, I believe that this would do the trick. For any newly-created
device on the system, regardless of ovirt bridges.

I now see that el7 has changed the default for IPV6INIT to "yes". We
should be more prudent and set IPV6INIT=no on all our devices.

Would you open a bug about this, so it is tracked?


I've opened bug 1219363 for this.

Regards,

Rik


--
Rik Theys
System Engineer
KU Leuven - Dept. Elektrotechniek (ESAT)
Kasteelpark Arenberg 10 bus 2440  - B-3001 Leuven-Heverlee
+32(0)16/32.11.07

<>
___
Users mailing list
Users@ovirt.org
http://lists.ovirt.org/mailman/listinfo/users

Re: [ovirt-users] hosted engine vm network access

[Yedidyah Bar David]

(Ccing users list)

- Original Message -
> From: "mike" 
> To: "Yedidyah Bar David" 
> Sent: Thursday, May 7, 2015 4:16:54 AM
> Subject: Re: hosted engine vm network access
> 
> Hello
> 
> 
> Thank you for your answers.
> 
> i tried everything from the beginning with the same probleme, i can ping
> my Centos7 Server but i cannot not have internet access neither to my GW
> on the created vm by hosted-engine --deploy. i can ONLY ping my centos 7
> server
> 
> 
> I have one Centos7 server installed and updated
> 
> 1 NIC eth0 at 192.168.42.114/24

But you wrote below that you have also eth1 ?

> 
> i installed the ovirt repo
> 
> i run hosted-engine, follow the questions, i set bridge at eth0 and set
> the FQDN of the vm at hevm.lab.lan (which is present in the /etc/hosts
> file of my Centos7 server with ip 192.168.42.116)
> 
> i can see on my Centos7 Server that i have a new interface named
> ovirtmgmgt with the the ip previously set for eth0 and now eth0 has no
> ip address but i can still ssh to my centos7 serverthrough ovirtmgmt
> interface i presume...

Indeed - this is a normal linux bridge

> 
> 
> here is my bridge conf of my centos7 server
> 
> #brctl show
> bridge name bridge id   STP enabled interfaces
> ;vdsmdummy; 8000.   no
> ovirtmgmt   8000.000c29de8495   no  eth0
> 
> 
> 
> Once my VM is created (hevm.lab.lan), i connect to it again through vnc
> and  i set an ip address manualy like:
> # ifconfig eth0 192.168.42.116/24
> # route add default gw 192.168.42.254
> 
> 
> to resume, i can only ping my centos7 server and no other host on the
> same subnet and no internet access.

I assume you can only ping the address of the ovirtmgmt bridge (192.168.42.114).
Can you ping the address of eth1?

> 
> 
> i hope i explained my problemin a maner that you could help me.

Still not enough details.

Where are eth0 and eth1 connected?

Please provide the details I asked for below. Thanks!

Best regards,

> 
> 
> sorry for my english
> 
> 
> thank you for the help.
> 
> 
> 
> 
> 
> 
> 
> 
> 
> 
> 
> 
> 
> 
> 
> 
> On 05/06/2015 07:37 AM, Yedidyah Bar David wrote:
> > - Original Message -
> >> From: "mike" 
> >> To: sbona...@redhat.com
> >> Cc: d...@redhat.com
> >> Sent: Tuesday, May 5, 2015 7:29:48 PM
> >> Subject: hosted engine vm network access
> >>
> >> Hie.
> >>
> >> i've been asked to email you instead of asking on the #ovirt channel.
> > That's fine, but usually in such cases people are supposed/advised to
> > email users@ovirt.org, Ccing specific people if needed/relevant :-)
> >
> >>
> >> i have followed this instructions:
> >> http://community.redhat.com/blog/2014/10/up-and-running-with-ovirt-3-5/
> >>
> >> the only differences are the IP addresse/mapping names of the begining
> >>
> >> _once i have installed the hosted engine through hosted-engine --deploy,
> >> i connect to it using vnc but i cannot have internet access to install
> >> the engine from ovirt repo neither ping my Main host_
> >>
> >> i have:
> >>
> >> 192.168.42.116virt1virt1.lab.lannfsmountnfsmount.lab.lan
> >> 192.168.42.117hevmhevm.lab.lan# this is the ip i would like
> >> to set for the hosted engine VM
> > You mean, you have this in /etc/hosts of all relevant machines?
> >
> >>
> >> on my Main Host cento7, i have
> >>
> >> eth1-> 192.168.42.116 (managment NIC for ssh access and nfs server)
> >> eth0-> with nothing
> >> ovirtmgmt-> 192.168.42.114
> > Which includes eth0? What's the output of 'brctl show'?
> >
> > Are eth0 and eth1 on the same switch/segment?
> >
> > What's the subnet of each?
> >
> >> bond0-> with nothing
> > That's normal
> >
> >>
> >> can you please help me?
> > What do you have on the engine VM?
> >
> > Can you ssh to the engine from the host? Or the other direction?
> >
> > Please check both addresses of the host.
> >
> > I am not sure, but for the engine VM to access 192.168.42.116, you might
> > need to add some routing somewhere.
> >
> > I'd say this isn't related to hosted-engine, but some general networking
> > issue. If you fail to diagnose this at this point, you can try to simply
> > boot some live cd in a VM (e.g. using virt-install or even qemu directly)
> > and debug from there. Once you have it working, start hosted-engine again.
> >
> > Please check (or post, if you want us to help) all relevant information,
> > including the above, and also things such as:
> >
> > Routing table of host/vm, and of all relevant routers
> >
> > Do you use dhcp? Does it work? On host/engine?
> >
> > Did you enable ip forwarding on the host (cat
> > /proc/sys/net/ipv4/ip_forward)?
> >
> > After you try ssh/ping, arp table on host/engine (arp -an)
> >
> > Check iptables config/logs
> >
> > Etc.
> >
> > Good luck.
> >
> > If you fail to solve this yourself, please Cc users@ with your reply.
> >
> > Thanks!
> >
> > Best,
> 
> 

-- 
Didi

___
Users mailing li