[Users] Multiple Engines, one web Interface?
by Darrell Budic
Now that we’ve got a self hosted engine, has any one given any thought to allowing one engine to remotely control another?
The scenario I’m imagining has one cluster at a DC with 4 nodes, and another cluster at a different DC with 3 nodes. Connectivity is normally pretty good, but it’s been known to drop for a bit every year. And I wouldn’t want to have the engine polling across it all the time as it’s a lower bandwidth link (100M vs 10G between nodes, so a bit). But what I would like is to have the one I’m sitting next to be able to control the remote one, and monitor it’s status, with each having it’s own self hosted engine so they are independent and migrations, etc, will continue to happen even if the two clusters get disconnected. This keeps all the polling local and fast, and avoids engine or cluster freak outs if/when the link is down between the two physically diverse clusters.
Cool bonus feature would be to enable a direct transfer between the two clusters storage facilities. Maybe with a “transport VM” button that suspends it to disk and moves it before restarting it (assuming you built your networks right, of course, complex but presumably possible), but I’d with a simple way to copy a shutdown VM over without going through an export/import process.
I could see this working with a separate manager engine, or just as an add in to the standard engine. Maybe a “cluster w/ local engine” type thing showing in the GUI.
Obviously two web interfaces on two separate engines aren’t a terribly hard thing to manage, but I just imagined the simple way and figured I’d throw it out there and see how hard or easy people thought it might be.
-Darrell
10 years, 11 months
[Users] Physical CDROM access
by Bob Doolittle
Hi,
With VMware ESX, when you edit the CD device you have of course the
option of attaching an ISO, but IIRC you also have the option of
selecting the physical CDROM device for the Host.
There seems no way to do this with oVirt. It would be a really nice
addition to the "Change CD" menu to have something like "/dev/cdrom". Is
there a technical issue preventing this? Of course that would prevent VM
migration such as HA while the CDROM device is attached, but that didn't
stop VMware and it would be a valuable addition.
Thanks,
Bob
10 years, 11 months
[Users] R: Re: How to - gluster
by tristano82＠libero.it
yes my idea is to start with 1 node ( storage+compute ) and then expand with
more server to add storage and compute.
what do you think ?
>----Messaggio originale----
>Da: vbellur(a)redhat.com
>Data: 16/12/2013 8.38
>A: "tristano82(a)libero.it"<tristano82(a)libero.it>, <users(a)ovirt.org>
>Ogg: Re: [Users] How to - gluster
>
>On 12/15/2013 10:30 PM, tristano82(a)libero.it wrote:
>> i have 1 phisical node with SSD disks, can i install ovirt with glusterFS
>> storage as backend . In a second time add new ovirt node ( same hardware )
>> and
>> attach to the first 1 to create a compute and storage HA ?
>>
>>
>
>gluster volumes can be expanded from a single brick to multiple bricks
>for providing HA.
>
>Do you plan to use a converged compute and storage cluster?
>
>Regards,
>Vijay
>
>
10 years, 11 months
[Users] feature request - support getting console.vv with a single click
by i iordanov
Hi guys,
Mobile browsers usually reserve the double-tap functionality for zooming
in/out. It's therefore difficult/impossible to obtain a console.vv file
from the user-mode web interface on a mobile device.
Would you consider adding functionality to connect to the console without a
double-click?
Thanks!
iordan
--
The conscious mind has only one thread of execution.
10 years, 11 months
[Users] Spice Proxy seems to try connect to host and not defined proxy
by Gianluca Cecchi
Hello,
wrking on 3.3.2 beta on F19 ovirt repo
For simplicity I set
- no Websocket Procy on engine
# engine-config -g WebSocketProxy
WebSocketProxy: version: general
- Stop my WebSocketProxy previosuly configured
# systemctl status ovirt-websocket-proxy.service
ovirt-websocket-proxy.service - oVirt Engine websockets proxy
Loaded: loaded
(/usr/lib/systemd/system/ovirt-websocket-proxy.service; enabled)
Active: inactive (dead) since Sun 2013-12-15 17:58:45 CET; 18min ago
Main PID: 23556 (code=exited, status=0/SUCCESS)
CGroup: name=systemd:/system/ovirt-websocket-proxy.service
- set SpiceProxyDefault
# engine-config -g SpiceProxyDefault
SpiceProxyDefault: http://10.4.4.63:3128 version: general
10.4.4.63 is a CentOS 5.10 server with squid enabled
[root@c510 squid]# diff squid.conf squid.conf.orig
578,581d577
<
< acl localnet src 10.4.3.0/24 # RFC1918 possible internal network
< acl localnet src 10.4.4.0/24 # RFC1918 possible internal network
<
624c620
< #http_access deny CONNECT !SSL_ports
---
> http_access deny CONNECT !SSL_ports
638d633
< http_access allow localnet
restart ovirt-engine service
My engine is 10.4.4.60
My host is fedora 19 with ip 10.4.4.59 and a running VM
My client is a Fedora 19 with firefox 26 and ip 10.4.3.24
If I open user portal in my client I see that "enable SpiceProxy" is
checked against this VM
(as I set it gobally via engine-config command)
But when I click console button I see this on my client:
$ netstat -an | egrep "10.4.4"
tcp 0 0 10.4.3.24:38061 10.4.4.60:443 TIME_WAIT
tcp 0 0 10.4.3.24:46284 10.4.4.60:22 ESTABLISHED
tcp 0 1 10.4.3.24:43066 10.4.4.59:5900 SYN_SENT
and an error in browser about
unable to connect to graphic server
so it attempts to go to host.
In client /var/log/messages:
Dec 15 18:21:21 tekkaman gnome-session[2864]: (plugin-container:8024):
SpiceXPI-CRITICAL **: controller connect: No such file or directory
Dec 15 18:21:21 tekkaman gnome-session[2864]: (plugin-container:8024):
SpiceXPI-CRITICAL **: controller connect: No such file or directory
Dec 15 18:21:21 tekkaman spice: starting remote-viewer --spice-controller ...
Dec 15 18:21:21 tekkaman spice: XDG_VTNR=1
Dec 15 18:21:21 tekkaman spice: ORBIT_SOCKETDIR=/var/tmp/orbit-g.cecchi
Dec 15 18:21:21 tekkaman spice: XSUNTRANSPORT=shmem
Dec 15 18:21:21 tekkaman spice: XDG_SESSION_ID=1
Dec 15 18:21:21 tekkaman spice: HOSTNAME=tekkaman.localdomain.local
Dec 15 18:21:21 tekkaman spice: GIO_LAUNCHED_DESKTOP_FILE_PID=6719
Dec 15 18:21:21 tekkaman spice: IMSETTINGS_INTEGRATE_DESKTOP=yes
Dec 15 18:21:21 tekkaman spice:
GPG_AGENT_INFO=/run/user/1000/keyring-uNXVI0/gpg:0:1
Dec 15 18:21:21 tekkaman spice: XDG_MENU_PREFIX=gnome-
Dec 15 18:21:21 tekkaman spice: TERM=dumb
Dec 15 18:21:21 tekkaman spice: SHELL=/bin/bash
Dec 15 18:21:21 tekkaman spice: HISTSIZE=1000
Dec 15 18:21:21 tekkaman spice: TMPDIR=/var/tmp
Dec 15 18:21:21 tekkaman spice: XRE_PROFILE_NAME=
Dec 15 18:21:21 tekkaman spice: GJS_DEBUG_OUTPUT=stderr
Dec 15 18:21:21 tekkaman spice:
GNOME_KEYRING_CONTROL=/run/user/1000/keyring-uNXVI0
Dec 15 18:21:21 tekkaman spice: QTDIR=/usr/lib64/qt-3.3
Dec 15 18:21:21 tekkaman spice: GJS_DEBUG_TOPICS=JS ERROR;JS LOG
Dec 15 18:21:21 tekkaman spice: QTINC=/usr/lib64/qt-3.3/include
Dec 15 18:21:21 tekkaman spice: MOZILLA_FIVE_HOME=/usr/lib64/firefox
Dec 15 18:21:21 tekkaman spice: NSS_SSL_CBC_RANDOM_IV=1
Dec 15 18:21:21 tekkaman spice: IMSETTINGS_MODULE=none
Dec 15 18:21:21 tekkaman spice: QT_GRAPHICSSYSTEM_CHECKED=1
Dec 15 18:21:21 tekkaman spice: XRE_PROFILE_LOCAL_PATH=
Dec 15 18:21:21 tekkaman spice: USER=g.cecchi
Dec 15 18:21:21 tekkaman spice: XRE_START_OFFLINE=
Dec 15 18:21:21 tekkaman spice:
LD_LIBRARY_PATH=/usr/lib64/firefox/xulrunner:/usr/lib64/firefox:/usr/lib64/firefox/plugins:/usr/lib64/firefox
Dec 15 18:21:21 tekkaman spice: SSH_AUTH_SOCK=/run/user/1000/keyring-uNXVI0/ssh
Dec 15 18:21:21 tekkaman spice: GNOME_DISABLE_CRASH_DIALOG=1
Dec 15 18:21:21 tekkaman spice:
SESSION_MANAGER=local/unix:@/tmp/.ICE-unix/2864,unix/unix:/tmp/.ICE-unix/2864
Dec 15 18:21:21 tekkaman spice: USERNAME=g.cecchi
Dec 15 18:21:21 tekkaman spice: LIBPATH=/usr/lib64/firefox:/usr/lib64/firefox
Dec 15 18:21:21 tekkaman spice: MOZ_APP_LAUNCHER=/usr/bin/firefox
Dec 15 18:21:21 tekkaman spice:
MOZ_PLUGIN_PATH=/usr/lib64/mozilla/plugins:/usr/lib64/firefox/plugins
Dec 15 18:21:21 tekkaman spice:
GIO_LAUNCHED_DESKTOP_FILE=/usr/share/applications/firefox.desktop
Dec 15 18:21:21 tekkaman spice:
FONTCONFIG_PATH=/etc/fonts:/usr/lib64/firefox/res/Xft
Dec 15 18:21:21 tekkaman spice: DESKTOP_SESSION=gnome
Dec 15 18:21:21 tekkaman spice:
PATH=/usr/lib64/qt-3.3/bin:/usr/local/bin:/usr/bin:/bin:/usr/local/sbin:/usr/sbin:/home/g.cecchi/.local/bin:/home/g.cecchi/bin
Dec 15 18:21:21 tekkaman spice: MAIL=/var/spool/mail/g.cecchi
Dec 15 18:21:21 tekkaman spice: QT_IM_MODULE=ibus
Dec 15 18:21:21 tekkaman spice: PWD=/home/g.cecchi
Dec 15 18:21:21 tekkaman spice: XMODIFIERS=@im=ibus
Dec 15 18:21:21 tekkaman spice: LANG=en_US.UTF-8
Dec 15 18:21:21 tekkaman spice: GNOME_KEYRING_PID=2860
Dec 15 18:21:21 tekkaman spice: KDE_IS_PRELINKED=1
Dec 15 18:21:21 tekkaman spice: XRE_PROFILE_PATH=
Dec 15 18:21:21 tekkaman spice: GDM_LANG=en_US.UTF-8
Dec 15 18:21:21 tekkaman spice: KDEDIRS=/usr
Dec 15 18:21:21 tekkaman spice: XSUNSMESIZE=512
Dec 15 18:21:21 tekkaman spice: GDMSESSION=gnome
Dec 15 18:21:21 tekkaman spice: HISTCONTROL=ignoredups
Dec 15 18:21:21 tekkaman spice: SHLVL=2
Dec 15 18:21:21 tekkaman spice: HOME=/home/g.cecchi
Dec 15 18:21:21 tekkaman spice: XDG_SEAT=seat0
Dec 15 18:21:21 tekkaman spice: GNOME_DESKTOP_SESSION_ID=this-is-deprecated
Dec 15 18:21:21 tekkaman spice:
DYLD_LIBRARY_PATH=/usr/lib64/firefox:/usr/lib64/firefox
Dec 15 18:21:21 tekkaman spice: MOZ_LAUNCHED_CHILD=
Dec 15 18:21:21 tekkaman spice: MOZ_GRE_CONF=/etc/gre.d/gre64.conf
Dec 15 18:21:21 tekkaman spice: QTLIB=/usr/lib64/qt-3.3/lib
Dec 15 18:21:21 tekkaman spice:
DBUS_SESSION_BUS_ADDRESS=unix:abstract=/tmp/dbus-G3RfQQysBc,guid=49804c78c033de622c5380b152addb59
Dec 15 18:21:21 tekkaman spice: SPICE_XPI_SOCKET=/tmp/spicec-tYz3e4/spice-xpi
Dec 15 18:21:21 tekkaman spice: NO_EM_RESTART=
Dec 15 18:21:21 tekkaman spice: LESSOPEN=||/usr/bin/lesspipe.sh %s
Dec 15 18:21:21 tekkaman spice: MOZ_CRASHREPORTER_RESTART_ARG_1=
Dec 15 18:21:21 tekkaman spice: SHLIB_PATH=/usr/lib64/firefox:/usr/lib64/firefox
Dec 15 18:21:21 tekkaman spice: XRE_BINARY_PATH=
Dec 15 18:21:21 tekkaman spice: MOZ_CRASHREPORTER_RESTART_ARG_0=/usr/bin/firefox
Dec 15 18:21:21 tekkaman spice: WINDOWPATH=1
Dec 15 18:21:21 tekkaman spice: DISPLAY=:0
Dec 15 18:21:21 tekkaman spice: XDG_RUNTIME_DIR=/run/user/1000
Dec 15 18:21:21 tekkaman spice:
QT_PLUGIN_PATH=/usr/lib64/kde4/plugins:/usr/lib/kde4/plugins
Dec 15 18:21:21 tekkaman spice: XUL_APP_FILE=
Dec 15 18:21:21 tekkaman spice:
MOZ_CRASHREPORTER_DATA_DIRECTORY=/home/g.cecchi/.mozilla/firefox/Crash
Reports
Dec 15 18:21:21 tekkaman spice:
MOZ_CRASHREPORTER_STRINGS_OVERRIDE=/usr/lib64/firefox/browser/crashreporter-override.ini
Dec 15 18:21:21 tekkaman spice: NO_AT_BRIDGE=1
Dec 15 18:21:21 tekkaman spice:
XAUTHORITY=/run/gdm/auth-for-g.cecchi-0Ht3a6/database
Dec 15 18:21:21 tekkaman spice: _=/usr/bin/env
Dec 15 18:21:34 tekkaman spice:
Dec 15 18:21:34 tekkaman spice: (remote-viewer:8709): GSpice-WARNING
**: Could not connect to 10.4.4.59: Socket I/O timed out
Dec 15 18:22:19 tekkaman spice: remote-viewer execution failed
Dec 15 18:22:19 tekkaman gnome-session[2864]: (plugin-container:8024):
SpiceXPI-CRITICAL **: could not get browser window, when trying to
call OnDisconnected
No logs in access.log of squid on 10.4.4.63 host
How to debug further?
What files to check on engine for Spice Proxy?
Tried also with a client that is on the same network as all servers
(10.4.4.0 where engine, proxy, hosts has their ips) and the spice
client opens but I see that it indeed connects to 10.4.4.9, so it is
not using proxy...
tcp 0 0 10.4.4.67:39613 10.4.4.59:5901 ESTABLISHED
Thanks,
Gianluca
10 years, 11 months
[Users] Polling vs Pushing engine events
by Sven Kieske
Hi,
we got the following problem:
we create / start / stop
hole vms /data centers / storage etc
(basically: everything ovirt can handle
via REST-API)
But if you want to know e.g. the status
of a vm (or anything) you need to constantly
poll the API.
This is not what we desire to do, as it
does not scale very well (e.g. polling
100 vms).
Is there a standardized way of pushing information
from the engine?
--
Mit freundlichen Grüßen / Regards
Sven Kieske
Systemadministrator
Mittwald CM Service GmbH & Co. KG
Königsberger Straße 6
32339 Espelkamp
T: +49-5772-293-100
F: +49-5772-293-333
https://www.mittwald.de
Geschäftsführer: Robert Meyer
St.Nr.: 331/5721/1033, USt-IdNr.: DE814773217, HRA 6640, AG Bad Oeynhausen
Komplementärin: Robert Meyer Verwaltungs GmbH, HRB 13260, AG Bad Oeynhausen
10 years, 11 months
[Users] Hosts going offline due to storage issues.
by Ian Mordey
Hi there
I've got a 3 node oVirt 3.3.1 setup connecting to an Equallogic iSCSI SAN
array. My nodes keep going offline with these errors:
Host reports about one of the Active Storage Domains as Problematic.
Host cannot access one of the Storage Domains attached to the Data Center
<DC> Setting Host state to Non-Operational
The host is then fenced, reboots and comes back up normally.
I see some of these errors prior to the event:
2013-12-10 13:36:49,442 ERROR
[org.ovirt.engine.core.vdsbroker.vdsbroker.ListVDSCommand]
(DefaultQuartzScheduler_Worker-14) Command ListVDS execution failed.
Exception: VDSNetworkException: java.net.ConnectException: Connection
refused
2013-12-10 13:36:49,448 WARN [org.ovirt.engine.core.vdsbroker.VdsManager]
(DefaultQuartzScheduler_Worker-14) Failed to refresh VDS , vds =
7f74b28a-149a-4f87-9689-b23db6d435f7 : <hostname> , VDS Network Error,
continuing.
2013-12-10 13:36:52,490 ERROR
[org.ovirt.engine.core.vdsbroker.vdsbroker.GetCapabilitiesVDSCommand]
(DefaultQuartzScheduler_Worker-59) Command GetCapabilitiesVDS execution
failed. Exception: VDSNetworkException: java.net.ConnectException:
Connection refused
Any ideas where to look? I can't see any errors on the switches or SAN..
Thanks
Ian
--
Listed on the London Stock Exchange: WAND<http://www.bloomberg.com/quote/WAND:LN>
THIS MESSAGE AND ANY ATTACHMENTS ARE CONFIDENTIAL, PROPRIETARY, AND MAY BE
PRIVILEGED. If this message was misdirected, WANdisco, Inc. and its
subsidiaries, ("WANdisco") does not waive any confidentiality or privilege.
If you are not the intended recipient, please notify us immediately and
destroy the message without disclosing its contents to anyone. Any
distribution, use or copying of this e-mail or the information it contains
by other than an intended recipient is unauthorized. The views and
opinions expressed in this e-mail message are the author's own and may not
reflect the views and opinions of WANdisco, unless the author is authorized
by WANdisco to express such views or opinions on its behalf. All email
sent to or from this address is subject to electronic storage and review by
WANdisco. Although WANdisco operates anti-virus programs, it does not
accept responsibility for any damage whatsoever caused by viruses being
passed.
10 years, 11 months
[Users] AcquireHostId problem
by Pascal Jakobi
Context :
vdsm-4.13.0-11.fc19.x86_64
sanlock-2.8-1.fc19.x86_64
Creating storage domain (NFS) fails with error in engine.log such as
Error code AcquireHostIdFailure and error message VDSGenericException:
VDSErrorException: Failed to CreateStorage
And
Cannot acquire host id: ('b4f05e3e-d714-45c4-adf9-8d269418338f',
SanlockException(-203, 'Sanlock lockspace add failure', 'Sanlock
exception'))
Reading the mail archive, I can see that this is a know issue but I do not
see the workaround. Can someone enlight me ?
Also, is there somewhere a coherent instruction set in order to make NFSv3
work (including iptables settings). I had to stop iptables to unblock
myself...
10 years, 11 months
