- Users - oVirt List Archives

[Users] Hosted Engine always reports "unknown stale-data"
by Andrew Lau 03 Feb '14

03 Feb '14

Hi, I was wondering if anyone has this same notice when they run: hosted-engine --vm-status The "engine status" will always be "unknown stale-data" even when the VM is powered on and the engine is online. engine-health will actually report the correct status. eg. --== Host 1 status ==-- Status up-to-date : False Hostname : 172.16.0.11 Host ID : 1 Engine status : unknown stale-data Is it some sort of blocked port causing this or is this by design? Thanks, Andrew

3 11

[Users] Full-size display Windows vs Fedora 20 guests
by Bob Doolittle 03 Feb '14

03 Feb '14

Hi, When I select View/Full Screen on a VM running a Windows guest, the display resolution automatically adjusts to fit the new canvas. However, when I do this on a VM running Fedora 20, it doesn't. Nor do I know how to query the new canvas size so that I can issue a manual xrandr command to fit it (without doing an ssh into the system and somehow finagling xwininfo to give me the size). In the guest I am running spice-vdagent-0.15.0-1, and restarting it has no effect. Shouldn't it be the one responsible for display optimization? -Bob

3 5

[Users] changing hostname in ovirt
by Sven Kieske 03 Feb '14

03 Feb '14

Hi, maybe I should write an RFE BZ for this but there might be a technical limitation, I don't know. What I want: I have a Host in ovirt which is in status "down" and/or "maintenance". I want to change the hostname/ip ovirt uses to connect to this host. 2. RFEs: 1. This is not possible via webadmin, the address field is grayed out (there seem to be conflicting design patterns for the webadmin, other buttons which don't work don't get grayed out for some reason, instead they throw errors..) 2. make it possible to change the hostname/ip via API from what I've seen so far I have to hack the database to make this change happen in 3.3.2 Is there any technical reason why this is not possible? ovirt should be happy with the UUIDs for the host and should not bother about the hostname. Is there a way to alter the hostname/ip which I don't know? Could someone point me to the right table and how to alter it without breaking it? Thank you! -- 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

4 9

[Users] ovirt test day: HA VM Reservation feature test summary
by Moti Asayag 03 Feb '14

03 Feb '14

Hi All, In the latest ovirt-test-day i've tested the HA VM resource reservation feature [1] according to the basic scenarios as described on [2]. The new feature notifies the admin via an event log about his cluster inability to preserve resources for HA VMs. I've reported 2 bugs based on the behavior: The cluster check doesn't consider the state of the cluster's hosts when it calculates the resources [3] and a minor issue of the audit log translation into a message [4]. [1] http://www.ovirt.org/Features/HA_VM_reservation [2] http://www.ovirt.org/OVirt_3.4_TestDay#SLA [3] Bug 1057579 - HA Vm reservation check ignores host status https://bugzilla.redhat.com/show_bug.cgi?id=1057579 [4] Bug 1057584 - HA Vm reservation event log is not well resolved https://bugzilla.redhat.com/show_bug.cgi?id=1057584 Thanks, Moti

2 3

[Users] Error message constantly being reported
by Jon Archer 03 Feb '14

03 Feb '14

Hi All, Constantly seeing this message in the logs: vdsm vds ERROR vdsm exception occured#012Traceback (most recent call last):#012 File "/usr/share/vdsm/BindingXMLRPC.py", line 952, in wrapper#012 res = f(*args, **kwargs)#012 File "/usr/share/vdsm/gluster/api.py", line 54, in wrapper#012 rv = func(*args, **kwargs)#012 File "/usr/share/vdsm/gluster/api.py", line 306, in tasksList#012 status = self.svdsmProxy.glusterTasksList(taskIds)#012 File "/usr/share/vdsm/supervdsm.py", line 50, in __call__#012 return callMethod()#012 File "/usr/share/vdsm/supervdsm.py", line 48, in <lambda>#012 **kwargs)#012 File "<string>", line 2, in glusterTasksList#012 File "/usr/lib64/python2.6/multiprocessing/managers.py", line 740, in _callmethod#012 raise convert_to_error(kind, result)#012GlusterCmdExecFailedException: Command execution failed#012error: tasks is not a valid status option#012Usage: volume status [all | <VOLNAME> [nfs|shd|<BRICK>]] [detail|clients|mem|inode|fd|callpool]#012return code: 1 looks like an option which isn't recognised by the "gluster volume status" command. Any ideas how to resolve? It's not causing any problems, but I would like to stop it. Cheers Jon

3 5

[Users] about the size of an offline snapshot
by Sandy Sun 03 Feb '14

03 Feb '14

Assign 20G virtio disk for VM, create an offline snapshot, find the true size of Vm-disk bigger than the virtual size (assign size)? I want to know how to compute the true size of vm-disk. Anbody can tell me the answer ? thanks.

4 5

[Users] oVirt Node Interface Renaming Problem
by David Li 03 Feb '14

03 Feb '14

Hi, I am using oVirt node 3.0.3. It seems systemd renames all the interfaces from ethx to something else. Not sure why but this creates lots of problems for some old scripts. For example: [root@localhost ~]# dmesg | grep -i eth0 [ 2.441579] bnx2 0000:10:00.0 eth0: Broadcom NetXtreme II BCM5709 1000Base-SX (C0) PCI Express found at mem fa000000, IRQ 30, node addr 5c:f3:fc:20:6e:58 [ 27.222803] systemd-udevd[822]: renamed network interface eth0 to enp16s0f0 Is there anyway to prevent this? Thanks.

3 2

[Users] How to scratch previous versions reports and dwh installation?
by Gianluca Cecchi 03 Feb '14

03 Feb '14

Hello, I have some problems, I think they are also due to PosgreSQL access rules changes from f18 to f19. In particular if one has passed from fedora 18 to Fedora 19 and form 3.2 to 3.3 as I did; see https://bugzilla.redhat.com/show_bug.cgi?id=1009335 Right now I have a problem with my AIO installation where trying to upgrade from 3.3.2 to 3.3.3 I got many problems related to RDBMS. Is there a way to completely get rid of prevous installtion and reinstall cleanly in 3.3.3 as I did on another system without getting problems. I tried removing packages: yum remove ovirt-engine-dwh ovirt-engine-reports and deleting related databases: drop database ovirt_engine_history; drop database ovirtenginereports; But then trying to install again packages and runnign setup I get this in log: 2014-01-30 00:37:30::DEBUG::common_utils::446::root:: running sql query on host: localhost, port: 5432, db: ovirt_engine_history, user: engine, query: 'select 1 from history_configuration;'. 2014-01-30 00:37:30::DEBUG::common_utils::907::root:: Executing command --> '/usr/bin/psql --pset=tuples_only=on --set ON_ERROR_STOP=1 --dbname ovirt_engine_history --host localhost --port 5432 --username engine -w -c select 1 from history_configuration;' in working directory '/root' 2014-01-30 00:37:30::DEBUG::common_utils::962::root:: output = 2014-01-30 00:37:30::DEBUG::common_utils::963::root:: stderr = ERROR: relation "history_configuration" does not exist LINE 1: select 1 from history_configuration; Can anyone tell if I have to run any other purge operation, suche as any users or table on engine db itself? And also what would be the expected configuration of /var/lib/pgsql/data/pg_hba.conf and eventually /var/lib/pgsql/data/postgresql.conf for a fedora 19 system so that a setup for reports and dwh could complete? Gianluca

2 5

[Users] ovirt 3.4.0-0.5.beta1 and OpenLDAP groups
by Winfried de Heiden - Voorwinde 03 Feb '14

03 Feb '14

Hi All, I managed to use OpenLDAP to integrate with oVirt 3.4.0-0.5.beta1. For this, I followed (more or less, I used a Raspberry Pi and Raspbian) instructions as found on http://www.ovirt.org/LDAP_Quick_Start It all seems to work well, I am able to connect to a domain, login etc. and assign some roles to users. However, I cannot use (ldap) groups it seems. I cann add a group in the ovirt gui, but (in the tab General) "Active" remain "false". A I missing something...? Winfried

3 3

[Users] EL6 manual hypervisor bare minimum with ovirt-hosted-engine packages
by Zach Musselman 03 Feb '14

03 Feb '14

Hello, I'd like to run the new ovirt-hosted-engine option in RHEV/Ovirt. This appears to not be a built in option of the Ovirt node image. I'd like to create an EL 6 install that as closely resembles the Node/Hypervisor image. How can I keep the amount of installed packages to a minimum and have the self hosted packages installed? Please post how I can get to the node/rhev-h image manually with as similar of packages and the ovirt-hosted-engine packages. # yum install ovirt-hosted-engine-setup # yum install ovirt-hosted-engine-ha Thanks -- Zach

2 1