- Users - oVirt List Archives

Re: [ovirt-users] Some questions
by Maor Lipchuk 18 Aug '14

18 Aug '14

It depends on the supportedENGINEs which getVdsCaps will return. Generally, VDSM for 3.x should keep supporting its 3.x previous minor versions. Regards, Maor ----- Original Message ----- From: "David cano bailen" <davidcano92(a)gmail.com> To: "Maor Lipchuk" <mlipchuk(a)redhat.com> Sent: Thursday, August 14, 2014 6:02:17 PM Subject: Re: [ovirt-users] Some questions hi, my concern is that we have some critical MV. If the node enters into non -operational state after upgrading the engine. Will be machines up? I've been searching and upgrading should be progressive, 3.0 to 3.1, 3.1 to 3.2 .... I guess if the nodes VDSM update to the latest version, and later I'll update the engine version by version nodes wouldnt enter on non-operational state and not suffer any court in MV. Am I right? Thank you very much for your help 2014-08-14 11:10 GMT+02:00 Maor Lipchuk <mlipchuk(a)redhat.com>: > Hi David, > > See my answers inline > > Regards, > Maor > > > ----- Original Message ----- > > From: "David cano bailen" <davidcano92(a)gmail.com> > > To: users(a)ovirt.org > > Sent: Wednesday, August 13, 2014 2:04:10 PM > > Subject: [ovirt-users] Some questions > > > > Hi, i installed Engine 3.0 and Node Ovirt 2.6, but it seems very > unestable. > > Sometimes machines doesnt starts, nodes doesnt vinculate or UI doesnt > work > > fine. I have only 2 nodes and storage its on the same physical machine > than > > engine. > > > > -¿Does Node Ovirt 3.4 work with Engine 3.0? > Engine 3.0 should support backward compatibility for ovirt 3.4. You can > verify that on the node be executing the following command : "vdsClient -s > 0 getVdsCaps" (-s for secure), > There, you should see a supportedENGINEs list. > > -¿Its necessary shutdown Nodes to do engine upgrade? > no, but take in consideration that after upgrade, those hosts might get > into non-operational state, in case VDSM version should be upgraded also. > > -¿Are there any problem having the same physical machine for engine and > > storage? ¿what are advantages have on diferents machines? > There is no limitation of doing that, > it's simply a question of the architecture you want to support. > Separating the engine from the Storage server, might be useful to make it > more high available. > > > > Thank you > > > > > > _______________________________________________ > > Users mailing list > > Users(a)ovirt.org > > http://lists.ovirt.org/mailman/listinfo/users > > >

2 1

ovirt with 389 server inactive groups
by Paul Robert Marino 18 Aug '14

18 Aug '14

I have ovirt engine running and connected to a 389 server with the memberof plugin enabled and working properly. I can add users and assign them to roles without any issues. when I look at a user I can see all the LDAP groups they are a member of. when I run engine-manage-domains -action=validate it tells me the domain is valid. here is my problem when I try to assign a role to an LDAP group it looks like it works but in the general tab when under the group it tells me the status is Inactive. dose any one know how to enable the group?

5 20

[SOLVED] RE: Reactivate a storage domain
by Dwight Schauer 18 Aug '14

18 Aug '14

Hello Ovirt users, Well, I solved the problem. When the NFS volume had filled up that knocked out non Ovirt managed VMs as well including DHCP servers. I'd been using DHCP for the management network. The Ovirt hypervisors never refreshed when I brought the DHCP servers back. (The management network was no longer connected). I went into one Ovirt hypervisor, configured it with static, and after a couple of minutes it become the SPM on its own without my involvement and the storage domains came back. Sincerely, Dwight Schauer -----Original Message----- From: users-bounces(a)ovirt.org [mailto:users-bounces@ovirt.org] On Behalf Of Dwight Schauer Sent: Sunday, August 17, 2014 12:37 PM To: users(a)ovirt.org Subject: [ovirt-users] Reactivate a storage domain Hello Ovirt users, How do I unlock a storage domain? The underlying NFS volume had filled up (I corrected that). I tried to reactivate the storage domain (it had deactivated to the volume filling up) but then it went into a locked state According to df -h: 8.0T 6.0T 2.1T 75% # 6T used, 2.1T free Ovirt thinks it only has 6.0T as is full (well, 1GB free, which is now false, but I can't convince it other wise). (I increased the size of the volume after VMs went into pause due to it filling up). Now no hosts wants to be the SPM... So my ovirt managed data center is down until I reactivate the NFS storage domain (that is the master). The storage domain reverts back to an inactive state after a while, but whenever I try to reactivate it goes to the locked state. Sincerely, Dwight Schauer _______________________________________________ Users mailing list Users(a)ovirt.org http://lists.ovirt.org/mailman/listinfo/users

1 0

Reactivate a storage domain
by Dwight Schauer 17 Aug '14

17 Aug '14

Hello Ovirt users, How do I unlock a storage domain? The underlying NFS volume had filled up (I corrected that). I tried to reactivate the storage domain (it had deactivated to the volume filling up) but then it went into a locked state According to df -h: 8.0T 6.0T 2.1T 75% # 6T used, 2.1T free Ovirt thinks it only has 6.0T as is full (well, 1GB free, which is now false, but I can't convince it other wise). (I increased the size of the volume after VMs went into pause due to it filling up). Now no hosts wants to be the SPM... So my ovirt managed data center is down until I reactivate the NFS storage domain (that is the master). The storage domain reverts back to an inactive state after a while, but whenever I try to reactivate it goes to the locked state. Sincerely, Dwight Schauer

1 0

list of answer file parameters?
by Jorick Astrego 17 Aug '14

17 Aug '14

Hi, I'm looking everywhere for a full list of "OVESETUP" parameters that can be used in the installer answerfile. Searched the red hat manual and found some in: "3.6. Passwords in Red Hat Enterprise Virtualization Manager" But there are more. I can get it from a finished install but some parameters only apply to some types of installation. I found this one form lzap: https://gist.github.com/lzap/7a8cbb7f44d41e4ad171 But I don't know if it's complete or just for his personal use. No hurry, I know you are all busy ;-) Kind regards, Jorick Astrego Netbulae B.V.

3 10

Instructions for novices
by OvirtAndKvm 17 Aug '14

17 Aug '14

--_=_swift_v4_1408258037_cf16571b98839ffb61d2087951ae8375_=_ Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: quoted-printable Hi, Can someone point me to instructions for building a C= entOS server that would be ready for installing the self hosted node= . I have a 7x 500GB HD, 16 GB RAM, i7 computer that I have inst= alled CentOS 6.5 (Software RAID6) to and would now like to install t= he self-hosted-engine. While I do have experience with D= ebian as a server and some experience with CentOS, I have not so far= found any information regarding how to build a basic server ready fo= r installing oVirt packages, for example configuring static IP addre= sses, hostname, hosts file, maybe NFS shares ? or iSCSI shares ?= http://www.ovirt.org/Download#Install_oVirt Step = 6 says "Follow the on screen prompts to configure and install the en= gine ", however knowing what to supply for the prompts and what to h= ave already configured on your server would be very helpful. ht= tp://www.ovirt.org/Hosted_Engine_Howto "During the deployment you'll= be asked for input on host name, storage path and other relevant in= formation ", it is how to set up the environment for answering these= questions that I would like assistance with (e.g. a HowTo ?)= I have once before installed the "ovirt-engine-setup-plug= in-allinone" in a VMware Workstation VM, but not without many challe= nges. I would now like to install the "ovirt-hosted-engine-= ha" The second prompt said "Please specify the full shared st= orage connection path to use (example: host:/path):", however it doe= s not say that it will create this or whether you need to have this = created earlier ?=C2=A0 At this time I don't have a static IP address = nor the hosts file containing the host name, this is just a newly bui= lt CentOS server with minimum GUI, updates applied and no other chan= ges.=C2=A0 So is there any documentation which would explain what othe= r changes are needed and how to do this? http://www.ovirt.o= rg/Quick_Start_Guide Two packages, ovirt-hosted-engine-setup and ovi= rt-hosted-engine-ha provide the setup and services necessary to deplo= y a self-hosted-engine oVirt environment. For more on the oVirt S= elf-Hosted Engine, see the oVirt Installation Guide.=20 https://a= ccess.redhat.com/documentation/en-US/Red_Hat_Enterprise_Virtualization/3= .3/html/Installation_Guide/Configuring_the_Self-Hosted_Engine.ht= ml http://blog.gluster.org/2014/05/ovirt-3-4-glusterized= / The Hosted Engine feature relies on NFS storage to house the m= anagement VM. --_=_swift_v4_1408258037_cf16571b98839ffb61d2087951ae8375_=_ Content-Type: text/html; charset=UTF-8 Content-Transfer-Encoding: quoted-printable <html> <head> <style type=3D"text/css"> body,p,td,div,span= { font-size:13px; font-family:Arial, Helvetica, sans-serif;= }; body p{ margin:0px; } </style> </head= > <body>Hi, Can someone point me to instructions for building= a CentOS server that would be ready for installing the self hosted node= . I have a 7x 500GB HD, 16 GB RAM, i7 computer that I have instal= led CentOS 6.5 (Software RAID6) to and would now like to install the sel= f-hosted-engine. While I do have experience with Debian as a serv= er and some experience with CentOS, I have not so far found any informat= ion regarding how to build a basic server ready for installing oVirt pac= kages, for example configuring static IP addresses, hostname, hosts file= , maybe NFS shares ? or iSCSI shares ? http://www.ovirt.org/Downl= oad#Install_oVirt Step 6 says "Follow the on screen prompts to co= nfigure and install the engine ", however knowing what to supply for= the prompts and what to have already configured on your server would be= very helpful. http://www.ovirt.org/Hosted_Engine_Howto "Durin= g the deployment you'll be asked for input on host name, storage path an= d other relevant information ", it is how to set up the environment = for answering these questions that I would like assistance with (e.g. a = HowTo ?) I have once before installed the "ovirt-engine-setup= -plugin-allinone" in a VMware Workstation VM, but not without many chall= enges. I would now like to install the "ovirt-hosted-engine-ha"<b= r> The second prompt said "Please specify the full shared storage con= nection path to use (example: host:/path):", however it does not say tha= t it will create this or whether you need to have this created earlier ?=   At this time I don't have a static IP address nor the hosts file = containing the host name, this is just a newly built CentOS server with = minimum GUI, updates applied and no other changes.  So is there any= documentation which would explain what other changes are needed and how= to do this? http://www.ovirt.org/Quick_Start_Guide Two packag= es, ovirt-hosted-engine-setup and ovirt-hosted-engine-ha=20 provide th= e setup and services necessary to deploy a self-hosted-engine=20 oVirt= environment. For more on the oVirt Self-Hosted Engine, see the=20 oVi= rt Installation Guide. https://access.redhat.com/documentati= on/en-US/Red_Hat_Enterprise_Virtualization/3.3/html/Installation_Guide/C= onfiguring_the_Self-Hosted_Engine.html http://blog.gluster.org/20= 14/05/ovirt-3-4-glusterized/ The Hosted Engine feature relies on NFS = storage to house the management VM. </body></html> --_=_swift_v4_1408258037_cf16571b98839ffb61d2087951ae8375_=_--

1 0

Vm:s can not be started
by Jonas Israelsson 15 Aug '14

15 Aug '14

I screwed up massively today when accidentally assigned a new virtual network (on my host) the same IP-address as on my iscsi SAN subnet and as a result the whole DC collapsed (I only have one host). Recovering was not quite as easy since the host network configuration only can be managed while the host is either in up, down or maintenance mode, and mine seem to be somewhere in between. I finally however managed to remove the network from the cluster tab marking the net on the hosts nic with a "?" and when the host came back online and I could later also be remove the faulty network from the nic. All VM:s now where either in mode pause or down, but I could not get any started to I just powered all of them down. Now everything DC-wise it up and running again still several VM:s can not be started. Out of about 30 vm;s I have only managed to get one up and running. The following it written to the engine.log when trying to fire up a non-working VM. 2014-08-15 15:26:23,638 INFO [org.ovirt.engine.core.vdsbroker.vdsbroker.DestroyVDSCommand] (DefaultQuartzScheduler_Worker-53) START, DestroyVDSCommand(HostName = ft.elementary.se, HostId = 97c43383-3770-4b0c-810c-a597bb187898, vmId=1b13210a-bf05-41f7-b42e-c4b34ca2214d, force=false, secondsToWait=0, gracefully=false), log id: 1aaa6470 2014-08-15 15:26:34,110 INFO [org.ovirt.engine.core.vdsbroker.vdsbroker.DestroyVDSCommand] (DefaultQuartzScheduler_Worker-53) FINISH, DestroyVDSCommand, log id: 1aaa6470 2014-08-15 15:26:34,117 INFO [org.ovirt.engine.core.dal.dbbroker.auditloghandling.AuditLogDirector] (DefaultQuartzScheduler_Worker-53) Correlation ID: null, Call Stack: null, Custom Event ID: -1, Message: VM moleman is down. Exit message: Requested operation is not valid: domain 'moleman' is already active. 2014-08-15 15:26:34,118 INFO [org.ovirt.engine.core.vdsbroker.VdsUpdateRunTimeInfo] (DefaultQuartzScheduler_Worker-53) Running on vds during rerun failed vm: null 2014-08-15 15:26:34,119 INFO [org.ovirt.engine.core.vdsbroker.VdsUpdateRunTimeInfo] (DefaultQuartzScheduler_Worker-53) VM moleman (1b13210a-bf05-41f7-b42e-c4b34ca2214d) is running in db and not running in VDS ft.elementary.se 2014-08-15 15:26:34,136 ERROR [org.ovirt.engine.core.vdsbroker.VdsUpdateRunTimeInfo] (DefaultQuartzScheduler_Worker-53) Rerun vm 1b13210a-bf05-41f7-b42e-c4b34ca2214d. Called from vds ft.elementary.se 2014-08-15 15:26:34,142 INFO [org.ovirt.engine.core.dal.dbbroker.auditloghandling.AuditLogDirector] (org.ovirt.thread.pool-6-thread-44) Correlation ID: 3335521e, Job ID: f334270e-2576-46c7-8c69-99c54d868d57, Call Stack: null, Custom Event ID: -1, Message: Failed to run VM moleman on Host ft.elementary.se. 2014-08-15 15:26:34,150 INFO [org.ovirt.engine.core.bll.RunVmCommand] (org.ovirt.thread.pool-6-thread-44) Lock Acquired to object EngineLock [exclusiveLocks= key: 1b13210a-bf05-41f7-b42e-c4b34ca2214d value: VM , sharedLocks= ] 2014-08-15 15:26:34,155 INFO [org.ovirt.engine.core.vdsbroker.IsVmDuringInitiatingVDSCommand] (org.ovirt.thread.pool-6-thread-44) START, IsVmDuringInitiatingVDSCommand( vmId = 1b13210a-bf05-41f7-b42e-c4b34ca2214d), log id: ee3a648 2014-08-15 15:26:34,156 INFO [org.ovirt.engine.core.vdsbroker.IsVmDuringInitiatingVDSCommand] (org.ovirt.thread.pool-6-thread-44) FINISH, IsVmDuringInitiatingVDSCommand, return: false, log id: ee3a648 2014-08-15 15:26:34,160 WARN [org.ovirt.engine.core.bll.RunVmCommand] (org.ovirt.thread.pool-6-thread-44) CanDoAction of action RunVm failed. Reasons:VAR__ACTION__RUN,VAR__TYPE__VM,VAR__ACTION__RUN,VAR__TYPE__VM,VAR__ACTION__RUN,VAR__TYPE__VM,SCHEDULING_ALL_HOSTS_FILTERED_OUT 2014-08-15 15:26:34,161 INFO [org.ovirt.engine.core.bll.RunVmCommand] (org.ovirt.thread.pool-6-thread-44) Lock freed to object EngineLock [exclusiveLocks= key: 1b13210a-bf05-41f7-b42e-c4b34ca2214d value: VM , sharedLocks= ] 2014-08-15 15:26:34,168 INFO [org.ovirt.engine.core.dal.dbbroker.auditloghandling.AuditLogDirector] (org.ovirt.thread.pool-6-thread-44) Correlation ID: 3335521e, Job ID: f334270e-2576-46c7-8c69-99c54d868d57, Call Stack: null, Custom Event ID: -1, Message: Failed to run VM moleman (User: isrjo). I was on Ovirt 3.4.0-0.13.rc.el6 and right or wrong I decided to upgrade the engine to a released version and I'm now on 3.4.0-1.el6, it however made no difference Any help would be greatly appreciated. Rgds Jonas

2 2

oVirt italian translation
by Giorgio Bersano 15 Aug '14

15 Aug '14

Hi oVirt UX team, we (Gianluca and me) are slowly progressing on the job of having italian as a fully translated language in oVirt. One of our problems is that we don't know what will be the outcome of our work until an italian embedding release will be available. Probably the first one will not be something to be really proud. To have a prototipe to get a glance of how it is now would be really useful, don't care about general quality. So I asked Sandro if could it be possible to have italian translation in the 3.5RC release at the point where it is now (almost 35% done) and then remove it from the official 3.5 release but he said it is not on his possibilities. An alternative he suggested was to have a "special" 3.5 development branch (something like ovirt-3.5-ita) just to give us the opportunity to get a taste of what it would be in the end. Please tell us what do you think about. Thanks, Giorgio.

4 17

cancel ticket in ovirt-engine
by mhd 15 Aug '14

15 Aug '14

hi.... my scenario: i have multi server installed rhev or ovirt-engine and one sever installed rhevm or ovirt-engine ..... and i bought thin client which connect by spice to rhev server and controll them... i can use thin client for connect to rhev server but the problem every connect i should make new ticket for spice protocol.... when i run vm by rhevm interface the log for vm in this path :/var/log/libvirt/qemu/vmname.log is seem run qemu command and enable options for vm,,, my question how i can add this option (disable-ticket) when user start run vm ??? another mean i want cancel ticket or make it lifetime .... thanks.. i my question not understaded please tell me...

2 1

Ovirt SSL Question
by Punit Dambiwal 15 Aug '14

15 Aug '14

Hi All, I have one question regarding the SSL settings in Ovirt....let me explain my environment first :- 1. Ovirt engine :- mgmt.3linux.com 2. Standalone websocket proxy :- web-proxy.3linux.com 3. Our Own Portal :- portal.3linux.com We have the above architecture...we fetch the VM console from the websocket proxy to our own portal through API....because still we are using selfsigned certificate...we need to trust the certificate every time,whenever we open the VM console... (https://<web-proxy.3linux.com >:<port>) When we initiate the VM console through our own web portal the url ( https://portal.3linux.com/content/ovirt/noVNC/vm-console.php?id=6e0caf73-ae… we accept the SSL certificate with https://<web-proxy.3linux.com>:<port> ....then it will open as expected but if we didn't accept the certificate manually...then it through failed to connect:1006 error... We don't want that every time end user will accept the certificate manually...as our link to open VM console is different then webproxy.... Now we want to replace the self signed certificate with valid SSL....can any one tell me where we need to put the certificates and how to generate the CSR for them and how many SSL we need to purchase to make this thing workable without accepting the certificate everytime.... Thanks, Punit

3 12