- Users - oVirt List Archives

[Users] oVirt 3.1 -> 3.2 upgrade path
by Alex Leonhardt 19 Feb '13

19 Feb '13

Hi, Anyone been doing upgrades of ovirt engine + nodes from 3.1 to 3.2 yet ? what is the proposed upgrade path ? nodes first, then engine or other way around ? As the "domain" is a 3.1 domain currently, the nodes (HVs) would become disconnected from the engine if the engine was upgrade first ? Or, if you upgrade the nodes (HVs) first to 3.2 will it be able to run in a "3.1 compatibility" mode to still be controlled by a 3.1 engine, and when engine is upgraded to 3.2 will the node (HV) then switch to 3.2 automatically / or is a restart / reconfiguration required ? Is a upgrade possible w/o taking down the node / e.g. do I have to stop all VMs running on that node (HV) ? Thanks, Alex -- | RHCE | Senior Systems Engineer | www.vcore.co | www.vsearchcloud.com |

1 0

[Users] Discussion... work appears to be needed in areas of recovery and cleanup
by Rick Beldin 19 Feb '13

19 Feb '13

Hi... I've been working with both RHEV and Ovirt and can see distinct changes and improvements with regards to how Ovirt is emerging and changing. One area that still disturbs me is in the area of recovery and cleanup, specifically around the ovirt-engine (rhevm) node. I've observed a few things about the project: - backup of a engine db seems to be a fairly manual process with many steps. There are many files listed that need to be backed up manually as listed in the RHEVM admin manual. pg_dump seems to automate the dumping of the db, but other config files are left up to the admin. - restoration appears to be equally complex - engine-cleanup (rhevm-cleanup) doesn't always cleanup - there appear to be instances where the only way to get a solid base is reinstallation. I've removed and reinstalled components so many times it is now second nature. ;) I'm curious about: - Are there technical issues (beyond the code not being written yet) that prevent backup and restore procedures from working correctly to save and restore an engine environment? I would expect that a test for that operation would be to setup an environment, take a snapshot, and restore the whole thing to another management node. - Are all the config files properly identified? The RHEVM Admin manual has an extensive list but is it exhaustive? - What technically prevents engine-cleanup from really *cleaning* a system? My expectation would be that I should be able to run engine-cleanup and completely wipe out the setup. All databases, certs, config files and the like returned to their original, pristine state. Perhaps another script - engine-restore-defaults? - needs to go beyond where engine-cleanup goes today? As a support professional, these are areas that I think warrant investment and I wonder what is gating the development here. Thanks, Rick -- +----------------------------------------------------------------------+ | Rick Beldin | Hewlett-Packard Company | | email: Rick.Beldin(a)hp.com | | | TCE&Q - Linux ERT | 5555 Windward Parkway West | | | Alpharetta, GA 30004 | +----------------------------------------------------------------------+

3 2

Re: [Users] Rest-api to fetch the hosts details ( active vm's , CPU , Physical memory etc.)
by Michael Pasternak 19 Feb '13

19 Feb '13

Hi Romil, On 02/19/2013 11:22 AM, Romil Gupta wrote: > Hi Michael , > > > Thanks for ur continuous help. I have one more query here .... > > I want to fetch the host elements like .. > Type > address > status > No. of CPUs > cpu->type > cpu -> name->topology->cores->sockets > memory > summary > version > > Is is the correct way of doing to get these details? > > hosts=api.hosts.list() > for host in hosts: > type=api.hosts.get(host.id <http://host.id>).type > address=api.hosts.get(host.id <http://host.id>).address no, please use getter methods instead as they're encapsulating attributes that may be called differently in python, for instance 'type' attribute in python renamed to 'type_' cause 'type' is preserved name. you can see all available get_X() methods by running code bellow: #> print [token for token in type(host.superclass).__dict__.keys() if token.startswith('get_')] > > I got some exception : > address=api.hosts.get(host.id <http://host.id>).get_address() > AttributeError: 'NoneType' object has no attribute 'get_address' this can only happen if your host lookup has failed. > > > Please suggest me the correct way of doing it ....if have referred > https://access.redhat.com/knowledge/docs/en-US/Red_Hat_Enterprise_Virtualiz… > > Regards, > Romil > > > > > > > On Wed, Jan 30, 2013 at 6:50 PM, Michael Pasternak <mpastern(a)redhat.com <mailto:mpastern@redhat.com>> wrote: > > On 01/30/2013 03:09 PM, Romil Gupta wrote: > > thanks for all your guidance , now I m able to fetch the details of a host using > > the below script : > > > > hosts=api.hosts.list() > > for host in hosts: > > print "host name--> %s id--->> %s \n"%(host.name <http://host.name> <http://host.name> , host.id <http://host.id> <http://host.id>) > > clusterid=api.hosts.get(host.name <http://host.name> <http://host.name>).cluster.id <http://cluster.id> <http://cluster.id> > > print clusterid > > > > hostname=api.hosts.get(host.name <http://host.name> <http://host.name>) > > statistic=hostname.statistics.list() > > i=0 > > while i < 14: > > print statistic[i].name > > print statistic[i].description > > print statistic[i].unit > > print statistic[i].values.value[0].datum > > i=i+1; > > > > > > summary=api.get_summary() > > print summary > > > > How I can print the summary , its only return the Object?? > > this is summary object structure: > > <summary> > <vms> > <total></total> > <active></active> > </vms> > <hosts> > <total></total> > <active></active> > </hosts> > <users> > <total></total> > <active></active> > </users> > <storage_domains> > <total></total> > <active></active> > </storage_domains> > </summary> > > you can access properties directly, like this: > summary.hosts.active > > > > > Thanks, > > Romil > > > > > > On Wed, Jan 30, 2013 at 4:52 PM, Michael Pasternak <mpastern(a)redhat.com <mailto:mpastern@redhat.com> <mailto:mpastern@redhat.com <mailto:mpastern@redhat.com>>> wrote: > > > > > > Romil, > > > > On 01/30/2013 12:18 PM, Romil Gupta wrote: > > > Hi, > > > > > > Is this is a right way to get it ?? > > > > > > statistics=params.Host(host.name <http://host.name> <http://host.name> <http://host.name>).get_statistic() > > > > 1. first you need to fetch the host to see it's statistics (by doing params.Host(...) you creating > > host parameters holder which is needed for adding new host to the system) > > > > 2. get_x() getters used to access object attributes, while collections are exposed as properties, do > > > > 1. myhost = api.hosts.get(name="xxx") > > 2. myhost.statistics.list() > > 3. loop over returned collection of statistics to find what you're looking for > > > > - note, statistic objects are complex types, you can look for data at: > > > > statistics[i].unit // the unit of the holder data > > statistics[i].values.value[0].datum // actual data > > > > > print statistics > > > > > > summary=params.Host(host.name <http://host.name> <http://host.name> <http://host.name>).get_summary() > > > > summary() is an api method, do: > > > > 1. api = API(url='', username='', password='') > > 2. api.get_summary() > > > > > > > print summary > > > > > > > > > Output is : none > > > > > > Thanks > > > Romil > > > > > > > > > On Wed, Jan 30, 2013 at 2:04 PM, Michael Pasternak <mpastern(a)redhat.com <mailto:mpastern@redhat.com> <mailto:mpastern@redhat.com <mailto:mpastern@redhat.com>> > <mailto:mpastern@redhat.com <mailto:mpastern@redhat.com> <mailto:mpastern@redhat.com <mailto:mpastern@redhat.com>>>> wrote: > > > > > > > > > Hi Romil, > > > > > > On 01/30/2013 10:17 AM, Romil Gupta wrote: > > > > Hi all , > > > > > > > > how I can get the hosts details like Active VM's , > > > > > > host doesn't have running vms attribute, instead you > > > can see in the guest on which host it's running, > > > > > > general system summary you can see at api.get_summary() > > > > > > Number of CPU's , CPU name , CPU type , > > > > > > these are host attributes > > > > > > Physical Memory (used , free ) , swap size and other parameters > > > > > > these are host.statistics attributes > > > > > > > using ovirt-engine-sdk-3.2.0.5-1. > > > > > > > > > > > > > > > > Regards, > > > > Romil > > > > > > > > -- > > > > I don't wish to be everything to everyone, but I would like to be something to someone. > > > > > > > > > -- > > > > > > Michael Pasternak > > > RedHat, ENG-Virtualization R&D > > > > > > > > > > > > > > > -- > > > I don't wish to be everything to everyone, but I would like to be something to someone. > > > > > > -- > > > > Michael Pasternak > > RedHat, ENG-Virtualization R&D > > > > > > > > > > -- > > I don't wish to be everything to everyone, but I would like to be something to someone. > > > -- > > Michael Pasternak > RedHat, ENG-Virtualization R&D > > > > > -- > I don't wish to be everything to everyone, but I would like to be something to someone. -- Michael Pasternak RedHat, ENG-Virtualization R&D

1 0

[Users] Connecting to portals through ip and consoles
by Gianluca Cecchi 18 Feb '13

18 Feb '13

Hello, I'm verifying this behavior with an oVirt 3.2 beta in all-in-one config if the client doesn't know the hostname of engine and uses instead its ip to connect to webadmin portal or user portal, the gui works, but then when it connects to a spice console of a VM it receives the error of unable to connect to connet to graphics server as soon as I put engine entry hostname ip in /etc/hosts of a fedora 17 client it can now connect to spice console. Is this expected behavior? Any workaround (other than hosts one I tested) in case of lab/test environment when no dns feasible? Thanks Gianluca

1 0

[Users] USB dongle (sigh...)
by Nicolas Ecarnot 18 Feb '13

18 Feb '13

Hi, A user of mine is trying to ruin oVirt's reputation by asking me how will I handle the USB dongle issue. I found nothing like USB passthrough drivers, and I don't intend to use such a bad solution. Is there anyone here who faced such a request? I saw http://www.dongleservice.com/ and http://www.digi.com/products/usb/anywhereusb, but I was wondering if anything lighter would exist (based on some dd if=/dev/goodOldLinuxTrickWeLove things...) The VM is a XP or 2003, and I already tried dd /dev/myUsbDevice of=dongle.iso, then mount it in the VM with daemon tool. The iso is mounting, but the content does not appear. Anyway, I'm not sure at all this will be enough emulation to convince the doubtful software *this* is its dongle. Regards, -- Nicolas Ecarnot

2 1

[Users] Restirct list of AD servers
by Keith Mitchell 18 Feb '13

18 Feb '13

I have a situation where the Active Directory domain I am trying to use as authentication for ovirt lists many servers all around the world. But... my server running ovirt is sitting behind a firewall that doesn't allow me to access all of them... only the local ones. We do have a "locater dns record" which we can query at a well known name and it will always return the local ip address of the AD server... but if you query the SRV records for the domain it will return all of the servers. I was able to add the domain using engine-manage-domains, and I tweaked the /etc/ovirt-engine/krb5.conf to only include the local AD servers where we can access, but that doesn't seem to be sufficient. Not sure if ovirt is querying the dns records on boot to get the list of servers to talk to or not, but it doesn't seem to be using /etc/ovirt-engine/krb.conf for this purpose. So... is there anyway to manually force it to use a certain server and not have it query dns? thanks.

2 3

[Users] oVirt 3.2 -- Now Available
by Mike Burns 18 Feb '13

18 Feb '13

The oVirt Team is pleased to announce that oVirt 3.2 is now available. There are a number of exciting new features in 3.2 [1]. Some of the highlights include: * UI Plugins * Port Mirroring * Networks as a top level Configuration * Japanese Localization * Storage Live Migration * Improved Gluster Integration * oVirt Node TUI Redesign * Node Plugins A Quick Start is available on the Wiki [2] Packages are currently available for Fedora 18 [3], with EL6 packages coming soon. Please see our Release Notes[4] for additional information If you run into any problems or need help, people are available on #ovirt on OFTC and users(a)ovirt.org Thanks The oVirt Team [1] http://www.ovirt.org/OVirt_3.2_release-management#Features [2] http://www.ovirt.org/Quick_Start_Guide [3] http://resources.ovirt.org/releases/stable/rpm/Fedora/18/ [4] http://www.ovirt.org/OVirt_3.2_release_notes

6 11

[Users] Failing on new setup
by Jaco 18 Feb '13

18 Feb '13

This is a multi-part message in MIME format. --------------040901050807070502000602 Content-Type: text/plain; charset=ISO-8859-1; format=flowed Content-Transfer-Encoding: 7bit Hi guys, New to oVirt - and "new" to Fedora after a *very long* break @ using Debian/Ubuntu almost exclusively for the last few years. Much has changed, so please pardon my ignorance - got some catching-up to do. I'm doing PoC of a number of systems, so need to give oVirt a fair shake, but I'm tripping up before I can even get going (does not bode well). I've tried following the QuickStart, all-in-one & other guides, & quickly came across issues. Next I'll have to start building from source - a prospect that will by no means present me with less issues. New F18, minimal headless server. Installed via yum as documented, but install failed at the last steps, as there seems to be an issue with the scripts' implementation of firewalld. A very nice chap @ the IRC channel managed to help be out to get over that initial hurdle - had to *completely* disable firewalld & select the 'None' option for the setup. (seems like a really bad idea to start off like that - disabling security - but this is a lab PoC, so OK, will let it slide for now) I've also disabled SELinux & IPtables via systemctl - so that we can eliminate that as an issue. Script(s) still don't complete without issue, but at lease the engine seems to come up & I can log in. After logging in, I try to (re)connect to the local hypervisor via (root@) $FQDN, localhost & 127.0.0.1, but it fails to come up (why root? OK....). It adds the host, but there seems to be connectivity issues. I'm seeing this in the GUI events: Host localhost installation failed. Command returned failure code 1 during SSH session 'root@localhost'. I'm seeing this in `tail -f /var/log/ovirt-engine/engine.log` http://pastebin.com/nVyNGSsf Relevant line being: ERROR [org.ovirt.engine.core.utils.ssh.SSHDialog] (pool-3-thread-31) SSH error running command root@localhost:'umask 0077; MYTMP="$(mktemp -t ovirt-XXXXXXXXXX)"; trap "chmod -R u+rwX \"${MYTMP}\" > /dev/null 2>&1; rm -fr \"${MYTMP}\" > /dev/null 2>&1" 0; rm -fr "${MYTMP}" && mkdir "${MYTMP}" && tar -C "${MYTMP}" -x && "${MYTMP}"/setup DIALOG/dialect=str:machine DIALOG/customization=bool:True': java.io.IOException: Command returned failure code 1 during SSH session 'root@localhost' This is a *fresh* box - I've reinstalled it a few times from scratch now - & I've re-installed oVirt through the manners documented, but I'm not getting anywhere fast. In this, google is not my friend, as results are sparse & not very helpful - i.e. if someone's found resolution, it has not been posted. Either there's a bug in the docco's or the code/scripts. Please help - I'd *really* like to give this a go Kind regards -Jaco --------------040901050807070502000602 Content-Type: text/html; charset=ISO-8859-1 Content-Transfer-Encoding: 7bit <html> <head> <meta http-equiv="content-type" content="text/html; charset=ISO-8859-1"> </head> <body bgcolor="#FFFFFF" text="#000000"> Hi guys, New to oVirt - and "new" to Fedora after a *very long* break @ using Debian/Ubuntu almost exclusively for the last few years. Much has changed, so please pardon my ignorance - got some catching-up to do. I'm doing PoC of a number of systems, so need to give oVirt a fair shake, but I'm tripping up before I can even get going (does not bode well). I've tried following the QuickStart, all-in-one & other guides, & quickly came across issues. Next I'll have to start building from source - a prospect that will by no means present me with less issues. New F18, minimal headless server. Installed via yum as documented, but install failed at the last steps, as there seems to be an issue with the scripts' implementation of firewalld. A very nice chap @ the IRC channel managed to help be out to get over that initial hurdle - had to *completely* disable firewalld & select the 'None' option for the setup. (seems like a really bad idea to start off like that - disabling security - but this is a lab PoC, so OK, will let it slide for now) I've also disabled SELinux & IPtables via systemctl - so that we can eliminate that as an issue. Script(s) still don't complete without issue, but at lease the engine seems to come up & I can log in. After logging in, I try to (re)connect to the local hypervisor via (root@) $FQDN, localhost & 127.0.0.1, but it fails to come up (why root? OK....). It adds the host, but there seems to be connectivity issues. I'm seeing this in the GUI events: <meta charset="utf-8"> Host localhost installation failed. Command returned failure code 1 during SSH session 'root@localhost'. I'm seeing this in `tail -f /var/log/ovirt-engine/engine.log` <a class="moz-txt-link-freetext" href="http://pastebin.com/nVyNGSsf">http://pastebin.com/nVyNGSsf</a> Relevant line being: ERROR [org.ovirt.engine.core.utils.ssh.SSHDialog] (pool-3-thread-31) SSH error running command root@localhost:'umask 0077; MYTMP="$(mktemp -t ovirt-XXXXXXXXXX)"; trap "chmod -R u+rwX \"${MYTMP}\" > /dev/null 2>&1; rm -fr \"${MYTMP}\" > /dev/null 2>&1" 0; rm -fr "${MYTMP}" && mkdir "${MYTMP}" && tar -C "${MYTMP}" -x &&  "${MYTMP}"/setup DIALOG/dialect=str:machine DIALOG/customization=bool:True': java.io.IOException: Command returned failure code 1 during SSH session 'root@localhost' This is a *fresh* box - I've reinstalled it a few times from scratch now - & I've re-installed oVirt through the manners documented, but I'm not getting anywhere fast. In this, google is not my friend, as results are sparse & not very helpful - i.e. if someone's found resolution, it has not been posted. Either there's a bug in the docco's or the code/scripts. Please help - I'd *really* like to give this a go Kind regards -Jaco </body> </html> --------------040901050807070502000602--

3 3

[Users] Upgrading to Fedora 18
by Kevin Daly 17 Feb '13

17 Feb '13

I have installed Ovirt 3.1 on a Fedora 17 server, and would like to upgrade to Fedora 18... The documentation I have found on the mailing list is unclear on how to do this. Does anyone have any pointers on how this can be done.. as well as upgrading to Ovirt 3.2 after we've upgraded to Fedora 18 Thanks In advance

3 2

[Users] Failed to migrate host
by Christian Hernandez 17 Feb '13

17 Feb '13

Hello, I got an error when trying to migrate a VM. On the "events" tab on the bottom it said. Failed to MigrateStatusVDS, error = Could not connect to peer VDS (VM: <vmname>, Source Host: <Hypervisor Host>). In the engine log it says the following. 2013-02-05 22:09:24,606 ERROR [org.ovirt.engine.core.vdsbroker.VdsUpdateRunTimeInfo] (QuartzScheduler_Worker-66) Rerun vm 6a9178cc-56cf-42f3-8a4c-098095d737f9. Called from vds VR2 2013-02-05 22:09:24,611 INFO [org.ovirt.engine.core.vdsbroker.vdsbroker.MigrateStatusVDSCommand] (pool-3-thread-48) START, MigrateStatusVDSCommand(HostName = VR2, HostId = 153b2742-6f2f-11e2-a248-52540038c812, vmId=6a9178cc-56cf-42f3-8a4c-098095d737f9), log id: 38675c5e 2013-02-05 22:09:24,656 ERROR [org.ovirt.engine.core.vdsbroker.vdsbroker.BrokerCommandBase] (pool-3-thread-48) Failed in MigrateStatusVDS method 2013-02-05 22:09:24,657 ERROR [org.ovirt.engine.core.vdsbroker.vdsbroker.BrokerCommandBase] (pool-3-thread-48) Error code noConPeer and error message VDSGenericException: VDSErrorException: Failed to MigrateStatusVDS, error = Could not connect to peer VDS 2013-02-05 22:09:24,662 INFO [org.ovirt.engine.core.vdsbroker.vdsbroker.BrokerCommandBase] (pool-3-thread-48) Command org.ovirt.engine.core.vdsbroker.vdsbroker.MigrateStatusVDSCommand return value Class Name: org.ovirt.engine.core.vdsbroker.vdsbroker.StatusOnlyReturnForXmlRpc mStatus Class Name: org.ovirt.engine.core.vdsbroker.vdsbroker.StatusForXmlRpc mCode 10 mMessage Could not connect to peer VDS 2013-02-05 22:09:24,663 INFO [org.ovirt.engine.core.vdsbroker.vdsbroker.BrokerCommandBase] (pool-3-thread-48) HostName = VR2 2013-02-05 22:09:24,664 ERROR [org.ovirt.engine.core.vdsbroker.VDSCommandBase] (pool-3-thread-48) Command MigrateStatusVDS execution failed. Exception: VDSErrorException: VDSGenericException: VDSErrorException: Failed to MigrateStatusVDS, error = Could not connect to peer VDS 2013-02-05 22:09:24,664 INFO [org.ovirt.engine.core.vdsbroker.vdsbroker.MigrateStatusVDSCommand] (pool-3-thread-48) FINISH, MigrateStatusVDSCommand, log id: 38675c5e

2 2