- Devel - oVirt List Archives

VM won't start because of -incoming flag after upgrading to alpha-2
by Christopher Pereira 08 Jul '15

08 Jul '15

This is a multi-part message in MIME format. --------------040500060504070207030504 Content-Type: text/plain; charset=utf-8; format=flowed Content-Transfer-Encoding: 7bit Hi, After upgrading from alpha-1 to alpha-2 and restarting services, a VM won't start because of the following qemu error: Domain id=5 is tainted: hook-script 2015-07-07T07:39:27.215579Z qemu-kvm: Unknown migration flags: 0 qemu: warning: error while loading state section id 2 2015-07-07T07:39:27.215681Z qemu-kvm: load of migration failed: Invalid argument The reason is that Engine was starting the VM with the -incoming flag (for "incoming migrations"). 1) Why did engine enable the -incoming flag? VM was configured with "Allow manual migration only". Maybe this ocurred because I set the host in maintenance mode before upgrading. 2) How can I clean the -incoming flag in oVirt? 3) Note that the "domain is tainted" error was also freezing the monitor socket between libvirt and vdsm, requiring a libvirtd + vdsmd restart. --------------040500060504070207030504 Content-Type: text/html; charset=utf-8 Content-Transfer-Encoding: 7bit <html> <head> <meta http-equiv="content-type" content="text/html; charset=utf-8"> </head> <body bgcolor="#FFFFFF" text="#000000"> Hi, After upgrading from alpha-1 to alpha-2 and restarting services, a VM won't start because of the following qemu error: <blockquote>Domain id=5 is tainted: hook-script 2015-07-07T07:39:27.215579Z qemu-kvm: Unknown migration flags: 0 qemu: warning: error while loading state section id 2 2015-07-07T07:39:27.215681Z qemu-kvm: load of migration failed: Invalid argument </blockquote> The reason is that Engine was starting the VM with the -incoming flag (for "incoming migrations"). 1) Why did engine enable the -incoming flag? VM was configured with "Allow manual migration only". Maybe this ocurred because I set the host in maintenance mode before upgrading. 2) How can I clean the -incoming flag in oVirt? 3) Note that the "domain is tainted" error was also freezing the monitor socket between libvirt and vdsm, requiring a libvirtd + vdsmd restart. </body> </html> --------------040500060504070207030504--

5 14

oVirt Node Weekly Meeting Minutes - 2015-07-06
by Fabian Deutsch 06 Jul '15

06 Jul '15

================================= #ovirt: oVirt Node Weekly Meeting ================================= Meeting started by fabiand at 14:04:04 UTC. The full logs are available at http://ovirt.org/meetings/ovirt/2015/ovirt.2015-07-06-14.04.log.html . Meeting summary --------------- * Agenda (fabiand, 14:06:04) * Jenkins status (fabiand, 14:06:12) * Other Items (fabiand, 14:06:16) * Jenkins status (fabiand, 14:06:21) * we now have all of our jobs (ovirt-node and the plugins) yamlized (fabiand, 14:08:07) * ACTION: tlitovsk to let us know the status of the publishers (fabiand, 14:10:00) * All master publishers are done (fabiand, 14:11:42) * ACTION: tlitovsk to modify the builds to let the rpms include the git hashes (fabiand, 14:13:57) * ACTION: rbarry to modify the jobs to extract the rpm manifest from the iso (fabiand, 14:14:16) Meeting ended at 14:16:47 UTC. Action Items ------------ * tlitovsk to let us know the status of the publishers * tlitovsk to modify the builds to let the rpms include the git hashes * rbarry to modify the jobs to extract the rpm manifest from the iso Action Items, by person ----------------------- * rbarry * rbarry to modify the jobs to extract the rpm manifest from the iso * tlitovsk * tlitovsk to let us know the status of the publishers * tlitovsk to modify the builds to let the rpms include the git hashes * **UNASSIGNED** * (none) People Present (lines said) --------------------------- * fabiand (47) * tlitovsk (7) * rbarry (5) * dougsland (2) * ovirtbot (2) Generated by `MeetBot`_ 0.1.4 .. _`MeetBot`: http://wiki.debian.org/MeetBot

1 0

Shared gluster storage domain for templates
by Christopher Pereira 04 Jul '15

04 Jul '15

Hi, I'm trying to setup a gluster data storage domain exclusively for templates (read only) and share it among different geo distributed ovirt data centers. I believe this is a common requirement. Do you know if there is a clean/efficient way to do this? My experiment: I geo-replicated the templates storage domain in order to keep a local copy on each data center for faster new VM instantiation or thin template provisioning. The problem I faced is it's not possible to import a replicated storage domain when the original Storage Domain UUID is already registered in the same Engine installation. A workarround was to detach the original SD (from Data Center 1) and attach the replicated SD (on Data Center 2)...or use another Engine installation. This works, but still requires to detach and reattach the storage domain. Is there anyway to copy a storage domain or change its UUID? I guess I can change the Storage Domain UUID (dirname and dom_md/metadata), but I'm afraid of the multiple references inside the binary file 'dom_md/ids'. BTW, using a gluster replica-n volume seems to be the cleanest solution because this way Engine would only be showing one single templates storage domain (ideal). The only reason I didn't go this way is because it is not designed between geo distributed data networks.

2 1

[AUTOMATION BLOCKER] vdsm 3.5 build failing on Fedora 20
by Sandro Bonazzola 03 Jul '15

03 Jul '15

Hi, vdsm build is currently failing on Fedora 20: /usr/bin/pep8 --exclude="${exclude}" --filename '*.py,*.py.in' . ./tests/functional/networkTests.py:933:56: E127 continuation line over-indented for visual indent ./tests/functional/networkTests.py:1007:56: E127 continuation line over-indented for visual indent ./tests/functional/networkTests.py:1008:55: E127 continuation line over-indented for visual indent make[2]: Leaving directory `/builddir/build/BUILD/vdsm-4.16.21' make[2]: *** [check-local] Error 1 make[1]: Leaving directory `/builddir/build/BUILD/vdsm-4.16.21' make[1]: *** [check-am] Error 2 make: *** [check-recursive] Error 1 RPM build errors: error: Bad exit status from /var/tmp/rpm-tmp.wl7CZV (%check) Bad exit status from /var/tmp/rpm-tmp.wl7CZV (%check) Child return code was: 1 Please fix ASAP. -- Sandro Bonazzola Better technology. Faster innovation. Powered by community collaboration. See how it works at redhat.com

4 4

trouble with already existing mac address in database
by Sven Kieske 03 Jul '15

03 Jul '15

-----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 Hi, jhernand was so kind to resolve this for me and asked for a mail to devel, because some people might be interested in it. so here it is: situation: you removed a vm somewhat dirty from ovirt by deleting it in the database, now you want to reuse it's mac address, but ovirt won't let you do it, because "Mac address is already in use." At first I thought I had not deleted all entries regarding this vm, but even when I did a dump of the whole engine db and grepped for the mac it didn't yield any results, so the mac was not in the db anymore, yet ovirt complained. what helped, was a restart of the engine service, because ovirt seems to cache some values in ram, like macs. so in case you got trouble like this, just restart the service. I wonder if it might be worth an RFE, so engine checks from time to time if values in ram are also in the db and if not cleans the ram? thanks again to jhernand for the (somewhat obvious) solution! site note: ovirt-engine is a really fast restarting service, pretty cool. - -- Mit freundlichen Grüßen / Regards Sven Kieske Systemadministrator Mittwald CM Service GmbH & Co. KG Königsberger Straße 6 32339 Espelkamp T: +495772 293100 F: +495772 293333 https://www.mittwald.de Geschäftsführer: Robert Meyer St.Nr.: 331/5721/1033, USt-IdNr.: DE814773217, HRA 6640, AG Bad Oeynhaus en Komplementärin: Robert Meyer Verwaltungs GmbH, HRB 13260, AG Bad Oeynhausen -----BEGIN PGP SIGNATURE----- Version: GnuPG v2.0.22 (GNU/Linux) iQIcBAEBAgAGBQJVlmDNAAoJEMby9TMDAbQR29kQAMyjBZh7RBxMRqoCp5+66rzB oZIv//v4d/y6cDypbFGJlC4k7N5EMNydZ3odl54pbnuOY806YFvnotyh5TT7xalH LLAaxz9oAUSod70sqBZX8A45mMioRR/MERyPyozEZn+YJuFYjhsAVjH1y75BQnhN NBqwpvmmxCMFBhLKpY4NK/oklFKZU8lDW0fuQr4oRGC0ngF0l13+QoC4OwNeYZlo OhECWP9yTIBpYJzSDTeaZCFHZ9bWG9ggL4ahwBbZ7w1AoB7gWAihV3kwlDIXGkql kgp/6xjPmqHrpeyB1dBb7bUzfLpuuN0lzUJAKcpvPU9p+n3ghq8zHMyJJOgMj6CX LpV7rvI5zBPnKbTg3HfFJUlFeb3rrasgJleW8FdDqUiuVpJ8DZr9vzx1HM6voNHW SQj/ZEKe5NNrHJW16djAhkemeT0/htsGF53xrWNPbeHiGky887ivhxhQ8DQ0B/4w Dcv/n4qKPJKCIUeQqGfXkkPpyn8hELysKpY1PIhjM/PXwEWWBFk0oGoNHt7enDaG VGuREioCfCasq4SFgbbw51+B0aQdQ1vYtt0rL8HNhVfl/RTlsZdSV474Dc8Or+Nb 3uxYo1kQrqZtYFCV/UchBzrAsqQLzq288XRFYPpoQAGyiSFSF+iMvwX1VEbiu1GF +90PfHwAV1ZDdjQ3sGd7 =spTC -----END PGP SIGNATURE-----

1 0

[AUTOMATION BLOCKER] ovirt-engine master - DAO tests failing on testSaveDetails(org.ovirt.engine.core.dao.gluster.GlusterGeoRepDaoTest)
by Sandro Bonazzola 03 Jul '15

03 Jul '15

Hi, http://jenkins.ovirt.org/job/ovirt-engine_master_dao-unit-tests_created/115… is currently failing on: 06:45:05 Failed tests: 06:45:05 testSaveDetails(org.ovirt.engine.core.dao.gluster.GlusterGeoRepDaoTest): expected:<org.ovirt.engine.core.common.businessentities.gluster.GlusterGeoRepSessionDetails@93d8f3f> but was:<org.ovirt.engine.core.common.businessentities.gluster.GlusterGeoRepSessionDetails@ea40d93f> 06:45:05 06:45:05 Tests run: 1571, Failures: 1, Errors: 0, Skipped: 0 Please fix ASAP -- Sandro Bonazzola Better technology. Faster innovation. Powered by community collaboration. See how it works at redhat.com

1 0

How to create FreeIPA user for ovirt engine (engine-manage-domains)?
by David Jaša 02 Jul '15

02 Jul '15

Hi, Pretty much any documentation around oVirt use of domains uses an undefined user (engine-manage-domains ... --user=[USER]) and maybe because of that, virtually all the ovirt tutorials that feature FreeIPA/IdM use "admin" user of FreeIPA (engine-manage-domains ... --provider=freeipa --user=admin). This leads to pretty scary situation of administrator password for your identity management system being stored for use by another system (ovirt-engine). So, the right way to do things should be use of a "service user" for engine that would have just enough privileges in FreeIPA to work correctly. So my questions are: 1. what are the necessary permissions for such a service user? 2. how to create such an user? Can it be done throught IPA web UI or does one need to go through the ldif/ldapmodify route? Best regards, David

2 1

[ANN] oVirt 3.5.4 First Release Candidate is now available for testing
by Sandro Bonazzola 01 Jul '15

01 Jul '15

The oVirt team is pleased to announce that the 3.5.4 First Release Candidate is now available for testing as of July 1st 2015. The release candidate is available now for Fedora 20, Red Hat Enterprise Linux 6.6, CentOS Linux 6.6 (or similar) and Red Hat Enterprise Linux 7.1, CentOS Linux 7.1 (or similar). This release of oVirt includes numerous bug fixes. See the release notes [1] for a list of the new features and bugs fixed. Please refer to release notes [1] for Installation / Upgrade instructions. A new oVirt Live ISO will be available as well[2]. Please note that mirrors[3] may need usually one day before being synchronized. Please refer to the release notes for known issues in this release. [1] http://www.ovirt.org/OVirt_3.5.4_Release_Notes [2] http://resources.ovirt.org/pub/ovirt-3.5-pre/iso/ [3] http://www.ovirt.org/Repository_mirrors#Current_mirrors -- Sandro Bonazzola Better technology. Faster innovation. Powered by community collaboration. See how it works at redhat.com

1 0

ovirt-engine-3.5.4 branch created
by Sandro Bonazzola 30 Jun '15

30 Jun '15

Hi, just a quick note reminding that ovirt-engine 3.5.4 branch has been created today. Any further patch targeted 3.5.4 must be cherry-picked there too. Thanks, -- Sandro Bonazzola Better technology. Faster innovation. Powered by community collaboration. See how it works at redhat.com

2 2

[ANN] oVirt 3.6.0 Second Alpha Release is now available for testing
by Sandro Bonazzola 30 Jun '15

30 Jun '15

The oVirt Project is pleased to announce the availability of the Second Alpha release of oVirt 3.6 for testing, as of June 29, 2015. oVirt is an open source alternative to VMware vSphere, and provides an excellent KVM management interface for multi-node virtualization. The Second Alpha release is available now for Fedora 20, Red Hat Enterprise Linux 6.6, CentOS 6.6 (or similar) and Red Hat Enterprise Linux 7.1, CentOS 7.1 (or similar). The Second Alpha release can also be deployed on Hypervisor Hosts running Fedora 21 and Fedora 22. Experimental support for Debian 8.1 Jessie has been added too. This Second Alpha release of oVirt includes numerous bug fixes. See the release notes [1] for an initial list of the new features and bugs fixed. Please refer to release notes [1] for Installation / Upgrade instructions. A new oVirt Live ISO[2] and Node ISO[3] are already available. Please note that mirrors[4] may need usually one day before being synchronized. Please refer to the release notes for known issues in this release. [1] http://www.ovirt.org/OVirt_3.6_Release_Notes [2] http://resources.ovirt.org/pub/ovirt-3.6-pre/iso/ovirt-live/el7-alpha2/ovir… [3] http://plain.resources.ovirt.org/pub/ovirt-3.6-pre/iso/ovirt-node/ovirt-nod… [4] http://www.ovirt.org/Repository_mirrors#Current_mirrors -- Sandro Bonazzola Better technology. Faster innovation. Powered by community collaboration. See how it works at redhat.com

1 2