Re: [one-users] Official Debian packages
Jaime Melis jme...@opennebula.org writes: Hmm, writing the question seems to gives my answer, it looks like the cherry-picking of master commits on the one-version branches are the reasons of the merge conflicts. Yes... that's the way we usually do it: by cherry picking the commits. Ok, I'll broke everything to rebase on master then. Unfortunately we don't think there's an easier way. Personally, I prefer to use DaggyFix[1][2] over cherry-picking. This requires to base the commits on the oldest supported release and then merge them everywhere needed, which can be seen as a lot of work, but not that much different from doing cherry-picking ;-) I think DaggyFix is more GIT friendly and respect the context as explain in an article I just found[3]. For example, here is the GIT graph to apply a foo-bar patch (issue #42): o-o-o-o-o-o-o-o-o-o-o-o-o-o-o-o-M₃ - master \ \ \ \ / \ \ \ o-o-o-o-M₂ /- prod/4/master (integration branch for version 4) \ \ \ / / \ \ \ o-o---+ - prod/3/42-foo-bar (short living branch) \ \ \ / \ \ \ o-o-o-o-M₁ - prod/3/master (integration branch for version 3) \ \ \ o-o-o-o-o-o-o-M₀ - prod/2/master (integration branch for version 2) \ \ / \ o-o-o-o - prod/2/41-fix-something \ o-o-o-o-o-o - prod/1/master (integration branch for version 1) in this example: - version 1 is not supported anymore - version 2 is supported but not concerned by #42: foo-bar A temporary branch is created based on version 3 named prod/3/42-foo-bar and the commits are done here. I use do to prefix my branch names by the name of the release branch it first apply to, it help to sort them when doing “git branch” ;-) Then, this 42/foo-bar branch is merge back into version 3 integration branch (commit M₁), to prepare a new release 3.X. This 42/foo-bar branch is merge into version 4 integration branch (commit M₂) if it apply to it like in my example. Finally, this 42/foo-bar branch is merge into the main development branch (commit M₃) if it apply to it like in my example. This is based on the “successful Git branching model[4]” I extended with the packaging branches. I hope it will help you, it took me time to find some kind of best practice I agree with, like “Is it better to rebase or merge?[5]”. Regards. Footnotes: [1] by mercurial http://mercurial.selenic.com/wiki/DaggyFixes [2] by monotone http://wiki.monotone.ca/DaggyFixes/ [3] https://queue.acm.org/detail.cfm?id=1595636 [4] http://nvie.com/posts/a-successful-git-branching-model/ [5] http://www.randyfay.com/node/91 -- Daniel Dehennin Récupérer ma clef GPG: gpg --recv-keys 0xCC1E9E5B7A6FE2DF Fingerprint: 3E69 014E 5C23 50E8 9ED6 2AAD CC1E 9E5B 7A6F E2DF signature.asc Description: PGP signature ___ Users mailing list Users@lists.opennebula.org http://lists.opennebula.org/listinfo.cgi/users-opennebula.org
Re: [one-users] Official Debian packages
Jaime Melis jmelis@opennebula.systems writes: Hi Alberto, I can confirm what Daniel said. We have no current plans of updating the offical Debian package of OpenNebula. Debian packaing guidelines are very strict and as Daniel mentioned it's a very big job. Daniel, I apologize, but somehow I was not aware of #3129. I will update the ticket now. Thanks, I have one issue with OpenNebula branches. I based my work on the 4.10 package[1] to compare the original package content with mine. Now, I would like to follow master and derive the packaging based on OpenNebula version, but I have a lot of conflicts when I merge master into my packaging branch. The base idea is to have one packaging branch following OpenNebula master branch. On a new release, a “release dedicated” packaging branch is derived from the master packaging branch to follow the OpenNebula release. Should I rebase[2] all my branches on master directly and then follow the OpenNebula releases or is there a path to merge master and one-4.10? Hmm, writing the question seems to gives my answer, it looks like the cherry-picking of master commits on the one-version branches are the reasons of the merge conflicts. Any hints? Footnotes: [1] http://downloads.opennebula.org/packages/opennebula-4.10.1/ [2] so changing history -- Daniel Dehennin Récupérer ma clef GPG: gpg --recv-keys 0xCC1E9E5B7A6FE2DF Fingerprint: 3E69 014E 5C23 50E8 9ED6 2AAD CC1E 9E5B 7A6F E2DF signature.asc Description: PGP signature ___ Users mailing list Users@lists.opennebula.org http://lists.opennebula.org/listinfo.cgi/users-opennebula.org
Re: [one-users] Official Debian packages
Alberto Zuin - Liste li...@albertozuin.eu writes: Hello all, Hello, just for information, I know there is a pre-compiled version of OpenNebula in your repository, but in the official Debian repository there is only an old version of OpenNebula for Wheezy (3.4) and only the contextualization package for the upcoming Jessie. There is any plan to have the package in the official Jessie repository? First, I'm neither from the OpenNebula team nor a Debian developper. I see no activity on the Debian repository[1]. I started[2] to rework the Debian packaging. It's a big work, for now lintian is far from being happy[3] and I don't know if the OpenNebula team is OK with it[4]. I do not even take the DFSG into account or the duplication of some components[5] with other Debian packages. So, I can't answer for others but it does not looks like it planed to make the fast moving OpenNebula fit the distribution releases. Regards. Footnotes: [1] http://anonscm.debian.org/cgit/pkg-opennebula/opennebula.git [2] https://github.com/baby-gnu/one/tree/pkg/debian/master [3] c.f. attachement [4] http://dev.opennebula.org/issues/3129#note-5 [5] https://bugs.debian.org/774114 -- Daniel Dehennin Récupérer ma clef GPG: gpg --recv-keys 0xCC1E9E5B7A6FE2DF Fingerprint: 3E69 014E 5C23 50E8 9ED6 2AAD CC1E 9E5B 7A6F E2DF signature.asc Description: PGP signature ___ Users mailing list Users@lists.opennebula.org http://lists.opennebula.org/listinfo.cgi/users-opennebula.org
Re: [one-users] Official Debian packages
Daniel Dehennin daniel.dehen...@baby-gnu.org writes: [...] It's a big work, for now lintian is far from being happy[3] [...] [3] c.f. attachement Missing attachement, sorry. -- Daniel Dehennin Récupérer ma clef GPG: gpg --recv-keys 0xCC1E9E5B7A6FE2DF Fingerprint: 3E69 014E 5C23 50E8 9ED6 2AAD CC1E 9E5B 7A6F E2DF P: opennebula source: source-contains-prebuilt-java-object src/oca/java/lib/xmlrpc-common-3.1.2.jar N: N:The source tarball contains a prebuilt Java class file. These are often N:included by mistake when developers generate a tarball without cleaning N:the source directory first. If there is no sign this was intended, N:consider reporting it as an upstream bug. N: N:Severity: pedantic, Certainty: possible N: N:Check: cruft, Type: source N: P: opennebula source: source-contains-prebuilt-java-object src/oca/java/lib/xmlrpc-client-3.1.2.jar P: opennebula source: source-contains-prebuilt-java-object src/oca/java/lib/ws-commons-util-1.0.2.jar P: opennebula source: source-contains-prebuilt-javascript-object src/sunstone/public/vendor/crypto-js/sha1-min.js N: N:The source tarball contains a prebuilt (minified) JavaScript object. N:They are usually left by mistake when generating the tarball by not N:cleaning the source directory first. You may want to report this as an N:upstream bug, in case there is no sign that this was intended. N: N:Severity: pedantic, Certainty: possible N: N:Check: cruft, Type: source N: E: opennebula source: source-is-missing src/sunstone/public/vendor/crypto-js/sha1-min.js N: N:The source of the following file is missing. Lintian checked a few N:possible paths to find the source, and do not find it. N: N:Please repack your package to include the source or add it to N:debian/missing-sources directory. N: N:If this is a false-positive, please report a bug against Lintian. N: N:Severity: serious, Certainty: possible N: N:Check: cruft, Type: source N: P: opennebula source: source-contains-prebuilt-javascript-object src/sunstone/public/vendor/crypto-js/enc-base64-min.js E: opennebula source: source-is-missing src/sunstone/public/vendor/crypto-js/enc-base64-min.js P: opennebula source: source-contains-prebuilt-javascript-object src/sunstone/public/vendor/crypto-js/core-min.js E: opennebula source: source-is-missing src/sunstone/public/vendor/crypto-js/core-min.js P: opennebula source: source-contains-prebuilt-javascript-object src/sunstone/public/bower_components/jquery-migrate/jquery-migrate.min.js P: opennebula source: source-contains-prebuilt-javascript-object src/sunstone/public/bower_components/jgrowl/jquery.jgrowl.min.js P: opennebula source: source-contains-prebuilt-javascript-object src/sunstone/public/bower_components/flot/excanvas.min.js P: opennebula source: source-contains-prebuilt-javascript-object src/sunstone/public/vendor/4.0/nouislider/jquery.nouislider.min.js E: opennebula source: source-is-missing src/sunstone/public/vendor/4.0/nouislider/jquery.nouislider.min.js P: opennebula source: source-contains-prebuilt-javascript-object src/sunstone/public/bower_components/sizzle/dist/sizzle.min.js P: opennebula source: source-contains-prebuilt-javascript-object src/sunstone/public/bower_components/no-vnc/include/logo.js mean line length is about 16184 characters E: opennebula source: source-is-missing src/sunstone/public/bower_components/no-vnc/include/logo.js P: opennebula source: source-contains-prebuilt-javascript-object src/sunstone/public/bower_components/no-vnc/include/keysymdef.js mean line length is about 3983 characters E: opennebula source: source-is-missing src/sunstone/public/bower_components/no-vnc/include/keysymdef.js P: opennebula source: source-contains-prebuilt-javascript-object src/sunstone/public/bower_components/jquery/dist/jquery.min.js P: opennebula source: source-contains-prebuilt-javascript-object src/sunstone/public/bower_components/foundation/js/foundation.min.js P: opennebula source: source-contains-prebuilt-javascript-object src/sunstone/public/bower_components/flot.tooltip/js/jquery.flot.tooltip.min.js P: opennebula source: source-contains-prebuilt-javascript-object src/sunstone/public/bower_components/flot.tooltip/js/excanvas.min.js E: opennebula source: source-is-missing src/sunstone/public/bower_components/flot.tooltip/js/excanvas.min.js P: opennebula source: source-contains-prebuilt-javascript-object src/sunstone/public/bower_components/no-vnc/include/web-socket-js/swfobject.js mean line length is about 10071 characters E: opennebula source: source-is-missing src/sunstone/public/bower_components/no-vnc/include/web-socket-js/swfobject.js P: opennebula source: source-contains-prebuilt-flash-object src/sunstone/public/bower_components/no-vnc/include/web-socket-js/WebSocketMain.swf N: N:The source tarball contains a prebuilt file in the Shockwave Flash (SWF) N:or Flash Video (FLV
Re: [one-users] Getting VNC token from Sunstone REST API
Daniel Dehennin daniel.dehen...@baby-gnu.org writes: We use jenkins to automate the installation of VM from ISO images[1] and we use something similar to jenkins.debian.net[2] which use vncdo[3]. [...] We would like to get the token from Sunstone REST API: https://nebula.example.net/vm/VMID/startvnc but we don't see how to manage the authentication? Hello, It looks like I need the csrftoken but I don't know how to retrieve it. Any hint? Regards. -- Daniel Dehennin Récupérer ma clef GPG: gpg --recv-keys 0xCC1E9E5B7A6FE2DF Fingerprint: 3E69 014E 5C23 50E8 9ED6 2AAD CC1E 9E5B 7A6F E2DF signature.asc Description: PGP signature ___ Users mailing list Users@lists.opennebula.org http://lists.opennebula.org/listinfo.cgi/users-opennebula.org
Re: [one-users] @installation problem
anagha b banag...@gmail.com writes: Hi, Hello, I am trying to intall opennebula-3.8.3 and got following error [...] I think the first thing the OpenNebula team will tell you is that 3.8 is old and you should get the 4.10.2[1]. Regards. Footnotes: [1] http://opennebula.org/software/ -- Daniel Dehennin Récupérer ma clef GPG: gpg --recv-keys 0xCC1E9E5B7A6FE2DF Fingerprint: 3E69 014E 5C23 50E8 9ED6 2AAD CC1E 9E5B 7A6F E2DF signature.asc Description: PGP signature ___ Users mailing list Users@lists.opennebula.org http://lists.opennebula.org/listinfo.cgi/users-opennebula.org
[one-users] Getting VNC token from Sunstone REST API
Hello, We use jenkins to automate the installation of VM from ISO images[1] and we use something similar to jenkins.debian.net[2] which use vncdo[3]. We retrieve the host where the VM is running and the VNC port from the XML RPC API and open a direct connection to the qemu VNC server. With our first “developer only” cluster, we did not have any issue since the cluster nodes are on the same network than our workstations. With our “soon in production” cluster, the nodes are not on a reachable network and we need to connect through the WebSocket proxy. The VNC token is not available through the XML RPC API[4], which could be seen as normal because it's managed by Sunstone and shared with websocketproxy on a possibly different machine than the OpenNebula daemon. We would like to get the token from Sunstone REST API: https://nebula.example.net/vm/VMID/startvnc but we don't see how to manage the authentication? Does someone manage to retrieve the token pragmatically? Regards. Footnotes: [1] to test the installation of our french distribution for schools [2] https://jenkins.debian.net [3] http://anonscm.debian.org/cgit/qa/jenkins.debian.net.git/tree/bin/g-i-installation.sh#n402 [4] http://dev.opennebula.org/issues/2655 -- Daniel Dehennin Récupérer ma clef GPG: gpg --recv-keys 0xCC1E9E5B7A6FE2DF Fingerprint: 3E69 014E 5C23 50E8 9ED6 2AAD CC1E 9E5B 7A6F E2DF signature.asc Description: PGP signature ___ Users mailing list Users@lists.opennebula.org http://lists.opennebula.org/listinfo.cgi/users-opennebula.org
Re: [one-users] Source image stuck in state CLONE and target image stuck in state LOCKED
Gerry O'Brien ge...@scss.tcd.ie writes: Hi, Hello, The oned (4.6.2) daemon was restarted during a clone and now the source image is stuck in state CLONE and target image is stuck in state LOCKED. The actual Linux copy completed so both images are OK. How to I reset them to the READY state? Is there an oneimage command or does the database have to be modified. If the database has to be modified are there maintenance commands to do this? I got this one time, I tried to disable then enable the images and then everything was OK. Regards. -- Daniel Dehennin Récupérer ma clef GPG: gpg --recv-keys 0xCC1E9E5B7A6FE2DF Fingerprint: 3E69 014E 5C23 50E8 9ED6 2AAD CC1E 9E5B 7A6F E2DF signature.asc Description: PGP signature ___ Users mailing list Users@lists.opennebula.org http://lists.opennebula.org/listinfo.cgi/users-opennebula.org
[one-users] Dynamic attach CD to VM
Hello, I'm trying to attach a CD to a running VM without success. I saw that IDE hotplug does not work. I found that we can predefine a disk/ without source[1] but it does not seems to work on 4.8. How are you doing to attach CD on running VM? Regards. Footnotes: [1] http://dev.opennebula.org/issues/304 -- Daniel Dehennin Récupérer ma clef GPG: gpg --recv-keys 0xCC1E9E5B7A6FE2DF Fingerprint: 3E69 014E 5C23 50E8 9ED6 2AAD CC1E 9E5B 7A6F E2DF signature.asc Description: PGP signature ___ Users mailing list Users@lists.opennebula.org http://lists.opennebula.org/listinfo.cgi/users-opennebula.org
Re: [one-users] Sharing VIP between VMs
Dmitri Chebotarov dcheb...@gmu.edu writes: Hi, Hello, Does anyone run pacemaker/corosync cluster in VMs hosted on Opennebula? Is it possible to share VIP? With ARP cache poisoning feature enabled (http://dev.opennebula.org/issues/2318) it's not possible to add aliased VIPs to the cluster. Unless I manually run 'ovs-ofctl add-flow br0 in_port=305,arp,dl_src=MAC,nw_src=VIP...' on the host. The ARP poisoning block can be globally disabled in /var/lib/one/remotes/vnm/OpenNebulaNetwork.conf, but is it possible to selectively disable it for a specific VMs only. There is a feature request to add multiple MAC or IP addresses[1] and another one to override configuration per networks[2]. Making it configurable per VM, so configurable by users, may be a little risky. I prefer the solution of sharing reserved addresses. Regards. Footnotes: [1] http://dev.opennebula.org/issues/2348 [2] http://dev.opennebula.org/issues/3387 -- Daniel Dehennin Récupérer ma clef GPG: gpg --recv-keys 0xCC1E9E5B7A6FE2DF Fingerprint: 3E69 014E 5C23 50E8 9ED6 2AAD CC1E 9E5B 7A6F E2DF signature.asc Description: PGP signature ___ Users mailing list Users@lists.opennebula.org http://lists.opennebula.org/listinfo.cgi/users-opennebula.org
Re: [one-users] Question on /var/lib/remotes/tm/shared/clone script
Steven C Timm t...@fnal.gov writes: [root@cloudworker1359 libvirt]# grep -v ^# qemu.conf | grep . dynamic_ownership = 0 You must set “user” and “group” to “oneadmin” like described in the documentation[1]. This way the qemu process run as oneadmin and has write access to the images. Regards. Footnotes: [1] http://docs.opennebula.org/4.10/design_and_installation/quick_starts/qs_ubuntu_kvm.html#configure-qemu -- Daniel Dehennin Récupérer ma clef GPG: gpg --recv-keys 0xCC1E9E5B7A6FE2DF Fingerprint: 3E69 014E 5C23 50E8 9ED6 2AAD CC1E 9E5B 7A6F E2DF signature.asc Description: PGP signature ___ Users mailing list Users@lists.opennebula.org http://lists.opennebula.org/listinfo.cgi/users-opennebula.org
Re: [one-users] Initial installation
Michael A Cooper nexusgu...@gmail.com writes: Hello Everyone, Hello, I found this link OpenNebula Ubuntu 14.04 Quickstart http://docs.opennebula.org/4.6/design_and_installation/quick_starts/qs_ubuntu_kvm.html to install open nebula on Ubuntu 14.04, Note that the current stable OpenNebula[1] is 4.10. This includes the frontend and how to add nodes. My question is according to this article, I can install OpenNebula frontend right on the Ubuntu Server with the KVM install, is this correct? Also is it possible to have just the frontend installed by itself (non-kvm server), then add nodes to it, I would prefer to it that way. If it is possible is there a step by step document for me to use to help me through the process? This article describes a multi-server setup: - the frontend which runs the management daemons, the opennebula web interface (sunstone) and a NFS server - the nodes which run libvirtd and use the NFS exported by the frontend In the NFS configuration[2], you have a warning about setting up what I call a « single node », i.e. everything on a single physical standalone machine. Some steps are already managed by the opennebula debian packages, like the Qemu configuration. The minimal requirement for a separated frontend is the ability for the user oneadmin on the frontend to connect over SSH on each nodes without passwords. If you want live migration, you must use shared storage[3][4]. More informations are available in “planning the installation”[5]. Regards. Footnotes: [1] http://docs.opennebula.org/stable/ [2] http://docs.opennebula.org/stable/design_and_installation/quick_starts/qs_ubuntu_kvm.html#configure-nfs [3] http://docs.opennebula.org/stable/administration/storage/index.html [4] personally I never share /var/lib/one as described in the quick start but /var/lib/one/datastores. [5] http://docs.opennebula.org/stable/design_and_installation/building_your_cloud/plan.html -- Daniel Dehennin Récupérer ma clef GPG: gpg --recv-keys 0xCC1E9E5B7A6FE2DF Fingerprint: 3E69 014E 5C23 50E8 9ED6 2AAD CC1E 9E5B 7A6F E2DF signature.asc Description: PGP signature ___ Users mailing list Users@lists.opennebula.org http://lists.opennebula.org/listinfo.cgi/users-opennebula.org
Re: [one-users] Issue with adding CDROM and boot it
Alexey Shalin c...@hoster.kg writes: Hello I set up a cluster of 3 nodes Opennebula and NFS. [...] /var/lib/one//datastores/0/38/disk.0 - points to iso file I do not know, why in deployment.0 having 2 cdroms, in the template only 1 disk added because of it .. i can not boot VM via CDROM The second CDROM is CONTEXT. We report an issue[1] to avoid adding TARGET=hda on boot ISO. Regards. Footnotes: [1] http://dev.opennebula.org/issues/3158 -- Daniel Dehennin Récupérer ma clef GPG: gpg --recv-keys 0xCC1E9E5B7A6FE2DF Fingerprint: 3E69 014E 5C23 50E8 9ED6 2AAD CC1E 9E5B 7A6F E2DF signature.asc Description: PGP signature ___ Users mailing list Users@lists.opennebula.org http://lists.opennebula.org/listinfo.cgi/users-opennebula.org
[one-users] Multiple probes running on a node
Hello, One of our ONE 4.8 node has tons of monitor_ds.sh probes running. All the “du” are marked in uninterruptible sleep 5903 ? Ss 0:00 \_ sshd: oneadmin [priv] 5905 ? S 0:00 \_ sshd: oneadmin@notty 5906 ? Ss 0:00 \_ bash -c if [ -x /var/tmp/one/im/run_probes ]; then /var/tmp/one/im/run_probes kvm /var/lib/one//datastores 4124 60 8 igor; else exit 42; fi 5907 ? S 0:00 \_ /bin/bash /var/tmp/one/im/run_probes kvm /var/lib/one//datastores 4124 60 8 igor 5914 ? S 0:00 \_ /bin/bash /var/tmp/one/im/run_probes kvm /var/lib/one//datastores 4124 60 8 igor 5915 ? S 0:00 \_ /bin/bash /var/tmp/one/im/run_probes kvm /var/lib/one//datastores 4124 60 8 igor 5920 ? S 0:00 \_ /bin/bash ./collectd-client_control.sh kvm /var/lib/one//datastores 4124 60 8 igor 5927 ? S 0:00 \_ /bin/bash /var/tmp/one/im/kvm.d/../run_probes kvm-probes /var/lib/one//datastores 4124 60 8 igor 5934 ? S 0:00 \_ /bin/bash /var/tmp/one/im/kvm.d/../run_probes kvm-probes /var/lib/one//datastores 4124 60 8 igor 5935 ? S 0:00 \_ /bin/bash /var/tmp/one/im/kvm.d/../run_probes kvm-probes /var/lib/one//datastores 4124 60 8 igor 5988 ? S 0:00 \_ /bin/bash ./monitor_ds.sh kvm-probes /var/lib/one//datastores 4124 60 8 igor 5991 ? S 0:00 \_ /bin/bash ./monitor_ds.sh kvm-probes /var/lib/one//datastores 4124 60 8 igor 5992 ? D 0:00 \_ du -sLm /var/lib/one//datastores 5993 ? S 0:00 \_ cut -f1 I think that instead of trying to run the probes again and again adding more contention on the disk, ONE should report the node as ERROR. Should I open a bug? Regards. -- Daniel Dehennin Récupérer ma clef GPG: gpg --recv-keys 0xCC1E9E5B7A6FE2DF Fingerprint: 3E69 014E 5C23 50E8 9ED6 2AAD CC1E 9E5B 7A6F E2DF signature.asc Description: PGP signature ___ Users mailing list Users@lists.opennebula.org http://lists.opennebula.org/listinfo.cgi/users-opennebula.org
Re: [one-users] Multiple probes running on a node
Jaime Melis j.me...@gmail.com writes: Absolutely, Ok, done[1]. and we will stop using du, hopefully in the next version. Great, it can be quite heavy, mostly over NFS. Regards. Footnotes: [1] http://dev.opennebula.org/issues/3393 -- Daniel Dehennin Récupérer ma clef GPG: gpg --recv-keys 0xCC1E9E5B7A6FE2DF Fingerprint: 3E69 014E 5C23 50E8 9ED6 2AAD CC1E 9E5B 7A6F E2DF signature.asc Description: PGP signature ___ Users mailing list Users@lists.opennebula.org http://lists.opennebula.org/listinfo.cgi/users-opennebula.org
Re: [one-users] OpenNebula lost ip addressess.
Hello, Leszek Master keks...@gmail.com writes: Today i've had lack of power, so long that my UPS batteries died, and so the servers. After rebooting everything i noticed that 0 of my VM's started automatically. So here's the first question, is there any way to manage autostart vm after infrastructure rebooting? For information, there is an opened issue for this: #1290[1]. The second thing is that when i started my vm's the ip addresses got messed up. How can it happen ? How to secure my infrastructure from changing ip addressed? I'll let the OpenNebula guys answer this part ;-) Regards. Footnotes: [1] http://dev.opennebula.org/issues/1290 -- Daniel Dehennin Récupérer ma clef GPG: gpg --recv-keys 0xCC1E9E5B7A6FE2DF Fingerprint: 3E69 014E 5C23 50E8 9ED6 2AAD CC1E 9E5B 7A6F E2DF signature.asc Description: PGP signature ___ Users mailing list Users@lists.opennebula.org http://lists.opennebula.org/listinfo.cgi/users-opennebula.org
Re: [one-users] Changing a onehost name within opennebula
Sergi s...@okitup.com writes: Hello, Hello, [...] I would like to change 'localhost' to its real uname name. Will it be seamless? Or are there implications I have to address before? I think this has impact on SSH authentication. If you change localhost to $(hostname -s) then you must accept the key for the new name. Regards. -- Daniel Dehennin Récupérer ma clef GPG: gpg --recv-keys 0xCC1E9E5B7A6FE2DF Fingerprint: 3E69 014E 5C23 50E8 9ED6 2AAD CC1E 9E5B 7A6F E2DF signature.asc Description: PGP signature ___ Users mailing list Users@lists.opennebula.org http://lists.opennebula.org/listinfo.cgi/users-opennebula.org
Re: [one-users] Private cloud with Front-end on Virtual MAchine HA
Bart b...@pleh.info writes: Hi Daniel, Hello, So basically, to sum it up, there is currently no way of running the OpenNebula management node (with all functionality inside one VM) on it's own virtualisation cluster (and thus managing itself along with the rest of the cluster). You can run the VM on the same hardware, but it will not be managed by OpenNebula, since you need OpenNebula to start OpenNebula VMs. That's why the documentation[1] explains the setup of an H-A system. If you have two physical servers, you may create two OpenNebula VMs, one master and one slave with replicated database between them. Regards. Footnotes: [1] http://docs.opennebula.org/4.10/advanced_administration/high_availability/oneha.html -- Daniel Dehennin Récupérer ma clef GPG: gpg --recv-keys 0xCC1E9E5B7A6FE2DF Fingerprint: 3E69 014E 5C23 50E8 9ED6 2AAD CC1E 9E5B 7A6F E2DF signature.asc Description: PGP signature ___ Users mailing list Users@lists.opennebula.org http://lists.opennebula.org/listinfo.cgi/users-opennebula.org
Re: [one-users] How do you change the '-vga cirrus' option on KVM
Gerry O'Brien ge...@scss.tcd.ie writes: Hi Daniel, Thanks for the info. In general, do you know what the mapping is between the XML 'model type' and the parameter that gets passed to KVM via the -vga switch? I found it in the libvirt documentation[1]. Regards. Footnotes: [1] http://libvirt.org/formatdomain.html#elementsVideo -- Daniel Dehennin Récupérer ma clef GPG: gpg --recv-keys 0xCC1E9E5B7A6FE2DF Fingerprint: 3E69 014E 5C23 50E8 9ED6 2AAD CC1E 9E5B 7A6F E2DF signature.asc Description: PGP signature ___ Users mailing list Users@lists.opennebula.org http://lists.opennebula.org/listinfo.cgi/users-opennebula.org
Re: [one-users] How do you change the '-vga cirrus' option on KVM
Gerry O'Brien ge...@scss.tcd.ie writes: Hi, Hello, How do you change the '-vga cirrus' option for KVM to something else, e.g. '-vga std'? Is it done via the RAW data in the template? That's how I do it: RAW=[TYPE=kvm,DATA=devicesvideomodel type=\vga\ heads=\1\/model/video/devices] It could be interesting to be able to select it in the template: 1. provide a default in /etc/one/vmm_exec/vmm_exec_kvm.conf (cirrus for example) 2. if graphics is “SPICE”, set the card to “qxl” in the template 2. user can override the card type for this template If you define the RAW DATA in /etc/one/vmm_exec/vmm_exec_kvm.conf, then you can not override it in the template because RAW are concatenated. Regards. -- Daniel Dehennin Récupérer ma clef GPG: gpg --recv-keys 0xCC1E9E5B7A6FE2DF Fingerprint: 3E69 014E 5C23 50E8 9ED6 2AAD CC1E 9E5B 7A6F E2DF signature.asc Description: PGP signature ___ Users mailing list Users@lists.opennebula.org http://lists.opennebula.org/listinfo.cgi/users-opennebula.org
[one-users] OpenVswitch and host crash
Hello, We had a power failure on a test ONE 4.8, the frontend and the two hosts crashed. I can not resume nor boot the powered off VMs, libvirt told me: error : virNetDevOpenvswitchRemovePort:187 : Unable to delete port (null) from OVS: Operation not permitted error : virCommandWait:2426 : internal error: Child process (ovs-vsctl --timeout=5 -- --may-exist add-port inetrnal vnet0 -- set Interface vnet0 'external-ids:attached-mac=02:00:c0:a8:00:6e' -- set Interface vnet0 'external-ids:iface-id=2a9321ad-2b7e-4ca5-9c81-a3f54d04807b' -- set Interface vnet0 'external-ids:vm-id=c486c111-e3fe-4338-99b0-8eed4c043cff' -- set Interface vnet0 external-ids:iface-status=active) unexpected exit status 1: ovs-vsctl: --may-exist add-port internal vnet0 but vnet0 is actually attached to bridge external To come back to normal, I had to run the following command: for switch in external internal do for port in $(ovs-vsctl list-ports $switch | grep '^vnet') do ovs-vsctl del-port $switch $port done done Any idea to perform such operation automatically after a recovery? Regards. -- Daniel Dehennin Récupérer ma clef GPG: gpg --recv-keys 0xCC1E9E5B7A6FE2DF Fingerprint: 3E69 014E 5C23 50E8 9ED6 2AAD CC1E 9E5B 7A6F E2DF signature.asc Description: PGP signature ___ Users mailing list Users@lists.opennebula.org http://lists.opennebula.org/listinfo.cgi/users-opennebula.org
[one-users] Extracting VM images
Hello, Looking at the marketplace documentation, I see that it does not store images. This means that I need to extract image files to put them on a public web server and reference them in the maketplace. Is there a way to extract them by other way than doing: 1. ssh one@nebula 2. IMG=$(oneimage show ID | awk -F':' '/^SOURCE/ { print $2}') 3. scp ${IMG} www-data@webserver:/somewhere/to/store/image/name.img Such workflow makes it impossible for simple users to extract their images from ONE and make them available for appliances. Is it possible to get some feedback on how you use maketplace with ONE? Regards. -- Daniel Dehennin Récupérer ma clef GPG: gpg --recv-keys 0xCC1E9E5B7A6FE2DF Fingerprint: 3E69 014E 5C23 50E8 9ED6 2AAD CC1E 9E5B 7A6F E2DF signature.asc Description: PGP signature ___ Users mailing list Users@lists.opennebula.org http://lists.opennebula.org/listinfo.cgi/users-opennebula.org
Re: [one-users] oneimage filter by path or source
Carlos Martín Sánchez cmar...@opennebula.org writes: [...] Actually, since the source is not a column, ticket 2372 would not apply. This would require a more general filter. Maxim, what would be your use case? If you need to do a one-time search, you can filter the data through the api [1] or using xpath: $ oneimage list -x | xpath -e '/IMAGE_POOL/IMAGE[PATH=/tmp/ttylinux.raw]/ID' +1 I used the xpath tool from libxml-xpath-perl to make several complicated XPath queries to find images not used by any templates. Finaly, I prefer this solution since it's more UNIX' spirit, piping several simple[1] commands. Regards. Footnotes: [1] In term of XPath, simple looks like a joke ;-) -- Daniel Dehennin Récupérer ma clef GPG: gpg --recv-keys 0xCC1E9E5B7A6FE2DF Fingerprint: 3E69 014E 5C23 50E8 9ED6 2AAD CC1E 9E5B 7A6F E2DF signature.asc Description: PGP signature ___ Users mailing list Users@lists.opennebula.org http://lists.opennebula.org/listinfo.cgi/users-opennebula.org
Re: [one-users] Multiple network cards with same MAC address
Ruben S. Montero rsmont...@opennebula.org writes: Hi Hello, MAC address is checked to not collide within a VNET - AR. If you want to set different MAC on different VNETs you need to define the MAC address ranges in such a way they do not overlap. Ok, thank you. -- Daniel Dehennin Récupérer ma clef GPG: gpg --recv-keys 0xCC1E9E5B7A6FE2DF Fingerprint: 3E69 014E 5C23 50E8 9ED6 2AAD CC1E 9E5B 7A6F E2DF signature.asc Description: PGP signature ___ Users mailing list Users@lists.opennebula.org http://lists.opennebula.org/listinfo.cgi/users-opennebula.org
Re: [one-users] Cloud view and available tabs
Hamada, Ondrej ondrej.ham...@acision.com writes: Hi, Hello, You should edit the file /etc/one/sunstone-views/cloud.yaml and restart sunstone. I already copy the user.yaml to enable tabs for a group, but the cloud.yaml is quite empty and I don't know what could be added. I'll try to add the vms-tab... Regards. -- Daniel Dehennin Récupérer ma clef GPG: gpg --recv-keys 0xCC1E9E5B7A6FE2DF Fingerprint: 3E69 014E 5C23 50E8 9ED6 2AAD CC1E 9E5B 7A6F E2DF signature.asc Description: PGP signature ___ Users mailing list Users@lists.opennebula.org http://lists.opennebula.org/listinfo.cgi/users-opennebula.org
Re: [one-users] oneimage filter by path or source
Maxim Terletskiy terlets...@emu.ru writes: Is it possible to list images with specific path or source? Good afternoon, This feature[1] is already registered in the backlog. Regards. Footnotes: [1] http://dev.opennebula.org/issues/2372 -- Daniel Dehennin Récupérer ma clef GPG: gpg --recv-keys 0xCC1E9E5B7A6FE2DF Fingerprint: 3E69 014E 5C23 50E8 9ED6 2AAD CC1E 9E5B 7A6F E2DF signature.asc Description: PGP signature ___ Users mailing list Users@lists.opennebula.org http://lists.opennebula.org/listinfo.cgi/users-opennebula.org
Re: [one-users] Sunstone Cloud View seems broken (Users Digest, Vol 79, Issue 36)
Rolandas Naujikas rolandas.nauji...@mif.vu.lt writes: Our users also confused about Create button. It should be invisible, when no template is selected. Also there is impossible to select a template, that have RAW attribute in it. I just create an issue[1] for this one. I was wondering since days why some template are working in cloud view and why other don't. Regards. Footnotes: [1] http://dev.opennebula.org/issues/3184 -- Daniel Dehennin Récupérer ma clef GPG: gpg --recv-keys 0xCC1E9E5B7A6FE2DF Fingerprint: 3E69 014E 5C23 50E8 9ED6 2AAD CC1E 9E5B 7A6F E2DF signature.asc Description: PGP signature ___ Users mailing list Users@lists.opennebula.org http://lists.opennebula.org/listinfo.cgi/users-opennebula.org
Re: [one-users] Sunstone Cloud View seems broken (Users Digest, Vol 79, Issue 36)
Daniel Dehennin daniel.dehen...@baby-gnu.org writes: Rolandas Naujikas rolandas.nauji...@mif.vu.lt writes: Our users also confused about Create button. It should be invisible, when no template is selected. Also there is impossible to select a template, that have RAW attribute in it. I just create an issue[1] for this one. I was wondering since days why some template are working in cloud view and why other don't. Should have checked the issue list[1], sorry for the noise. Footnotes: [1] http://dev.opennebula.org/issues/3154 -- Daniel Dehennin Récupérer ma clef GPG: gpg --recv-keys 0xCC1E9E5B7A6FE2DF Fingerprint: 3E69 014E 5C23 50E8 9ED6 2AAD CC1E 9E5B 7A6F E2DF signature.asc Description: PGP signature ___ Users mailing list Users@lists.opennebula.org http://lists.opennebula.org/listinfo.cgi/users-opennebula.org
[one-users] Multiple network cards with same MAC address
Hello, Using ONE 4.8 on two nodes with Open vSwitch networks, a crossover cable between the nodes as trunk. I have a router template using 4 different networks, each network is isolated in dedicated VLAN, but 3 network cards use the same MAC address. On another try two of them use the same MAC address. This result is miss-function on that router, the firt card is tagged with the VLAN ID two times, the next one is in trunk mode: Mon Sep 8 11:00:14 2014 [Z0][VMM][I]: post: Executed sudo ovs-vsctl set Port vnet44 tag=655. Mon Sep 8 11:00:14 2014 [Z0][VMM][I]: post: Executed sudo ovs-vsctl set Port vnet45 tag=659. Mon Sep 8 11:00:14 2014 [Z0][VMM][I]: post: Executed sudo ovs-vsctl set Port vnet45 tag=660. Mon Sep 8 11:00:14 2014 [Z0][VMM][I]: post: Executed sudo ovs-vsctl set Port vnet47 tag=661. Mon Sep 8 11:00:14 2014 [Z0][VMM][I]: ExitCode: 0 The “vnet45” should bet tagged with “659”: ovs-vsctl find Port name=vnet45 _uuid : b5b33023-03ff-4229-8039-f2268096fb77 bond_downdelay : 0 bond_fake_iface : false bond_mode : [] bond_updelay: 0 external_ids: {} fake_bridge : false interfaces : [349e0f2d-3cd7-4865-bb8d-72b82aebc437] lacp: [] mac : [] name: vnet45 other_config: {} qos : [] statistics : {} status : {} tag : 660 trunks : [] vlan_mode : [] The “vnet46” has the same MAC address as “vnet45” and is in trunk mode: ovs-vsctl find Port name=vnet46 _uuid : cad4a6c7-5ab5-440f-81e7-59bfb19e00a6 bond_downdelay : 0 bond_fake_iface : false bond_mode : [] bond_updelay: 0 external_ids: {} fake_bridge : false interfaces : [1b0a1deb-fdb3-44ea-b406-67afa953ee5f] lacp: [] mac : [] name: vnet46 other_config: {} qos : [] statistics : {} status : {} tag : [] trunks : [] vlan_mode : [] Here is one of the networks definitions: oneadmin@nebula:~$ onevnet show 659 VIRTUAL NETWORK 659 INFORMATION ID : 659 NAME : dmz1 USER : auser GROUP : users CLUSTER: - BRIDGE : internal VLAN : Yes USED LEASES: 2 PERMISSIONS OWNER : um- GROUP : --- OTHER : --- VIRTUAL NETWORK TEMPLATE BRIDGE=internal PHYDEV= PUBLIC=NO VLAN=YES VLAN_ID= ADDRESS RANGE POOL AR TYPESIZE LEASES MAC IP GLOBAL_PREFIX 0 ETHER254 1 02:00:2b:94:f7:fa - - LEASES AR OWNERMAC IP IP6_GLOBAL 0 VM : 6113 02:00:2b:94:f7:fa - - I could define random mac prefixes for each network but I assumed that ONE will pick different MAC for each cards. Regards. -- Daniel Dehennin Récupérer ma clef GPG: gpg --recv-keys 0xCC1E9E5B7A6FE2DF Fingerprint: 3E69 014E 5C23 50E8 9ED6 2AAD CC1E 9E5B 7A6F E2DF signature.asc Description: PGP signature ___ Users mailing list Users@lists.opennebula.org http://lists.opennebula.org/listinfo.cgi/users-opennebula.org
Re: [one-users] Thin provisioning and image size
Javier Fontan jfon...@opennebula.org writes: Which was the size of the original image? I think that when you do a save_as (deferred disk snapshot) it just copies the size of the original image to the new one. I started with empty qcow2 disk of several virtual sizes, but on disk they are all 196KB. Regards. -- Daniel Dehennin Récupérer ma clef GPG: gpg --recv-keys 0xCC1E9E5B7A6FE2DF Fingerprint: 3E69 014E 5C23 50E8 9ED6 2AAD CC1E 9E5B 7A6F E2DF ___ Users mailing list Users@lists.opennebula.org http://lists.opennebula.org/listinfo.cgi/users-opennebula.org
Re: [one-users] Can't loginto sunstone after upgrade to 4.8
Daniel Molina dmol...@opennebula.org writes: [...] You can just use this command to change the password oneadmin@nebula:~$ oneuser passwd serveradmin -r serveradmin.pwd --sha1 Ok, I did not understand why it was not working without the --sha1. Thanks. -- Daniel Dehennin Récupérer ma clef GPG: gpg --recv-keys 0xCC1E9E5B7A6FE2DF Fingerprint: 3E69 014E 5C23 50E8 9ED6 2AAD CC1E 9E5B 7A6F E2DF signature.asc Description: PGP signature ___ Users mailing list Users@lists.opennebula.org http://lists.opennebula.org/listinfo.cgi/users-opennebula.org
[one-users] Thin provisioning and image size
Hello, The size of my qcow2 images is always reported as 1M: IMAGE 114 INFORMATION ID : 114 NAME : debian-wheezy USER : nebula GROUP : nebula DATASTORE : default TYPE : OS REGISTER TIME : 04/10 10:18:26 PERSISTENT : No SOURCE : /var/lib/one/datastores/1/5d2be6f9746fc3447be11c6621da87df FSTYPE : save_as SIZE : 1M STATE : rdy RUNNING_VMS: 0 PERMISSIONS OWNER : um- GROUP : u-- OTHER : u-- IMAGE TEMPLATE DEV_PREFIX=vd DRIVER=qcow2 SAVED_DISK_ID=0 SAVED_IMAGE_ID=17 SAVED_VM_ID=347 SAVE_AS=YES As far as I understand, the size is calculated only at creation time and never updated. I think we could differentiate the real size from the virtual size in case of thin provisioning like qcow2, qemu-img actually reports: qemu-img info /var/lib/one/datastores/1/5d2be6f9746fc3447be11c6621da87df image: /var/lib/one/datastores/1/5d2be6f9746fc3447be11c6621da87df file format: qcow2 virtual size: 40G (42949672960 bytes) disk size: 629M cluster_size: 65536 Format specific information: compat: 1.1 lazy refcounts: false Regards. -- Daniel Dehennin Récupérer ma clef GPG: gpg --recv-keys 0xCC1E9E5B7A6FE2DF Fingerprint: 3E69 014E 5C23 50E8 9ED6 2AAD CC1E 9E5B 7A6F E2DF signature.asc Description: PGP signature ___ Users mailing list Users@lists.opennebula.org http://lists.opennebula.org/listinfo.cgi/users-opennebula.org
Re: [one-users] Can't loginto sunstone after upgrade to 4.8
Ale ale...@gmail.com writes: Hello all, After upgrade opennebula from 4.6 to 4.8 through deb, i follow the steps on the upgrade guide. oned.log report bad decript http://pastebin.com/KfVYpeSL Probably i've made some mistake, any suggestion? This means that the /var/lib/one/.one/sunstone_auth does not match the password in the database. Does the CLIs work for oneadmin user? oneadmin@nebula:~$ onehost list I personally reset the serveradmin password with the following: # Generate a random password oneadmin@nebula:~$ pwgen -s 42 1 serveradmin.pwd # Make serveradmin using core auth to change the password oneadmin@nebula:~$ oneuser chauth serveradmin core -r serveradmin.pwd # Switch back to server_cipher oneadmin@nebula:~$ oneuser chauth serveradmin server_cipher # Store new password oneadmin@nebula:~$ for f in sunstone ec2 occi oneflow onegate; do \ echo serveradmin:$(cat serveradmin.pwd) ~/.one/${f}_auth ; \ done Then you can restart Sunstone. Regards. -- Daniel Dehennin Récupérer ma clef GPG: gpg --recv-keys 0xCC1E9E5B7A6FE2DF Fingerprint: 3E69 014E 5C23 50E8 9ED6 2AAD CC1E 9E5B 7A6F E2DF signature.asc Description: PGP signature ___ Users mailing list Users@lists.opennebula.org http://lists.opennebula.org/listinfo.cgi/users-opennebula.org
Re: [one-users] Debian Wheezy KVM 2.1
Arnold Bechtoldt m...@arnoldbechtoldt.com writes: Hey guys, Are there any known issues running Wheezy together with QEMU/ KVM 2.1 from wheezy-backports and ONE 4.8? Hello, I added the linux-image-3.14-0.bpo.2-amd64 and it works fine, don't forget to enable boot on HD and CDROM when installing[1]. With previous version of qemu, when only HD was sufficient when you had an empty “vda” and an installation ISO as “hda”. Regards. Footnotes: [1] http://dev.opennebula.org/issues/2952 -- Daniel Dehennin Récupérer ma clef GPG: gpg --recv-keys 0xCC1E9E5B7A6FE2DF Fingerprint: 3E69 014E 5C23 50E8 9ED6 2AAD CC1E 9E5B 7A6F E2DF signature.asc Description: PGP signature ___ Users mailing list Users@lists.opennebula.org http://lists.opennebula.org/listinfo.cgi/users-opennebula.org
Re: [one-users] web UI does not work for zone/vDC if sunstone server is behind a reverse proxy server
Liu, Gene gene@alcatel-lucent.com writes: Thanks Daniel! Could you please specify how? Using nginx, I have this in /etc/nginx/site-available/opennebula-sunstone: #+begin_src upstream sunstone { server 127.0.0.1:9869; } server { listen 80 default_server; listen [::]:80 default_server ipv6only=on; root /usr/share/nginx/html; index index.html index.htm; server_name nebula.example.net; access_log /var/log/nginx/opennebula-sunstone-access.log; error_log /var/log/nginx/opennebula-sunstone-error.log; # Pass headers with underscore underscores_in_headers on; # To upload ISO files, increase for VMs images client_max_body_size 1G; location / { include proxy_params; # Forward ZONE_NAME to ONE process proxy_set_header ZONE_NAME $http_zone_name; proxy_pass http://sunstone; } } #+end_src Regards. Gene On 14-08-25 06:16 AM, Daniel Molina wrote: ZONE_NAME http header ___ Users mailing list Users@lists.opennebula.org http://lists.opennebula.org/listinfo.cgi/users-opennebula.org -- Daniel Dehennin Récupérer ma clef GPG: gpg --recv-keys 0xCC1E9E5B7A6FE2DF Fingerprint: 3E69 014E 5C23 50E8 9ED6 2AAD CC1E 9E5B 7A6F E2DF signature.asc Description: PGP signature ___ Users mailing list Users@lists.opennebula.org http://lists.opennebula.org/listinfo.cgi/users-opennebula.org
Re: [one-users] Could not read from CDROM (code 0004)
IBC Service Network / Zolotuhin Aleksey a.zolotu...@ibc.ru writes: Hello! Hello, And, in my own opinion, forcing CONTEXT be the last CDROM device will eliminate the situation that I have encountered. I just open a request[1]. Regards. Footnotes: [1] http://dev.opennebula.org/issues/3158 -- Daniel Dehennin Récupérer ma clef GPG: gpg --recv-keys 0xCC1E9E5B7A6FE2DF Fingerprint: 3E69 014E 5C23 50E8 9ED6 2AAD CC1E 9E5B 7A6F E2DF signature.asc Description: PGP signature ___ Users mailing list Users@lists.opennebula.org http://lists.opennebula.org/listinfo.cgi/users-opennebula.org
Re: [one-users] SSL proxy
Johan Kooijman m...@johankooijman.com writes: As a follow up: the connection can be made, telnet works. Firefox tells me: GET https://theonepoc.cloud.nl:29876/ [HTTP/1.0 403 Forbidden 35ms] New state 'loaded', was 'disconnected'. Msg: noVNC ready: native WebSockets, canvas rendering util.js:110 New state 'connect', was 'loaded'. util.js:110 Skipping unsupported WebSocket binary sub-protocol util.js:111 Firefox can't establish a connection to the server at wss:// theonepoc.cloud.nl:29876/?token=xbw7pps1nuzhxz5b9nds. websock.js:333 As far as I understand, theonepoc.cloud.nl:29876 is not managed by nginx but by the python-websocket, nginx just serve the javascript code which open the WSS connection. Maybe with nginx support[1] you can reverse proxy the WebSocket: 1) make python-websocket listen on 127.0.0.1 only 2) configure nginx to listen on theonepoc.cloud.nl:29876 with SSL and proxy_pass it to python-websocket on 127.0.0.1 I do not test it, but a simpler approach could be to configure wss in /etc/one/sunstone-server.conf to use the same certificate and key than nginx? Regards. Footnotes: [1] http://nginx.org/en/docs/http/websocket.html -- Daniel Dehennin Récupérer ma clef GPG: gpg --recv-keys 0xCC1E9E5B7A6FE2DF Fingerprint: 3E69 014E 5C23 50E8 9ED6 2AAD CC1E 9E5B 7A6F E2DF signature.asc Description: PGP signature ___ Users mailing list Users@lists.opennebula.org http://lists.opennebula.org/listinfo.cgi/users-opennebula.org
Re: [one-users] Could not read from CDROM (code 0004)
IBC Service Network / Zolotuhin Aleksey a.zolotu...@ibc.ru writes: Many thanks for the explanation. The problem really was that the VM tried to boot from the context CD. After forcing the bootable CD as hda, everything worked as expected. Hello, That's the reason we force TARGET=hda on all installation ISO, I just can not have two of them used in the same template. Could it be better to force CONTEXT to be last CDROM device? Regards. -- Daniel Dehennin Récupérer ma clef GPG: gpg --recv-keys 0xCC1E9E5B7A6FE2DF Fingerprint: 3E69 014E 5C23 50E8 9ED6 2AAD CC1E 9E5B 7A6F E2DF signature.asc Description: PGP signature ___ Users mailing list Users@lists.opennebula.org http://lists.opennebula.org/listinfo.cgi/users-opennebula.org
[one-users] Failing ONE 4.7.80-1 on Debian
Hello, As I'm very interested in the changes to VNET[1][2] in the next ONE 4.8, I tried the Debian package I found in the new repos[3]. I don't know if it's in a state for testing but unfortunately, oned segfault. I attach the logs. Regards. Footnotes: [1] dev.opennebula.org/issues/2858 [2] https://github.com/OpenNebula/docs/blob/master/source/user/virtual_resource_management/vgg.rst [3] http://downloads.opennebula.org/repo/4.8/ -- Daniel Dehennin Récupérer ma clef GPG: gpg --recv-keys 0xCC1E9E5B7A6FE2DF Fingerprint: 3E69 014E 5C23 50E8 9ED6 2AAD CC1E 9E5B 7A6F E2DF root@test-nebula:~# apt-cache policy opennebula opennebula: Installed: 4.7.80-1 Candidate: 4.7.80-1 Version table: *** 4.7.80-1 0 500 http://downloads.opennebula.org/repo/4.8/Debian/7/ stable/opennebula amd64 Packages 100 /var/lib/dpkg/status 3.4.1-3.1 0 500 http://ftp.fr.debian.org/debian/ wheezy/main amd64 Packages # Enable debug root@test-nebula:~# sed -e '2a set -x\n' /usr/bin/one root@test-nebula:~# /bin/sh -x /etc/init.d/opennebula start + PATH=/sbin:/usr/sbin:/bin:/usr/bin + DESC=OpenNebula cloud + NAME=one + DAEMON=/usr/bin/one + DAEMON_ARGS= + PIDFILE=/var/run/one.pid + SCRIPTNAME=/etc/init.d/one + [ -x /usr/bin/one ] + . /lib/init/vars.sh + TMPTIME=0 + SULOGIN=no + DELAYLOGIN=no + UTC=yes + VERBOSE=no + FSCKFIX=no + [ -f /etc/default/rcS ] + . /etc/default/rcS + unset EDITMOTD + unset RAMRUN + unset RAMLOCK + [ -r /proc/cmdline ] + cat /proc/cmdline + [ ] + [ ] + . /lib/lsb/init-functions + run-parts --lsbsysinit --list /lib/lsb/init-functions.d + [ -r /lib/lsb/init-functions.d/20-left-info-blocks ] + . /lib/lsb/init-functions.d/20-left-info-blocks + FANCYTTY= + [ -e /etc/lsb-base-logging.sh ] + true + [ no != no ] + do_start + mkdir -p /var/run/one /var/lock/one + chown oneadmin /var/run/one /var/lock/one + su oneadmin -s /bin/sh -c one start + '[' -z '' ']' + ONE_PID=/var/run/one/oned.pid + ONE_SCHEDPID=/var/run/one/sched.pid + ONE_CONF=/etc/one/oned.conf + ONE_DB=/var/lib/one/one.db + ONE_LOG=/var/log/one/oned.log + ONE_SCHED_LOG=/var/log/one/sched.log + ONE_XMLRPC_LOG=/var/log/one/one_xmlrpc.log + ONED=/usr/bin/oned + ONE_SCHEDULER=/usr/bin/mm_sched + LOCK_FILE=/var/lock/one/one + KILL_9_SECONDS=5 + BACKUP=true + '[' start = -f ']' + case $1 in + setup ++ dirname /var/run/one/oned.pid + ONE_PID_DIR=/var/run/one + mkdir -p /var/run/one + '[' '!' -w /var/run/one ']' + '[' -f /var/lock/one/one ']' + start + '[' '!' -x /usr/bin/oned ']' + '[' '!' -x /usr/bin/mm_sched ']' + '[' '!' -f /var/lib/one/one.db ']' + '[' '!' -f /var/lib/one/.one/one_auth ']' + start_oned + '[' true = true ']' + '[' -f /var/log/one/oned.log ']' + '[' -f /var/log/one/one_xmlrpc.log ']' + LASTRC=0 + LASTPID=7652 + '[' 0 -ne 0 ']' + echo 7652 + start_sched + /usr/bin/oned -f + '[' true = true ']' + '[' -f /var/log/one/sched.log ']' + LASTRC=0 + LASTPID=7653 + '[' 0 -ne 0 ']' + echo 7653 + sleep 3 + /usr/bin/mm_sched /usr/bin/one: line 130: 7652 Segmentation fault $ONED -f 21 + STARTED=true ++ cat /var/run/one/oned.pid + ps 7652 + '[' 1 -ne 0 ']' + echo 'oned failed to start' oned failed to start + STARTED=false ++ cat /var/run/one/sched.pid + ps 7653 + '[' 0 -ne 0 ']' + '[' false == false ']' + stop + stop_oned + '[' -f /var/run/one/oned.pid ']' ++ cat /var/run/one/oned.pid + PID=7652 + kill 7652 + counter=0 + ps 7652 + rm -f /var/run/one/oned.pid + stop_sched + '[' -f /var/run/one/sched.pid ']' ++ cat /var/run/one/sched.pid + kill 7653 + rm -f /var/run/one/sched.pid /usr/bin/one: line 119: 7653 Terminated $ONE_SCHEDULER + exit -1 + : root@test-nebula:~# cat /var/log/one/oned.log Sat Jul 19 19:09:06 2014 [Z0][ONE][I]: Starting OpenNebula 4.7.80 OpenNebula Configuration File AUTH_MAD=AUTHN=ssh,x509,ldap,server_cipher,server_x509,EXECUTABLE=one_auth_mad DATASTORE_BASE_PATH=/var/lib/one//datastores DATASTORE_CAPACITY_CHECK=yes DATASTORE_LOCATION=/var/lib/one//datastores DATASTORE_MAD=ARGUMENTS=-t 15 -d dummy,fs,vmfs,lvm,ceph,dev,EXECUTABLE=one_datastore DB=BACKEND=sqlite DEFAULT_CDROM_DEVICE_PREFIX=hd DEFAULT_DEVICE_PREFIX=hd DEFAULT_IMAGE_TYPE=OS DEFAULT_UMASK=177 ENABLE_OTHER_PERMISSIONS=YES FEDERATION=MASTER_ONED=,MODE=STANDALONE,ZONE_ID=0 HM_MAD=EXECUTABLE=one_hm HOST_MONITORING_EXPIRATION_TIME=43200 HOST_PER_INTERVAL=15 IMAGE_RESTRICTED_ATTR=SOURCE IM_MAD=ARGUMENTS=-p 4124 -f 5 -t 50 -i 20,EXECUTABLE=collectd,NAME=collectd IM_MAD=ARGUMENTS=-r 3 -t 15 kvm,EXECUTABLE=one_im_ssh,NAME=kvm INHERIT_DATASTORE_ATTR=CEPH_HOST INHERIT_DATASTORE_ATTR=CEPH_SECRET INHERIT_DATASTORE_ATTR=CEPH_USER INHERIT_DATASTORE_ATTR=GLUSTER_HOST INHERIT_DATASTORE_ATTR=GLUSTER_VOLUME INHERIT_VNET_ATTR=VLAN_TAGGED_ID KEEPALIVE_MAX_CONN=30 KEEPALIVE_TIMEOUT=15 LOG=DEBUG_LEVEL=3,SYSTEM=file MAC_PREFIX=02:00 MANAGER_TIMER=15 MAX_CONN=15 MAX_CONN_BACKLOG=15 MESSAGE_SIZE=1073741824 MONITORING_INTERVAL=60 MONITORING_THREADS
Re: [one-users] Failing ONE 4.7.80-1 on Debian
Javier Fontan jfon...@opennebula.org writes: What version of debian are you using? We have tried both 7.1 and 7.6 without problems. Ok, I tried on testing, I'll give a try on Wheezy. Thanks. -- Daniel Dehennin Récupérer ma clef GPG: gpg --recv-keys 0xCC1E9E5B7A6FE2DF Fingerprint: 3E69 014E 5C23 50E8 9ED6 2AAD CC1E 9E5B 7A6F E2DF signature.asc Description: PGP signature ___ Users mailing list Users@lists.opennebula.org http://lists.opennebula.org/listinfo.cgi/users-opennebula.org
Re: [one-users] Failing ONE 4.7.80-1 on Debian
Daniel Dehennin daniel.dehen...@baby-gnu.org writes: Ok, I tried on testing, I'll give a try on Wheezy. Hello, After your answer on IRC, I tried to strace the oned process to see what could be wrong, I modifies the /usr/bin/one on line 187 with: strace -f -o /tmp/oned.strace.log $ONED -f 21 Here is the output: 3933 open(/var/lib/one, O_RDONLY|O_CLOEXEC) = 5 3933 fsync(5) = 0 3933 close(5) = 0 3933 lseek(4, 0, SEEK_SET) = 0 3933 write(4, \331\325\5\371 \241c\327\0\0\0\3, 12) = 12 3933 fsync(4) = 0 3933 lseek(3, 0, SEEK_SET) = 0 3933 write(3, SQLite format 3\0\4\0\1\1\0@ \0\0\0\37\0\0\0+..., 1024) = 1024 3933 lseek(3, 39936, SEEK_SET) = 39936 3933 write(3, \r\0\0\0\1\3k\0\3k\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0..., 1024) = 1024 3933 lseek(3, 40960, SEEK_SET) = 40960 3933 write(3, \n\0\0\0\1\3\362\0\3\362\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0..., 1024) = 1024 3933 fsync(3) = 0 3933 close(4) = 0 3933 unlink(/var/lib/one/one.db-journal) = 0 3933 fcntl(3, F_SETLK, {type=F_RDLCK, whence=SEEK_SET, start=1073741826, len=510}) = 0 3933 fcntl(3, F_SETLK, {type=F_UNLCK, whence=SEEK_SET, start=1073741824, len=2}) = 0 3933 fcntl(3, F_SETLK, {type=F_UNLCK, whence=SEEK_SET, start=0, len=0}) = 0 3933 --- SIGSEGV (Segmentation fault) @ 0 (0) --- 3933 +++ killed by SIGSEGV +++ I don't know if this help, it's on Debian wheezy AMD64, uname -a: Linux test-nebula 3.14-1-amd64 #1 SMP Debian 3.14.12-1 (2014-07-11) x86_64 GNU/Linux Regards. -- Daniel Dehennin Récupérer ma clef GPG: gpg --recv-keys 0xCC1E9E5B7A6FE2DF Fingerprint: 3E69 014E 5C23 50E8 9ED6 2AAD CC1E 9E5B 7A6F E2DF signature.asc Description: PGP signature ___ Users mailing list Users@lists.opennebula.org http://lists.opennebula.org/listinfo.cgi/users-opennebula.org
Re: [one-users] creating a template with install cd
Daniel Dehennin daniel.dehen...@baby-gnu.org writes: [...] Or libvirt, forcing OS=[BOOT=cdrom] makes it working. I have an issue when setting OS=[BOOT=cdrom], when the installation is finished, it reboot on the ISO and not the hard drive and I can not detach the ISO. We are testing ONE on Ubuntu 14.04. Regards. -- Daniel Dehennin Récupérer ma clef GPG: gpg --recv-keys 0xCC1E9E5B7A6FE2DF Fingerprint: 3E69 014E 5C23 50E8 9ED6 2AAD CC1E 9E5B 7A6F E2DF signature.asc Description: PGP signature ___ Users mailing list Users@lists.opennebula.org http://lists.opennebula.org/listinfo.cgi/users-opennebula.org
Re: [one-users] creating a template with install cd
Stefan Kooman ste...@bit.nl writes: Quoting Daniel Dehennin (daniel.dehen...@baby-gnu.org): Daniel Dehennin daniel.dehen...@baby-gnu.org writes: [...] Or libvirt, forcing OS=[BOOT=cdrom] makes it working. I have an issue when setting OS=[BOOT=cdrom], when the installation is finished, it reboot on the ISO and not the hard drive and I can not detach the ISO. Yeah, installing a VM with ONE is a bit cumbersome. The only way I know is to shutdown the vm (or delete it, if it's already powered off), update the template and re-instantiate. Having the possibility to do (offline) attach/detach and changing of boot device would help in install scenarios. I found a solution: #+begin_src OS=[ARCH=x86_64,BOOT=hd] RAW=[DATA= os boot dev='cdrom'/ /os ,TYPE=kvm] #+end_src This result in the following xml: #+begin_src domain type='kvm' xmlns:qemu='http://libvirt.org/schemas/domain/qemu/1.0' nameone-20/name cputune shares512/shares /cputune memory1048576/memory os type arch='x86_64'hvm/type boot dev='hd'/ /os devices emulator/usr/bin/kvm/emulator disk type='file' device='disk' source file='/var/lib/one//datastores/100/20/disk.0'/ target dev='vda'/ driver name='qemu' type='qcow2' cache='default'/ /disk disk type='file' device='cdrom' source file='/var/lib/one//datastores/100/20/disk.1'/ target dev='hda'/ readonly/ driver name='qemu' type='raw' cache='default'/ /disk disk type='file' device='cdrom' source file='/var/lib/one//datastores/100/20/disk.2'/ target dev='hdb'/ readonly/ driver name='qemu' type='raw'/ /disk interface type='bridge' source bridge='vswitch'/ virtualport type='openvswitch'/ mac address='02:00:ac:10:2a:65'/ model type='virtio'/ /interface graphics type='vnc' listen='0.0.0.0' port='5920' keymap='fr'/ input type='tablet' bus='usb'/ /devices features acpi/ /features os boot dev='cdrom'/ /os /domain #+end_src In that case the VM try to boot on hard drive first and then CDROM. Regards. -- Daniel Dehennin Récupérer ma clef GPG: gpg --recv-keys 0xCC1E9E5B7A6FE2DF Fingerprint: 3E69 014E 5C23 50E8 9ED6 2AAD CC1E 9E5B 7A6F E2DF signature.asc Description: PGP signature ___ Users mailing list Users@lists.opennebula.org http://lists.opennebula.org/listinfo.cgi/users-opennebula.org
Re: [one-users] creating a template with install cd
Stefan Kooman ste...@bit.nl writes: [...] Did you do a onehost sync (--force) after this change? No, the /etc/one/vmm_exec/vmm_exec_kvm.conf are not synced to hosts IIRC. In the mean time I found the issue[1]. Regards. Footnotes: [1] http://dev.opennebula.org/issues/2951 -- Daniel Dehennin Récupérer ma clef GPG: gpg --recv-keys 0xCC1E9E5B7A6FE2DF Fingerprint: 3E69 014E 5C23 50E8 9ED6 2AAD CC1E 9E5B 7A6F E2DF signature.asc Description: PGP signature ___ Users mailing list Users@lists.opennebula.org http://lists.opennebula.org/listinfo.cgi/users-opennebula.org
Re: [one-users] User parameterized template
Carlos Martín Sánchez cmar...@opennebula.org writes: [...] That syntax is only supported inside the CONTEXT section. You can however use the network name only: NIC = [ NETWORK = foo-net ] If you don't provide the network_uname or network_uid, opennebula will look for the network owned by the user instantiating the template. Bear in mind that all users must own a network with the same name, otherwise the deployment will fail. Great, that's exactly what I was trying to use but the network name is prefixed with the user name. We will make some try. Regards. -- Daniel Dehennin Récupérer ma clef GPG: gpg --recv-keys 0xCC1E9E5B7A6FE2DF Fingerprint: 3E69 014E 5C23 50E8 9ED6 2AAD CC1E 9E5B 7A6F E2DF signature.asc Description: PGP signature ___ Users mailing list Users@lists.opennebula.org http://lists.opennebula.org/listinfo.cgi/users-opennebula.org
Re: [one-users] creating a template with install cd
Javier Fontan jfon...@opennebula.org writes: Set in ubuntu CDROM the TARGET to hda. The bios should catch it. Also make sure that the cd driver is raw: TARGET=hda DRIVER=raw It's not working here after an upgrade to 4.6, this template was working on 4.4: #+begin_src disk type='file' device='disk' source file='/var/lib/one//datastores/101/65/disk.0'/ target dev='vda'/ driver name='qemu' type='qcow2' cache='writeback' io='native'/ /disk disk type='file' device='cdrom' source file='/var/lib/one//datastores/101/65/disk.1'/ target dev='hda'/ readonly/ driver name='qemu' type='raw' cache='none'/ /disk disk type='file' device='cdrom' source file='/var/lib/one//datastores/101/65/disk.2'/ target dev='hdb'/ readonly/ driver name='qemu' type='raw'/ /disk #+end_src It could be the qemu backport to 2.0 which makes problem. Regards. -- Daniel Dehennin Récupérer ma clef GPG: gpg --recv-keys 0xCC1E9E5B7A6FE2DF Fingerprint: 3E69 014E 5C23 50E8 9ED6 2AAD CC1E 9E5B 7A6F E2DF signature.asc Description: PGP signature ___ Users mailing list Users@lists.opennebula.org http://lists.opennebula.org/listinfo.cgi/users-opennebula.org
Re: [one-users] creating a template with install cd
Daniel Dehennin daniel.dehen...@baby-gnu.org writes: [...] It could be the qemu backport to 2.0 which makes problem. Or libvirt, forcing OS=[BOOT=cdrom] makes it working. Regards. -- Daniel Dehennin Récupérer ma clef GPG: gpg --recv-keys 0xCC1E9E5B7A6FE2DF Fingerprint: 3E69 014E 5C23 50E8 9ED6 2AAD CC1E 9E5B 7A6F E2DF signature.asc Description: PGP signature ___ Users mailing list Users@lists.opennebula.org http://lists.opennebula.org/listinfo.cgi/users-opennebula.org
[one-users] User parameterized template
Hello, Trying to factorize the VM template on ONE 4.6.1 for our internal use, I'm trying to create a user parametric[1] template for network. We have one network topology model and want to ease running their VMs in their own pre-defined networks[2]. I'm trying to define the template with the following NIC definition: NIC=[NETWORK_UNAME=${UNAME},NETWORK=${UNAME}-net] When a user foo instanciate the VM template, I would like ONE to “resolve” the NIC definition to: NIC=[NETWORK_UNAME=foo,NETWORK=foo-net] Is there a way to achive something like this? Regards. Footnotes: [1] like for context http://docs.opennebula.org/4.6/user/references/template.html#context-section [2] not like asked in http://dev.opennebula.org/issues/2125 -- Daniel Dehennin Récupérer ma clef GPG: gpg --recv-keys 0xCC1E9E5B7A6FE2DF Fingerprint: 3E69 014E 5C23 50E8 9ED6 2AAD CC1E 9E5B 7A6F E2DF signature.asc Description: PGP signature ___ Users mailing list Users@lists.opennebula.org http://lists.opennebula.org/listinfo.cgi/users-opennebula.org
Re: [one-users] Unwanted upgrade from 4.4.1 to 4.6.0
Javier Fontan jfon...@opennebula.org writes: That's very unfortunate. We should have to rethink the repository process and maybe create a new one for each major release. +1 This would permit to have an “unstable” repository to test new features when they arrive and a “testing” one for RC. Managing repositories is easy with reprepro[1]. Regards. Footnotes: [1] https://wiki.debian.org/SettingUpSignedAptRepositoryWithReprepro -- Daniel Dehennin Récupérer ma clef GPG: gpg --recv-keys 0xCC1E9E5B7A6FE2DF Fingerprint: 3E69 014E 5C23 50E8 9ED6 2AAD CC1E 9E5B 7A6F E2DF signature.asc Description: PGP signature ___ Users mailing list Users@lists.opennebula.org http://lists.opennebula.org/listinfo.cgi/users-opennebula.org
Re: [one-users] OpenNebula 4.5.85-1 on Ubuntu Trusty Tahr
Javier Fontan jfon...@opennebula.org writes: Hi, Hello, Thanks for taking the time to test it. We have already created packages for Ubuntu 14.04 and we are now in the process of testing them. There was no time for RC but they will be available for the final release. Is the packaging code[1] available somewhere? In #2843[2] someone asked for RC packages but I would like to help in the process of packaging for distribution, at least Debian based. I setup a poor-man wanna-build[3] at home to ease the process of building packages in clean environments, but I did not find the packaging of ONE in git. Regards. Footnotes: [1] i.e the debian/ directory [2] http://dev.opennebula.org/issues/2843 [3] Debian wanna-build is to complex to setup for now. http://git.baby-gnu.net/gitweb/gitweb.cgi?p=packaging-with-dvcs.git;a=summary -- Daniel Dehennin Récupérer ma clef GPG: gpg --recv-keys 0xCC1E9E5B7A6FE2DF Fingerprint: 3E69 014E 5C23 50E8 9ED6 2AAD CC1E 9E5B 7A6F E2DF signature.asc Description: PGP signature ___ Users mailing list Users@lists.opennebula.org http://lists.opennebula.org/listinfo.cgi/users-opennebula.org
Re: [one-users] Resume paused VM due to full system datastore
Ruben S. Montero rsmont...@opennebula.org writes: Hi Hello, [...] Note also that we can re-send from OpenNebula a deploy action (virsh create, in your case) with onevm boot (or through the play icon in Sunstone) for UNKNOWN, however there is no way to send from OpenNebula a 'virsh resume' command. I was looking for this, or at least, the kvm monitoring could display that the VMs is in “paused” state. I'll emulate a DS full to see which information can be reported to the user. BTW OpenNebula 4.4 should control the system datastore usage (even not shared) and not schedule VMs to prevent taking all the space, you can even limit manually the size used by OpenNebula. Is this not working for you? Unfortunately, this cluster is using ONE 4.2. The problem occurs not on new scheduled VM but when qcow2 images grow up, when the DS is full, any VM trying to write, making their disk bigger, are “paused” by libvirt/kvm. Regards. -- Daniel Dehennin Récupérer ma clef GPG: gpg --recv-keys 0xCC1E9E5B7A6FE2DF Fingerprint: 3E69 014E 5C23 50E8 9ED6 2AAD CC1E 9E5B 7A6F E2DF signature.asc Description: PGP signature ___ Users mailing list Users@lists.opennebula.org http://lists.opennebula.org/listinfo.cgi/users-opennebula.org
[one-users] Resume paused VM due to full system datastore
Hello, I just encounter an issue with KVM based VMs when the non shared system datastore became full. The libvirt/kvm paused the VMs trying to write on their discs and I have to run: for vm in $(virsh -c qemu:///system list | awk '/paused/ {print $1}') do virsh -c qemu:///system resume ${vm} done In ONE they was in UNKNOWN state. Shouldn't it be handled by ONE directly? Regards. -- Daniel Dehennin Récupérer ma clef GPG: gpg --recv-keys 0xCC1E9E5B7A6FE2DF Fingerprint: 3E69 014E 5C23 50E8 9ED6 2AAD CC1E 9E5B 7A6F E2DF signature.asc Description: PGP signature ___ Users mailing list Users@lists.opennebula.org http://lists.opennebula.org/listinfo.cgi/users-opennebula.org
Re: [one-users] Request for comments: Sunstone dashboard
Stefan Kooman ste...@bit.nl writes: [...] Hello, As Stefan point it out so clearly, I'll just “plus” him ;-) - Make it interactive Currently it's a static page with graphs being updated every $update_interval. It would be nice to be able to click Storage, Users, Network, Hosts, Virtual Machines and get there right away. This will save you at least one click. +1 - Drill down funtionality The dashboard could function as a gateway to a (new) Statistics page. Clicking on for example Virtual Machines would give you all relevant information about VM statistics. Clicking on an individual vm on that page would give you all relevant info about that particular machine. With all statistics for this vm nicely organized (network, disk, cpu, mem, etc.). +1 As I do not use zones and VDC, I have no opinion about them. Regards. -- Daniel Dehennin Récupérer ma clef GPG: gpg --recv-keys 0xCC1E9E5B7A6FE2DF Fingerprint: 3E69 014E 5C23 50E8 9ED6 2AAD CC1E 9E5B 7A6F E2DF signature.asc Description: PGP signature ___ Users mailing list Users@lists.opennebula.org http://lists.opennebula.org/listinfo.cgi/users-opennebula.org
Re: [one-users] Adding test system datastore
Carlos Martín Sánchez cmar...@opennebula.org writes: Hi there, Maybe we could try to replicate the 'onehost disable' behaviour. That is, a disabled system DS would be ignored by the scheduler, but you can still deploy VMs manually there. What do you think? Should I open a ticket for this? Am'I right about group and resource providers? If yes, disabling a DS is not useful for my usecase. But remains interesting for general cluster management, for example if something goes wrong with the backend, system DS could be disabled and VMs restarted to use another one. Regards. -- Daniel Dehennin Récupérer ma clef GPG: gpg --recv-keys 0xCC1E9E5B7A6FE2DF Fingerprint: 3E69 014E 5C23 50E8 9ED6 2AAD CC1E 9E5B 7A6F E2DF signature.asc Description: PGP signature ___ Users mailing list Users@lists.opennebula.org http://lists.opennebula.org/listinfo.cgi/users-opennebula.org
[one-users] Adding test system datastore
Hello, I would like to add a new system datastore to make some tests on my ONE 4.2 cluster. My problem is that it will be automatically used when added to my cluster. I would like to define some requirements to make it selectable only for specific VM. Is there a way to mark a system datastore to not be used by default, except for template with the “good” requirements? Regards. -- Daniel Dehennin Récupérer ma clef GPG: gpg --recv-keys 0xCC1E9E5B7A6FE2DF Fingerprint: 3E69 014E 5C23 50E8 9ED6 2AAD CC1E 9E5B 7A6F E2DF signature.asc Description: PGP signature ___ Users mailing list Users@lists.opennebula.org http://lists.opennebula.org/listinfo.cgi/users-opennebula.org
Re: [one-users] Adding test system datastore
Stefan Kooman ste...@bit.nl writes: [...] You can, in your vm template you can put the following: SCHED_DS_REQUIREMENTS=NAME=NAMEOFYOURTESTDATASTORE. All other DS'es will be filtered out. This will force a VM to use this one when it have this SCHED_DS_REQUIREMENTS, but what about all other VMs which do not have any requirements set? As far as I understand, they will use it too. [...] Actually I was looking for this as well :). For example I would like to be able to link certain datastores to groups, i.e. use datastore A for group A by default instead of having it defined in every single template. If users of group A have the possibility to define their own templates and they would forget the SCHED_DS_REQUIREMENTS to their datastore it would end up somewhere else (if at least they have enough rights on other (system) datastores). Basically it would be nice to have even more filtering capabitilites to ensure correct placement/deployment. If I remember correctly, I heard about some group and resource providers mechanisms for 4.6, it may address this. Regards. -- Daniel Dehennin Récupérer ma clef GPG: gpg --recv-keys 0xCC1E9E5B7A6FE2DF Fingerprint: 3E69 014E 5C23 50E8 9ED6 2AAD CC1E 9E5B 7A6F E2DF signature.asc Description: PGP signature ___ Users mailing list Users@lists.opennebula.org http://lists.opennebula.org/listinfo.cgi/users-opennebula.org
Re: [one-users] Sunstone with nginx as reverse-proxy - server exception
Daniel Molina dmol...@opennebula.org writes: On 4 February 2014 17:58, ML mail mlnos...@yahoo.com wrote: I have found out the problem with the 403 forbidden errors, I needed to pass the following HTTP headers in my nginx reverse proxy configuration: proxy_set_headerX-Real-IP $remote_addr; proxy_set_headerX-Forwarded-For $proxy_add_x_forwarded_for; proxy_set_headerHost $http_host; Great, could you share with the list how did you configure nginx and sunstone, and the whole configuration? Personnaly I added a “client_max_body_size” to upload images: #+begin_src conf # /etc/nginx/sites-available/opennebula-sunstone # Please see /usr/share/doc/nginx-doc/examples/ for more detailed examples. ## upstream sunstone { server 127.0.0.1:9869; } server { listen 80 default_server; listen [::]:80 default_server ipv6only=on; root /usr/share/nginx/html; index index.html index.htm; server_name nebula.baby-gnu.net; access_log /var/log/nginx/opennebula-sunstone-access.log; error_log /var/log/nginx/opennebula-sunstone-error.log; # To upload ISO files, must be increased for VMs images client_max_body_size 1G; location / { include proxy_params; proxy_pass http://sunstone; } } #+end_src Regards. -- Daniel Dehennin Récupérer ma clef GPG: gpg --keyserver pgp.mit.edu --recv-keys 0x7A6FE2DF pgpFrkVOyyCUo.pgp Description: PGP signature ___ Users mailing list Users@lists.opennebula.org http://lists.opennebula.org/listinfo.cgi/users-opennebula.org
Re: [one-users] SPICE support
Javier Fontan jfon...@opennebula.org writes: Hi, Hello, [...] The proposed solution is adding an special raw parameter that is automatically added when SPICE is selected. Another option is having a predefined set of options that can be configured like the number of USBs or the sound card model. I am not used to SPICE and can not set the parameters more used nor understand values like the channel name (is it always the same?). Could you help me selecting which parameters should we add? Do you prefer a RAW like value where to add anything you want? To use it, I would like to make a selection in some predefined items. 1. select SPICE graphics, this display a list of predefined items 2. checkbox “enable vdagent support”, if selected: display the channel name with a default to “com.redhat.spice.0” which seems to be the default, at leat on Debian[1] 3. checkbox “enable guest agent”, if selected, the target name must be “org.qemu.guest_agent.0” to communicate with guest agent[2] 4. checkbox “enable USB ports”, if selected: ask the number of ports 5. checkbox “enable sound”, if selected: list card models, with a default selected For the UI, I'm not sure about the fixed list of checkboxes, maybe some kind of list to select items to enable which add a bloc in the wizard: 1. select SPICE graphics 2. display a list: “Enable list of items” 3. user select one item in the list, for example “vdagent support” 4. a bloc is added in the wizard: * with the title “vdagent support (X)”, the (X) represent a “remove this items” * with an input field labeled “channel name:” and a default value of “com.redhat.spice.0” 5. the “vdagent support” is removed from the selection list to avoid adding several times the same item Then, we could be mad and imagine some kind of “RAW wizard creator”, to let users define their own RAW templates: - define a type: “KVM” - define label: “enable USB ports” - checkbox “require user input”, if selected, for each input * define id: “usb_port_numbers” * define label: “number of port” * define type: “integer” * define default: “3” - define a RAW template (as Jinja template since I don't know ruby ones): devices {% for port in usb_port_numbers %} redirdev bus='usb' type='spicevmc'/ {% endfor %} /devices But I'm getting off-topic of the thread ;-) Regards. Footnotes: [1] http://anonscm.debian.org/gitweb/?p=collab-maint/spice-vdagent.git;a=blob;f=debian/spice-vdagent.init#l28 [2] http://libvirt.org/formatdomain.html#elementCharChannel -- Daniel Dehennin Récupérer ma clef GPG: gpg --keyserver pgp.mit.edu --recv-keys 0x7A6FE2DF pgpmgKzU44Ghe.pgp Description: PGP signature ___ Users mailing list Users@lists.opennebula.org http://lists.opennebula.org/listinfo.cgi/users-opennebula.org
[one-users] Sunstone errors after upgrade to 4.4.1 on Debian Jessie
Hello, I just upgrade my ONE installation to 4.4.1 from OpenNebula deb repository[1] and my sunstone does not work anymore. The CLI is working. According to my logs, it looks like a problem with ruby Tilt, mine is in version 2.0.0-1 from debian, here are the sunstone logs: root@yggdrasil:/etc/one# cat /var/log/one/sunstone.error == Sinatra/1.4.3 has taken the stage on 9869 for development with backup from Thin WARN: tilt autoloading 'tilt/erb' in a non thread-safe way; explicit require 'tilt/erb' suggested. NoMethodError - undefined method `mappings' for Tilt:Module: /usr/lib/ruby/vendor_ruby/sinatra/base.rb:755:in `find_template' /usr/lib/ruby/vendor_ruby/sinatra/base.rb:829:in `block in compile_template' /usr/lib/ruby/vendor_ruby/tilt.rb:88:in `fetch' /usr/lib/ruby/vendor_ruby/sinatra/base.rb:816:in `compile_template' /usr/lib/ruby/vendor_ruby/sinatra/base.rb:797:in `render' /usr/lib/ruby/vendor_ruby/sinatra/base.rb:807:in `block in render' /usr/lib/ruby/vendor_ruby/sinatra/base.rb:807:in `catch' /usr/lib/ruby/vendor_ruby/sinatra/base.rb:807:in `render' /usr/lib/ruby/vendor_ruby/sinatra/base.rb:657:in `erb' /usr/lib/one/sunstone/sunstone-server.rb:261:in `block in main' /usr/lib/ruby/vendor_ruby/sinatra/base.rb:1541:in `call' /usr/lib/ruby/vendor_ruby/sinatra/base.rb:1541:in `block in compile!' /usr/lib/ruby/vendor_ruby/sinatra/base.rb:950:in `[]' /usr/lib/ruby/vendor_ruby/sinatra/base.rb:950:in `block (3 levels) in route!' /usr/lib/ruby/vendor_ruby/sinatra/base.rb:966:in `route_eval' /usr/lib/ruby/vendor_ruby/sinatra/base.rb:950:in `block (2 levels) in route!' /usr/lib/ruby/vendor_ruby/sinatra/base.rb:987:in `block in process_route' /usr/lib/ruby/vendor_ruby/sinatra/base.rb:985:in `catch' /usr/lib/ruby/vendor_ruby/sinatra/base.rb:985:in `process_route' /usr/lib/ruby/vendor_ruby/sinatra/base.rb:948:in `block in route!' /usr/lib/ruby/vendor_ruby/sinatra/base.rb:947:in `each' /usr/lib/ruby/vendor_ruby/sinatra/base.rb:947:in `route!' /usr/lib/ruby/vendor_ruby/sinatra/base.rb:1059:in `block in dispatch!' /usr/lib/ruby/vendor_ruby/sinatra/base.rb:1041:in `block in invoke' /usr/lib/ruby/vendor_ruby/sinatra/base.rb:1041:in `catch' /usr/lib/ruby/vendor_ruby/sinatra/base.rb:1041:in `invoke' /usr/lib/ruby/vendor_ruby/sinatra/base.rb:1056:in `dispatch!' /usr/lib/ruby/vendor_ruby/sinatra/base.rb:882:in `block in call!' /usr/lib/ruby/vendor_ruby/sinatra/base.rb:1041:in `block in invoke' /usr/lib/ruby/vendor_ruby/sinatra/base.rb:1041:in `catch' /usr/lib/ruby/vendor_ruby/sinatra/base.rb:1041:in `invoke' /usr/lib/ruby/vendor_ruby/sinatra/base.rb:882:in `call!' /usr/lib/ruby/vendor_ruby/sinatra/base.rb:870:in `call' /usr/lib/ruby/vendor_ruby/rack/commonlogger.rb:33:in `call' /usr/lib/ruby/vendor_ruby/sinatra/base.rb:212:in `call' /usr/lib/ruby/vendor_ruby/rack/session/abstract/id.rb:225:in `context' /usr/lib/ruby/vendor_ruby/rack/session/abstract/id.rb:220:in `call' /usr/lib/ruby/vendor_ruby/rack/protection/xss_header.rb:18:in `call' /usr/lib/ruby/vendor_ruby/rack/protection/path_traversal.rb:16:in `call' /usr/lib/ruby/vendor_ruby/rack/protection/json_csrf.rb:18:in `call' /usr/lib/ruby/vendor_ruby/rack/protection/base.rb:50:in `call' /usr/lib/ruby/vendor_ruby/rack/protection/base.rb:50:in `call' /usr/lib/ruby/vendor_ruby/rack/protection/frame_options.rb:31:in `call' /usr/lib/ruby/vendor_ruby/rack/nulllogger.rb:9:in `call' /usr/lib/ruby/vendor_ruby/rack/head.rb:11:in `call' /usr/lib/ruby/vendor_ruby/rack/methodoverride.rb:21:in `call' /usr/lib/ruby/vendor_ruby/sinatra/showexceptions.rb:21:in `call' /usr/lib/ruby/vendor_ruby/sinatra/base.rb:175:in `call' /usr/lib/ruby/vendor_ruby/sinatra/base.rb:1949:in `call' /usr/lib/ruby/vendor_ruby/sinatra/base.rb:1449:in `block in call' /usr/lib/ruby/vendor_ruby/sinatra/base.rb:1726:in `synchronize' /usr/lib/ruby/vendor_ruby/sinatra/base.rb:1449:in `call' /usr/lib/ruby/vendor_ruby/thin/connection.rb:80:in `block in pre_process' /usr/lib/ruby/vendor_ruby/thin/connection.rb:78:in `catch' /usr/lib/ruby/vendor_ruby/thin/connection.rb:78:in `pre_process' /usr/lib/ruby/vendor_ruby/eventmachine.rb:1037:in `call' /usr/lib/ruby/vendor_ruby/eventmachine.rb:1037:in `block in spawn_threadpool' Regards. Footnotes: [1] deb http://downloads.opennebula.org/repo/Debian/7 stable opennebula -- Daniel Dehennin Récupérer ma clef GPG: gpg --keyserver pgp.mit.edu --recv-keys 0x7A6FE2DF pgpIFGNW9T133.pgp Description: PGP signature ___ Users mailing list Users@lists.opennebula.org http
Re: [one-users] error installing sunstone
Carlos Martín Sánchez cmar...@opennebula.org writes: Hello, [...] Well, what do the logs say? On my personal system, after the upgrade to 4.4.1 on a Debian Jessie, I have the following: root@yggdrasil:~# cat /var/log/one/occi-server.error Port 4567 busy. root@yggdrasil:~# netstat -lpn | grep ':4567' tcp0 0 127.0.0.1:4567 0.0.0.0:* LISTEN 17233/ruby root@yggdrasil:~# ps -p 17233 www PID TTY STAT TIME COMMAND 17233 ?Sl 0:00 ruby /usr/lib/one/ruby/cloud/econe/econe-server.rb So, it looks like econe-server and occi-server want to use the same port. Regards. -- Daniel Dehennin Récupérer ma clef GPG: gpg --keyserver pgp.mit.edu --recv-keys 0x7A6FE2DF pgpGn5Bs52DzF.pgp Description: PGP signature ___ Users mailing list Users@lists.opennebula.org http://lists.opennebula.org/listinfo.cgi/users-opennebula.org
Re: [one-users] [SOLVED] Sunstone errors after upgrade to 4.4.1 on Debian Jessie
Ruben S. Montero rsmont...@opennebula.org writes: Hi Daniel OpenNebula 4.4.1, has been certified using tilt up to 1.4.4 (wheezy shipd with 1.3 something, I believe). The upgrade to opennebula which requires ruby-sinatra, and in turns it requires ruby-tilt, may have upgraded the tilt gem in your setup... We should look at it, for the upcoming jessie ( http://dev.opennebula.org/issues/2716). If possible downgrade tilt Thanks a lot, I solved it be downgrading: echo 'deb http://ftp.fr.debian.org/debian/ wheezy main' /etc/apt/sources.list.d/wheezy.list apt-get update apt-get install ruby-tilt=1.3.3-2 ruby-sinatra=1.3.2-2 apt-mark hold ruby-tilt invoke-rc.d opennebula-sunstone restart A question is asked about the downgrade. Regards. -- Daniel Dehennin Récupérer ma clef GPG: gpg --keyserver pgp.mit.edu --recv-keys 0x7A6FE2DF pgpatgX7wLoiE.pgp Description: PGP signature ___ Users mailing list Users@lists.opennebula.org http://lists.opennebula.org/listinfo.cgi/users-opennebula.org
Re: [one-users] AppMarket 2.0 Beta (1.9.80) release
Daniel Molina dmol...@opennebula.org writes: [...] Feedback is always more than welcome. Feature requests for the AppMarket are handled through the Github issues. https://github.com/OpenNebula/addon-appmarket/issues?state=open Thanks, noted. [...] Currently OneFlow services are not supported, but it wouldn't be difficult to add this option. Currently, multi-image support means you can provide more than one image/datablock/cdrom for a given VM (1 VM Template - Multiple Disks) and the relationship is defined in the opennebula_template attribute. We could include an opennebula_service_template attribute also. I'll add a Github request for this. [...] Appliances currently have an attribute version, you can update the appliances and increment this value, but the original appliances will disappear. Why do you want to keep old versions of the appliance? This would permit to manage different states for an application: - old-stable - stable - next [...] By default AppMarket does not store any image, just the metadata of the appliance, therefore the url is just a reference to the file and it could be stored in a third party server. This is how the OpenNebula Marketplace works [1]. When you import an appliance from the AppMarket to OpenNebula, a new image is create an the path attribute contains the url of the file; this file will be downloaded to the specified datastore by the DS driver. In AppMarket 2.0 a new component was included, AppMarket Worker. This component allows you to create appliances from an OVA file. The OVA file is provided as an url and the worker will download, unpack and store the images and will update the appliance with the new urls for the files. The storage used by this component is not related to OpenNebula. So you can create appliances in two different ways: 1. Providing the urls of the files and the opennebula_template. In this case the Worker will not do anything. 2. Providing an OVA file. In this case the Worker will unpack and store the files and will generate the opennebula_template and files section of the appliance. Thanks for the clarification. [...] Is there a way to push appliances from ONE/Sunstone, I read From AppMarket 2.0 creating an appliance can be also done through the AppMarket tab in Sunstone. Providing a new VM/OneFlow action Store in AppMarket could be great ;-) Currently, this is not supported but it would be a nifty feature, Export VM/Service, a step forward in terms of portability To avoid duplicating images, it could be interesting that ONE provides public URLs for images usable as AppMarket URL. Thanks. -- Daniel Dehennin Récupérer ma clef GPG: gpg --keyserver pgp.mit.edu --recv-keys 0x7A6FE2DF pgpwRsroiNFcG.pgp Description: PGP signature ___ Users mailing list Users@lists.opennebula.org http://lists.opennebula.org/listinfo.cgi/users-opennebula.org
Re: [one-users] AppMarket 2.0 Beta (1.9.80) release
Jaime Melis jme...@opennebula.org writes: Dear community, Hello, we are pleased to announce the release of AppMarket 2.0 Beta: http://opennebula.org/appmarket-2-0-beta-1-9-80/ I discuss with some colleagues used to some very expensive proprietary software, I gather their needs to study the opportunity to switch part of their infrastructure to ONE. What they call “catalogue” is very important to them and they have a use-flow based on it, I'm preparing some redmine requests but I take the opportunity of this thread to ask some questions. NB: I do not setup an AppMarket yet. This release brings many new goodies to the AppMarket: AppMarket - Multi-image Appliances: AppMarket Appliances can now have multiple disks, which creates new possibilities to upload and register more complex and feature complete Appliances. Does this mean we can store and distribute a complete OneFlow service? Is there a way to clearly identify and group appliances by version, like: - one entry per appliance - one sub-entry per appliance version, with some sorting newest first/last I'm wondering about storage size, does AppMarket use the same storage as ONE or the disks are copied from ONE to AppMarket? I do not see clearly what is the storage requirement and how things get together between AppMarket/ONE/Sunstone. - Sunstone Import: users will be able to import registered Appliances to OpenNebula via Sunstone by using pre-filled Image and Template creation forms. - Sunstone Views: two new Sunstone views: an AppMarket *admin* view, that allows full control of the AppMarket, and a *user* view, that allows end users to import registered appliances. In the documentation “Create an Appliance”[1], we must provide “the URLs of the files and the OpenNebula template in the Appliance creation form”. Is there a way to push appliances from ONE/Sunstone, I read “From AppMarket 2.0 creating an appliance can be also done through the AppMarket tab in Sunstone”. Providing a new VM/OneFlow action “Store in AppMarket” could be great ;-) AppMarket Worker A new horizontally-scalable component that subscribes to the AppMarket and executes jobs. It enhances the AppMarket functionality by adding: - OVA processing: If a URL to an OVA package is supplied, the AppMarket will be able completely process it and integrate it to its repository. This involves: downloading and unpacking of the OVA package, parsing of the OVF file and creating a corresponding OpenNebula template. - Format conversion: appliances registered in a specific format, can be converted to a new format. - Extensibility: All the operations described above are implemented within the framework of an extensible driver engine, which allows further customization and integration by the administrators. Great feature, many thanks to Produban for funding. Regards. Footnotes: [1] https://github.com/OpenNebula/addon-appmarket/blob/master/doc/usage.md#create-an-appliance-user -- Daniel Dehennin Récupérer ma clef GPG: gpg --keyserver pgp.mit.edu --recv-keys 0x7A6FE2DF pgp_pPPw6lqkr.pgp Description: PGP signature ___ Users mailing list Users@lists.opennebula.org http://lists.opennebula.org/listinfo.cgi/users-opennebula.org
[one-users] libvirt/kvm: Permission denied on PERSISTENT images
Hello, Using ONE 4.2 on Debian wheezy. I just upgrade the qemu and libvirt to wheezy backports to solve bad interaction between emulated card and guest kernel[1]. Now I can not boot VMs with PERSISTENT images. I found a thread[2] on the mailing-list which looked like my problem, I made the checks without success. My environment: --- OS: Debian Wheezy qemu-kvm: 1.7.0+dfsg-2~bpo70+2 libvirt-bin: 1.2.1-1~bpo70+1 ONE: 4.2, from OpenNebula deb packages Datastores: - 0: local disk of each host - 1: NFS TM_MAD=qcow2 Test 1: --- IMAGE_ID=114 PERSISTENT=NO The VM boots fine Test 2: --- IMAGE_ID=114 PERSISTENT=YES The VM FAILED. The only difference is the use of a symlink, does someone know any issue with libvirt/qemu to use disk images though symlink? Regards. oneadmin@nebula:~$ cat vms/1937/transfer.0.prolog LN qcow2 benjamin:/var/lib/one/datastores/1/5d2be6f9746fc3447be11c6621da87df igor:/var/lib/one//datastores/0/1937/disk.0 1937 1 CONTEXT ssh /var/lib/one/vms/1937/context.sh igor:/var/lib/one//datastores/0/1937/disk.1 1937 0 oneadmin@nebula:~$ cat vms/1937/deployment.0 domain type='kvm' xmlns:qemu='http://libvirt.org/schemas/domain/qemu/1.0' nameone-1937/name vcpu1/vcpu cputune shares21/shares /cputune memory524288/memory os type arch='x86_64'hvm/type boot dev='hd'/ /os devices emulator/usr/bin/kvm/emulator disk type='file' device='disk' source file='/var/lib/one//datastores/0/1937/disk.0'/ target dev='vda'/ driver name='qemu' type='qcow2' cache='writeback'/ /disk disk type='file' device='cdrom' source file='/var/lib/one//datastores/0/1937/disk.1'/ target dev='hda'/ readonly/ driver name='qemu' type='raw'/ /disk interface type='bridge' source bridge='vsw'/ virtualport type='openvswitch'/ mac address='02:00:0a:00:2a:0e'/ model type='virtio'/ /interface graphics type='vnc' listen='0.0.0.0' port='7837' keymap='fr'/ /devices features acpi/ /features /domain oneadmin@nebula:~$ cat /var/log/one/1937.log Wed Jan 29 15:06:31 2014 [DiM][I]: New VM state is ACTIVE. Wed Jan 29 15:06:32 2014 [LCM][I]: New VM state is PROLOG. Wed Jan 29 15:06:46 2014 [LCM][I]: New VM state is BOOT Wed Jan 29 15:06:46 2014 [VMM][I]: Generating deployment file: /var/lib/one/vms/1937/deployment.0 Wed Jan 29 15:06:46 2014 [VMM][I]: ExitCode: 0 Wed Jan 29 15:06:46 2014 [VMM][I]: Successfully execute network driver operation: pre. Wed Jan 29 15:06:46 2014 [VMM][I]: Command execution fail: cat EOT | /var/tmp/one/vmm/kvm/deploy '/var/lib/one//datastores/0/1937/deployment.0' 'igor' 1937 igor Wed Jan 29 15:06:46 2014 [VMM][I]: error: Failed to create domain from /var/lib/one//datastores/0/1937/deployment.0 Wed Jan 29 15:06:46 2014 [VMM][I]: error: Failed to open file '/var/lib/one//datastores/0/1937/disk.0': Premission denied Wed Jan 29 15:06:46 2014 [VMM][E]: Could not create domain from /var/lib/one//datastores/0/1937/deployment.0 Wed Jan 29 15:06:46 2014 [VMM][I]: ExitCode: 255 Wed Jan 29 15:06:46 2014 [VMM][I]: Failed to execute virtualization driver operation: deploy. Wed Jan 29 15:06:46 2014 [VMM][E]: Error deploying virtual machine: Could not create domain from /var/lib/one//datastores/0/1937/deployment.0 Wed Jan 29 15:06:47 2014 [DiM][I]: New VM state is FAILED oneadmin@host:~/datastores/0/1937$ virsh -c qemu:///system create deployment.0 error: Failed to create domain from deployment.0 error: Failed to open file '/var/lib/one//datastores/0/1937/disk.0': Permission denied oneadmin@host:~/datastores/0/1937$ ls -l disk.0 lrwxrwxrwx 1 oneadmin oneadmin 58 jan. 29 15:06 disk.0 - /var/lib/one/datastores/1/5d2be6f9746fc3447be11c6621da87df oneadmin@host:~/datastores/0/1937$ ls -lL disk.0 -rw-rw 1 oneadmin oneadmin 1357250560 nov. 18 17:42 disk.0 root@host:~# grep -E '^(user|groud|dynamic)' /etc/libvirt/qemu.conf user = oneadmin dynamic_ownership = 0 Footnotes: [1] http://paste.debian.net/plain/78845 [2] http://lists.opennebula.org/pipermail/users-opennebula.org/2013-February/022168.html -- Daniel Dehennin Récupérer ma clef GPG: gpg --keyserver pgp.mit.edu --recv-keys 0x7A6FE2DF pgpmX0zLVTBC_.pgp Description: PGP signature ___ Users mailing list Users@lists.opennebula.org http://lists.opennebula.org/listinfo.cgi/users-opennebula.org
Re: [one-users] libvirt/kvm: Permission denied on PERSISTENT images
Hello, again, The problem is libvirt specific: 2014-01-29 15:10:52.458+: 6566: debug : virStorageFileGetMetadata:1114 : path=/var/lib/one//datastores/0/1937/disk.0 format=9 uid=9869 gid=9869 probe=0 2014-01-29 15:10:52.458+: 6566: debug : virStorageFileGetMetadataRecurse:1046 : path=/var/lib/one//datastores/0/1937/disk.0 format=9 uid=9869 gid=9869 probe=0 2014-01-29 15:10:52.458+: 6566: debug : virStorageFileIsSharedFSType:1322 : Check if path /var/lib/one//datastores/0/1937/disk.0 with FS magic 61267 is shared 2014-01-29 15:10:52.458+: 6566: error : virStorageFileGetMetadataRecurse:1060 : Failed to open file '/var/lib/one//datastores/0/1937/disk.0': Permission denied 2014-01-29 15:10:52.458+: 6566: debug : virObjectUnref:256 : OBJECT_UNREF: obj=0x7ffa8b845330 2014-01-29 15:10:52.458+: 6566: debug : virFileClose:90 : Closed fd 42 2014-01-29 15:10:52.458+: 6566: debug : virObjectRef:293 : OBJECT_REF: obj=0x7ffa8b845330 I can do an “md5sum /var/lib/one//datastores/0/1937/disk.0”. I'm asking on #virt@OFTC Regards. -- Daniel Dehennin Récupérer ma clef GPG: gpg --keyserver pgp.mit.edu --recv-keys 0x7A6FE2DF pgpX15odCXFyU.pgp Description: PGP signature ___ Users mailing list Users@lists.opennebula.org http://lists.opennebula.org/listinfo.cgi/users-opennebula.org
Re: [one-users] [SOLVED] libvirt/kvm: Permission denied on PERSISTENT images
Daniel Dehennin daniel.dehen...@baby-gnu.org writes: Hello, again, The problem is libvirt specific: [...] I can do an “md5sum /var/lib/one//datastores/0/1937/disk.0”. As “oneadmin”, but libvirtd is running as root and new version make some checks on the file, which failed because of a “root_squash”[1] Changing the exportfs parameter solve my issue, sorry for the noises. Footnotes: [1] http://dev.opennebula.org/issues/2381 -- Daniel Dehennin Récupérer ma clef GPG: gpg --keyserver pgp.mit.edu --recv-keys 0x7A6FE2DF pgpd50kyn68N5.pgp Description: PGP signature ___ Users mailing list Users@lists.opennebula.org http://lists.opennebula.org/listinfo.cgi/users-opennebula.org
Re: [one-users] Default RAW section for all VMs
Daniel Dehennin daniel.dehen...@baby-gnu.org writes: Hello, Hello, We encounter an issue with Ubuntu 12.04: the display is frozen just after the boot. After some tests, I found that defining the following RAW section solve our issue: devicesvideomodel type='vga' heads='1'//video/devices I would like to add it for all VMs, so I modify /etc/one/vmm_exec/vmm_exec_kvm.conf: RAW = devicesvideomodel type='vga' heads='1'//video/devices But just modifying this file does not work, is it required to restart opennebula? I have my answer: - /etc/one/vmm_exec/vmm_exec_kvm.conf is applied this moring after restarting ONE that night - RAW section in /etc/one/vmm_exec/vmm_exec_kvm.conf is not overridden, resulting in multiple video cards: devices video model type='vga' heads='1'/ /video /devices devicesvideomodel type='vga' heads='1'//video/devices Is it the expected behaviour? According to the configuration file, defining a raw section in a VM template should be overridden: # Default configuration attributes for the KVM driver # (all domains will use these values as defaults). These values can # be overridden in each VM template. Valid atributes are: Regards. -- Daniel Dehennin Récupérer ma clef GPG: gpg --keyserver pgp.mit.edu --recv-keys 0x7A6FE2DF pgpjjlaW5mpZo.pgp Description: PGP signature ___ Users mailing list Users@lists.opennebula.org http://lists.opennebula.org/listinfo.cgi/users-opennebula.org
Re: [one-users] Default RAW section for all VMs
Ruben S. Montero rsmont...@opennebula.org writes: Hi Hello, You are right, the values are overriden expect for RAW, which is appended. IMHO it is better to keep this behavior and change the comments on that file. This way, you can put basic needed RAW attributes for your setup and then add specific ones for each guest... We will go this way for now, but some kind of template inheritance like in my comment on #2149[1]. Defining the RAW section in my /etc/one/vmm_exec/vmm_exec_kvm.conf prevent my SPICE VMs from booting since their template define the video model to “qxl”. For this specific issue, I just need to define the global RAW video to “qxl” which is compatible to “vga”[2]. Regards. Footnotes: [1] http://dev.opennebula.org/issues/2149 [2] vga for libvirt and std for qemu -- Daniel Dehennin Récupérer ma clef GPG: gpg --keyserver pgp.mit.edu --recv-keys 0x7A6FE2DF pgpEpSZEJWfvc.pgp Description: PGP signature ___ Users mailing list Users@lists.opennebula.org http://lists.opennebula.org/listinfo.cgi/users-opennebula.org
[one-users] Default RAW section for all VMs
Hello, We encounter an issue with Ubuntu 12.04: the display is frozen just after the boot. After some tests, I found that defining the following RAW section solve our issue: devicesvideomodel type='vga' heads='1'//video/devices I would like to add it for all VMs, so I modify /etc/one/vmm_exec/vmm_exec_kvm.conf: RAW = devicesvideomodel type='vga' heads='1'//video/devices But just modifying this file does not work, is it required to restart opennebula? Regards. -- Daniel Dehennin Récupérer ma clef GPG: gpg --keyserver pgp.mit.edu --recv-keys 0x7A6FE2DF pgpR4lkMI2jXn.pgp Description: PGP signature ___ Users mailing list Users@lists.opennebula.org http://lists.opennebula.org/listinfo.cgi/users-opennebula.org
Re: [one-users] SPICE + QXL accessible VM
ML mail mlnos...@yahoo.com writes: Hello, Hello, I would like to use the SPICE protocol to access a VM in OpenNebula which will be used as remote/virtual destop. As far as I know I need to set the vga parameter of KVM to QXL but in the Sunstone interface I can't find any vga option in the VM template. How can I setup a VM from Sunstone which can be accessible through the SPICE protocol? There is an issue[1] on ONE redmine for this. Regards. Footnotes: [1] http://dev.opennebula.org/issues/2485 -- Daniel Dehennin Récupérer ma clef GPG: gpg --keyserver pgp.mit.edu --recv-keys 0x7A6FE2DF pgpzN8eTPgOMU.pgp Description: PGP signature ___ Users mailing list Users@lists.opennebula.org http://lists.opennebula.org/listinfo.cgi/users-opennebula.org
Re: [one-users] Monitor continually cycles through finding machines RUNNING and stat UNKNOWN
Javier Fontan jfon...@opennebula.org writes: It seems that there are more people having this problem and we are taking a look on several ways to fix this. One problem with /var/run is that it is normally owned by root and a process started by oneadmin user can not write there. In the frontend a new directory for OpenNebula pid files is created but in the nodes it does not exist. Hello, What do you think about a “ONE node setup init script”? On my debian systems I have opennebula-common and opennebula-node on each nodes. The -node could include an init script to setup an opennebula directory in /var/run[1] with proper owner and permissions. My 2¢. Regards. Footnotes: [1] /run now on debian system -- Daniel Dehennin Récupérer ma clef GPG: gpg --keyserver pgp.mit.edu --recv-keys 0x7A6FE2DF pgp4QS4lYcl1v.pgp Description: PGP signature ___ Users mailing list Users@lists.opennebula.org http://lists.opennebula.org/listinfo.cgi/users-opennebula.org
Re: [one-users] Closing OpenNebula 4.4 Cycle, Planning for 4.6
Tino Vazquez cvazq...@c12g.com writes: [...] As discussed with the OpenNebula Conf attendees (see? you need to come next year ;) ), we want to give our users the chance to shape the roadmap for the upcoming OpenNebula 4.6. Hello, Thanks a lot for giving us the opportunity to participate on the roadmap plan. It already seems quite busy, looking at issues not in 4.6 roadmap I would say, for my personal use: #1290 “Allowing autostart setting for KVM deployed VMs with persistent images, to restart them upon node reboot” It ease the “reboot after upgrade” for single node configuration. Today, on my single node, I need to stop all my VMs before doing the reboot and resume them by hand when reboot is done for my professional use: #2417 “Move images between datastores” -- It may be somewhat related to #2591 “Share resources between clusters” #2545 “Decorrelate networks TYPE and IP management” --- I'm quite sure this one is big enough to be delayed after 4.6. Ideally, the VM name and its IPs should match between ONE leases and my DHCP server but I don't figure it out. I think it's related to #2348 “Add static MAC address to” (I forget the VM NIC) Regards. -- Daniel Dehennin Récupérer ma clef GPG: gpg --keyserver pgp.mit.edu --recv-keys 0x7A6FE2DF pgpf0JWXR92Gu.pgp Description: PGP signature ___ Users mailing list Users@lists.opennebula.org http://lists.opennebula.org/listinfo.cgi/users-opennebula.org
[one-users] Missing tag ID on OpenvSwitch networks
Hello, On an ONE 4.2 cluster we have some network with VLAN enabled but the VLAN ID is missing and I'm not sure to understand what's happening. Some vnet interfaces are tagged with random like VLAN ID in “ovs-vsctl show” and some are tagged at all, the later are in trunk mode which is a problem. How does ONE manage such (miss-)?configuration? I suppose that ONE assign a VLAN ID to VLAN enabled network without one, in that case, is there a place where I can find this “dynamic VLAN IDs allocation”? Regards. -- Daniel Dehennin Récupérer ma clef GPG: gpg --keyserver pgp.mit.edu --recv-keys 0x7A6FE2DF ___ Users mailing list Users@lists.opennebula.org http://lists.opennebula.org/listinfo.cgi/users-opennebula.org
Re: [one-users] [4.2] [Sunstone] oneadmin does not see any VMs
Daniel Molina dmol...@opennebula.org writes: Hi Daniel, Hello, Could you check if there is any error in the browser console? In firefox 26 I have this: 12:00:36.024 TypeError: graphics.TYPE is undefined vms-tab.js:3217 Regards. -- Daniel Dehennin Récupérer ma clef GPG: gpg --keyserver pgp.mit.edu --recv-keys 0x7A6FE2DF pgpEWUvmAsjBv.pgp Description: PGP signature ___ Users mailing list Users@lists.opennebula.org http://lists.opennebula.org/listinfo.cgi/users-opennebula.org
Re: [one-users] [4.2] [Sunstone] oneadmin does not see any VMs
Daniel Molina dmol...@opennebula.org writes: [...] As the message says the problem is here: https://github.com/OpenNebula/one/blob/release-4.2/src/sunstone/public/js/plugins/vms-tab.js#L3217 Do you have any VM with a GRAPHICS section defined but with no TYPE attribute? Is there a way to query ONE? I do not plan to review 300 VM templates ;-) You can fix it changing that line to: if (graphics graphics.TYPE graphics.TYPE.toLowerCase() == vnc $.inArray(state, VNCstates)!=-1){ This fix the problem. Regards. -- Daniel Dehennin Récupérer ma clef GPG: gpg --keyserver pgp.mit.edu --recv-keys 0x7A6FE2DF pgppAnQuyWwKx.pgp Description: PGP signature ___ Users mailing list Users@lists.opennebula.org http://lists.opennebula.org/listinfo.cgi/users-opennebula.org
Re: [one-users] Missing tag ID on OpenvSwitch networks
Daniel Dehennin daniel.dehen...@baby-gnu.org writes: Hello, Hello, [...] Some vnet interfaces are tagged with random like VLAN ID in “ovs-vsctl show” and some are tagged at all, the later are in trunk mode which is a problem. Sorry for the noise, in fact the networks have VLAN=no, so no dynamic VLAN ID are generated :-/ Regards. -- Daniel Dehennin Récupérer ma clef GPG: gpg --keyserver pgp.mit.edu --recv-keys 0x7A6FE2DF pgpvZ_T5caHd9.pgp Description: PGP signature ___ Users mailing list Users@lists.opennebula.org http://lists.opennebula.org/listinfo.cgi/users-opennebula.org
[one-users] [4.2] [Sunstone] oneadmin does not see any VMs
Hello, In sunstone, the oneadmin user do not see any VMs in the vm-tab. It works With the CLI, so the problem looks to be specific to sunstone. I should have done something but can't figure what, do you have any hints on what should I check? Regards. -- Daniel Dehennin Récupérer ma clef GPG: gpg --keyserver pgp.mit.edu --recv-keys 0x7A6FE2DF pgpW3Jpr17e8Y.pgp Description: PGP signature ___ Users mailing list Users@lists.opennebula.org http://lists.opennebula.org/listinfo.cgi/users-opennebula.org
Re: [one-users] [4.2] [Sunstone] oneadmin does not see any VMs
Stefan Kooman ste...@bit.nl writes: Quoting Daniel Dehennin (daniel.dehen...@baby-gnu.org): Hello, In sunstone, the oneadmin user do not see any VMs in the vm-tab. It works With the CLI, so the problem looks to be specific to sunstone. I should have done something but can't figure what, do you have any hints on what should I check? Have you recently upgraded from one 4.2 - 4.4? No, we are lagging behind, we install 3.8 one week before 4.0 was out, upgrade to 4.2 1 month before 4.4 was out... Do you have the following in your sunstone-server.conf: # Default language setting :lang: en_US # Default table order :table_order: desc If not, add them and restart sunstone (if you're running 4.4 that is). No as it's 4.2. The last things I did was to create a dedicated view for a user group to add informations like VM logs etc. I modify some ACLs but: - oneadmin group seems not concerned by them - it works in CLI Regards. -- Daniel Dehennin Récupérer ma clef GPG: gpg --keyserver pgp.mit.edu --recv-keys 0x7A6FE2DF pgpujoJyqWZz9.pgp Description: PGP signature ___ Users mailing list Users@lists.opennebula.org http://lists.opennebula.org/listinfo.cgi/users-opennebula.org
[one-users] Booting stopped VMs a second time after an error cause problem with CDROM link
Hello, On ONE 4.2, testing why the stop did not work[1]: 1. start a non persistent VM 2. stop it = successfully copied to frontend datastores/0 3. resume the VM, transfer failed = the VM state is back to STOPPED 4. fix the transfer problem[2] 5. resume the VM = [TM][I]: Command execution fail: /var/lib/one/remotes/tm/ssh/context /var/lib/one/vms/1565/context.sh grichka:/var/lib/one//datastores/0/1565/disk.1 1565 0 [TM][I]: context: Generating context block device at grichka:/var/lib/one//datastores/0/1565/disk.1 [TM][E]: context: Command ln -s /var/lib/one/datastores/0/1565/disk.1 /var/lib/one/datastores/0/1565/disk.1.iso failed: ln: unable to create symlink “/var/lib/one/datastores/0/1565/disk.1.iso”: file exists [TM][E]: Error creating ISO symbolic link [TM][I]: ExitCode: 1 [TM][E]: Error executing image transfer script: Error creating ISO symbolic link [DiM][I]: New VM state is FAILED Maybe a clean should be done on host before resuming from STOPPED? Regards. Footnotes: [1] each node must be able to SSH password-less to them self [2] was a host key verification failed -- Daniel Dehennin Récupérer ma clef GPG: gpg --keyserver pgp.mit.edu --recv-keys 0x7A6FE2DF pgpBH4Dg_fWBV.pgp Description: PGP signature ___ Users mailing list Users@lists.opennebula.org http://lists.opennebula.org/listinfo.cgi/users-opennebula.org
Re: [one-users] Booting stopped VMs a second time after an error cause problem with CDROM link
Javier Fontan jfon...@opennebula.org writes: This bug is solved in one 4.4: http://dev.opennebula.org/issues/2462 Thanks a lot and sorry for the noise, I'll finish by subscribe to some RSS feed to be aware ;-) Regards. -- Daniel Dehennin Récupérer ma clef GPG: gpg --keyserver pgp.mit.edu --recv-keys 0x7A6FE2DF pgpp_xjFtvxXi.pgp Description: PGP signature ___ Users mailing list Users@lists.opennebula.org http://lists.opennebula.org/listinfo.cgi/users-opennebula.org
[one-users] Monitoring of a host can fail during a VM shutdown
Hello, On a ONE 4.2, we just encountered an transiant issue: [InM][I]: Monitoring host grichka (9) [InM][I]: Command execution fail: 'if [ -x /var/tmp/one/im/run_probes ]; then /var/tmp/one/im/run_probes kvm 9 grichka; else exit 42; fi' [InM][I]: error: failed to get domain 'one-1547' [InM][I]: error: Domain not found: no domain with matching name 'one-1547' [InM][I]: ../../vmm/kvm/poll:70:in `block in get_vm_info': undefined method `[]' for nil:NilClass (NoMethodError) [InM][I]: from ../../vmm/kvm/poll:68:in `each' [InM][I]: from ../../vmm/kvm/poll:68:in `get_vm_info' [InM][I]: from ../../vmm/kvm/poll:35:in `get_all_vm_info' [InM][I]: from ../../vmm/kvm/poll:401:in `print_all_vm_template' [InM][I]: from ../../vmm/kvm/poll:440:in `main' [InM][E]: Error executing poll.sh [InM][I]: ExitCode: 1 [ONE][E]: Error monitoring Host grichka (9): Error executing poll.sh It looks like something list all the running VM and then call “get_vm_info” for each one. This result in an error if the VM disappears in the meantime, because of a shutdown for example. Is this something plausible, in which case I'll open an issue on the bug tracker? Regards. -- Daniel Dehennin Récupérer ma clef GPG: gpg --keyserver pgp.mit.edu --recv-keys 0x7A6FE2DF pgpkL8XquZvt6.pgp Description: PGP signature ___ Users mailing list Users@lists.opennebula.org http://lists.opennebula.org/listinfo.cgi/users-opennebula.org
Re: [one-users] Controlling KVM VMs from network
Jaime Melis jme...@opennebula.org writes: Hi, I just found out about this, so you might be able to use QMP after all, via Libvirt: $ virsh help qemu-monitor-command Great, I'm not sure if several “control” monitors could be used at the same time. If not, it will interfere with ONE management of VMs :-/ Regards. -- Daniel Dehennin Récupérer ma clef GPG: gpg --keyserver pgp.mit.edu --recv-keys 0x7A6FE2DF pgpWpXymJjWYP.pgp Description: PGP signature ___ Users mailing list Users@lists.opennebula.org http://lists.opennebula.org/listinfo.cgi/users-opennebula.org
Re: [one-users] Controlling KVM VMs from network
Jaime Melis jme...@opennebula.org writes: Oh, I see what you mean. Well, it depends on what type of operations you want to execute. If you stop the virtual machine, OpenNebula will move it to UNKNOWN state, for example. But if you use it to manipulate parameters inside the virtual machine, OpenNebula will probably not care about that. Can you elaborate a bit more on what you are planning to do? We plan to build an automatic test framework based on jenkins.debian.net, but managing VMs with ONE. We are looking at what's the best to send keys to VMs, mostly to select a boot entry of an ISO image. Using vncdo seems to not provide feedback, unlike QMP. Regards. -- Daniel Dehennin Récupérer ma clef GPG: gpg --keyserver pgp.mit.edu --recv-keys 0x7A6FE2DF pgpN0ddW33v9Y.pgp Description: PGP signature ___ Users mailing list Users@lists.opennebula.org http://lists.opennebula.org/listinfo.cgi/users-opennebula.org
[one-users] Controlling KVM VMs from network
Hello, We are trying to build an automated integration test environment and are facing one issue: controlling KVM VMs. Looking around, we saw that jenkins.debian.net use “vncdo” to send keys to automated their qemu based tests[1]. Another solution could be the use of the Qemu Machine Protocol[2]. Is is possible do control VMs in this way with ONE? Regards. Footnotes: [1] http://anonscm.debian.org/gitweb/?p=users/holger/jenkins.debian.net.git;a=blob;f=bin/g-i-installation.sh;hb=HEAD [2] http://wiki.qemu.org/QMP -- Daniel Dehennin Récupérer ma clef GPG: gpg --keyserver pgp.mit.edu --recv-keys 0x7A6FE2DF pgpu74uJbCvHF.pgp Description: PGP signature ___ Users mailing list Users@lists.opennebula.org http://lists.opennebula.org/listinfo.cgi/users-opennebula.org
[one-users] Performance with system datastore on NFS
Hello, I just finalize the migration our 3.8.3 ONE to 4.2 and was forced to change my plan to put datastore 0 on NFS4 because of slow accesses. For non persistent images they use local datastore 0, but persistent ones use NFS (via the symlink). All my images are qcow2. I have a dedicated VLAN for storage access, and my nodes are mounting the datastores as: 10.255.255.2:/one-datastores on /var/lib/one/datastores type nfs4 (rw,relatime,vers=4,rsize=262144,wsize=262144,namlen=255,hard,proto=tcp,port=0,timeo=600,retrans=2,sec=sys,clientaddr=10.255.255.4,minorversion=0,fsc,local_lock=none,addr=10.255.255.2,_netdev) I remember seeing a document about different cache scenarios and performance but I can not remember where. My setup is a tree nodes ONE: - one not very powerfull frontend - two quite powerful nodes to run VMs (core i7 + 16Go RAM) My storage is what they call “workgroup NAS” with dual gigabit nics configured in bonding. I'm wondering about using RAID1+0 instead of the RAID5 to improve disk access performances and lower CPU usage. Any hints or idea? Regards. -- Daniel Dehennin Récupérer ma clef GPG: gpg --keyserver pgp.mit.edu --recv-keys 0x7A6FE2DF pgpXQ2K0JUWS7.pgp Description: PGP signature ___ Users mailing list Users@lists.opennebula.org http://lists.opennebula.org/listinfo.cgi/users-opennebula.org
[one-users] Tricks to move images between datastores
Hello, I added a request[1] to move images between datastores, but for now, I'm migrating a 3.8 to 4.2 and wonder if someone has a tricks to move all my CDROM images from 1 to 101. Regards. Footnotes: [1] http://dev.opennebula.org/issues/2417 -- Daniel Dehennin Récupérer ma clef GPG: gpg --keyserver pgp.mit.edu --recv-keys 0x7A6FE2DF pgpKO7BjdVbzv.pgp Description: PGP signature ___ Users mailing list Users@lists.opennebula.org http://lists.opennebula.org/listinfo.cgi/users-opennebula.org
[one-users] [SOLVED] Re: Context CDROM seen as hard drive
Javier Fontan jfon...@opennebula.org writes: Do not use virtio as a device driver for the CDROM as it does not emulate a CD. Use IDE (hda, for example). I finally found the solution by trying, I set in /etc/one/oned.conf: DEFAULT_DEVICE_PREFIX = vd I switch to: DEFAULT_DEVICE_PREFIX = hd and it works. I must admit I'm a bit confused between some settings in /etc/ and some in /var/lib/one/remotes, I need to study more ;-) Regards. -- Daniel Dehennin Récupérer ma clef GPG: gpg --keyserver pgp.mit.edu --recv-keys 0x7A6FE2DF pgpoUGZ0bbCpE.pgp Description: PGP signature ___ Users mailing list Users@lists.opennebula.org http://lists.opennebula.org/listinfo.cgi/users-opennebula.org
[one-users] Context CDROM seen as hard drive
Hello, Running the debian installer with some preseed under ONE 4.2/Wheezy[1] gives me some troubles during partitioning. Here is my deployment.0: #+begin_src xml domain type='kvm' xmlns:qemu='http://libvirt.org/schemas/domain/qemu/1.0' nameone-21/name vcpu2/vcpu cputune shares512/shares /cputune memory524288/memory os type arch='x86_64'hvm/type boot dev='hd'/ /os devices emulator/usr/bin/kvm/emulator disk type='file' device='disk' source file='/var/lib/one//datastores/0/21/disk.0'/ target dev='vda'/ driver name='qemu' type='qcow2' cache='none' io='threads'/ /disk disk type='file' device='cdrom' source file='/var/lib/one//datastores/0/21/disk.1'/ target dev='vdb'/ readonly/ driver name='qemu' type='raw'/ /disk interface type='bridge' source bridge='baby-gnu'/ virtualport type='openvswitch'/ mac address='02:00:0a:02:00:00'/ model type='virtio'/ /interface graphics type='vnc' listen='0.0.0.0' port='5921' keymap='fr'/ /devices features acpi/ /features /domain #+end_src The disk.1 is seen by Linux as a disk: root@wheezy:/home/system# udevadm info -q env -p /block/vdb DEVLINKS=/dev/disk/by-label/CONTEXT /dev/disk/by-path/pci-:00:05.0-virtio-pci-virtio2 DEVNAME=/dev/vdb DEVPATH=/devices/pci:00/:00:05.0/virtio2/block/vdb DEVTYPE=disk ID_FS_LABEL=CONTEXT ID_FS_LABEL_ENC=CONTEXT ID_FS_TYPE=iso9660 ID_FS_USAGE=filesystem ID_FS_VERSION=Joliet\x20Extension ID_PATH=pci-:00:05.0-virtio-pci-virtio2 ID_PATH_TAG=pci-_00_05_0-virtio-pci-virtio2 MAJOR=254 MINOR=16 SUBSYSTEM=block UDEV_LOG=3 USEC_INITIALIZED=2487053 It's missing the “ID_CDROM=1”. The Debian installer is then not completely automatic since it ask to select which disk to use. Any hints to make it appear as a CDROM? Regards. Footnotes: [1] libvirt-bin version 0.9.12-11+deb7u4 -- Daniel Dehennin Récupérer ma clef GPG: gpg --keyserver pgp.mit.edu --recv-keys 0x7A6FE2DF pgpK6e6KaoIbs.pgp Description: PGP signature ___ Users mailing list Users@lists.opennebula.org http://lists.opennebula.org/listinfo.cgi/users-opennebula.org
Re: [one-users] VM in Opennebula for Esxi 5.1 failed
Catalina Quinde catalinaqui...@gmail.com writes: First of all hello, I do not have proprietary system so I'll only give you information I have. Please help me,I created an OpenNebula cluster in which I have added the host Esxi, I've also created two datastores, one for system and one for images, with features: ID 100 system TM_MAD = vmfs ID 101 image DS_MAD = vmfs TM_MAD = vmfs These two datastores I have added to the cluster created, the comand onedatastore list indicates CAPACITY datastore = 0M And in the path / var / lib / one / datastores here are not these two datastores as directories. Reading the documentation[1], I understand that: 1. oneadmin must be able to SSH ESX servers without password[2] 2. nothing is under /var/lib/one/datastores of the OpenNebula frontend because it's managed by ESX servers, ONE access to VMFS datastores though API Maybe if you describe a little more your setup will help people helping you ;-) I'm using OpenNebula Esxi 4.2 and 5.1. I suppose you mean that you use: - OpenNebula 4.2 - ESXi 5.1 Right? Regards. Footnotes: [1] http://opennebula.org/documentation:rel4.2:vmware_ds [2] http://opennebula.org/documentation:rel4.2:hostsubsystem -- Daniel Dehennin Récupérer ma clef GPG: gpg --keyserver pgp.mit.edu --recv-keys 0x7A6FE2DF pgpgYGX1b1Lfl.pgp Description: PGP signature ___ Users mailing list Users@lists.opennebula.org http://lists.opennebula.org/listinfo.cgi/users-opennebula.org
[one-users] KVM process at 99% CPU
Hello, On our 3.8 dual-node ONE I got this morning a OVERLOADED node with load average of 35. I found one my own VMs eating 99% CPU, it was just booted and waiting at ISO boot menu prompt. It looks to be related to KVM as I reproduce it on a 4.2. I can reproduce it by defining a template using an Ubuntu Server AMD64 ISO and instanciate it. I just let the VMs at ISO boot menu, and the KVM process quickly eat 99% of CPU. If someone has an idea. Here is the template: oneadmin@nebula:~$ onetemplate show 17 TEMPLATE 17 INFORMATION ID : 17 NAME : Install ISO AMD64 20G USER : eole GROUP : oneadmin REGISTER TIME : 10/11 19:29:17 PERMISSIONS OWNER : um- GROUP : --- OTHER : --- TEMPLATE CONTENTS CONTEXT=[ NETWORK=YES, SSH_PUBLIC_KEY=$USER[SSH_PUBLIC_KEY] ] CPU=0.2 DISK=[ IMAGE_ID=0 ] DISK=[ CACHE=none, IMAGE_ID=43 ] FEATURES=[ ACPI=yes, PAE=no ] GRAPHICS=[ KEYMAP=fr, LISTEN=0.0.0.0, TYPE=VNC ] INPUT=[ BUS=usb, TYPE=tablet ] MEMORY=768 NIC=[ MODEL=virtio, NETWORK_ID=1 ] OS=[ ARCH=x86_64, BOOT=hd ] VCPU=2 And here the ISO: oneadmin@nebula:~$ oneimage show 43 IMAGE 43 INFORMATION ID : 43 NAME : Ubuntu 13.10 server AMD64 USER : eole GROUP : oneadmin DATASTORE : ISO TYPE : CDROM REGISTER TIME : 11/07 16:29:06 PERSISTENT : No SOURCE : /var/lib/one//datastores/100/39670c51a99879ca3e221d5fc8027dce PATH : /var/tmp/thin-body20131107-1782-yqfomp SIZE : 672M STATE : used RUNNING_VMS: 1 PERMISSIONS OWNER : um- GROUP : --- OTHER : --- IMAGE TEMPLATE DEV_PREFIX=hd DRIVER=raw VIRTUAL MACHINES ID USER GROUPNAMESTAT UCPUUMEM HOSTTIME 117 eole oneadmin userver-testrunn 99768M nebula 0d 00h06 Regards. -- Daniel Dehennin Récupérer ma clef GPG: gpg --keyserver pgp.mit.edu --recv-keys 0x7A6FE2DF pgp2fs2gWG7nw.pgp Description: PGP signature ___ Users mailing list Users@lists.opennebula.org http://lists.opennebula.org/listinfo.cgi/users-opennebula.org
Re: [one-users] Missing log when no error happens
Ruben S. Montero rsmont...@opennebula.org writes: So, your proposal is to have, all messages as in: [TM][I]: clone: Cloning /var/lib/one/datastores/1/e72972d1dc14203d83520f9ae6395d94 in yggdrasil:/var/lib/one//datastores/0/15/disk.1 This is pre-pend with the clone: string, right? If this is the case, this could be done in the core as: [TM][I]: ERROR IN CLONE [TM][I]: Command execution fail: /var/lib/one/remotes/tm/qcow2/clone yggdrasil:/var/lib/one/datastores/1/e72972d1dc14203d83520f9ae6395d94 yggdrasil:/var/lib/one//datastores/0/15/disk.1 15 1 [TM][I]: clone: Cloning /var/lib/one/datastores/1/e72972d1dc14203d83520f9ae6395d94 in yggdrasil:/var/lib/one//datastores/0/15/disk.1 ... [TM][I]: ERROR IN CLONE What do you think? I thought about wrapping all executed commands to log: - INFO: the description of the command, like “Cloning /var/lib/one/datastores/1/e72972d1dc14203d83520f9ae6395d94 in yggdrasil:/var/lib/one//datastores/0/15/disk.1” - DEBUG: the command to execute, like: “ssh@yggdrasil: COMMAND” This could output something like the following (line wrapped to ease reading): [TM][I]: Command: /var/lib/one/remotes/tm/qcow2/clone yggdrasil:/var/lib/one/datastores/1/e72972d1dc14203d83520f9ae6395d94 yggdrasil:/var/lib/one//datastores/0/15/disk.1 15 1 [TM][I]: clone: Cloning /var/lib/one/datastores/1/e72972d1dc14203d83520f9ae6395d94 in yggdrasil:/var/lib/one//datastores/0/15/disk.1 [TM][D]: clone: ssh@yggdrasil cd /var/lib/one/datastores/0/15; qemu-img create -b /var/lib/one/datastores/1/e72972d1dc14203d83520f9ae6395d94 -f qcow2 /var/lib/one/datastores/0/15/disk.1 [TM][E]: ERROR IN CLONE [TM][E]: Command execution fail: /var/lib/one/remotes/tm/qcow2/clone yggdrasil:/var/lib/one/datastores/1/e72972d1dc14203d83520f9ae6395d94 yggdrasil:/var/lib/one//datastores/0/15/disk.1 15 1: qemu-img: Could not open '/var/lib/one/datastores/1/e72972d1dc14203d83520f9ae6395d94' [TM][E]: Error copying yggdrasil:/var/lib/one/datastores/1/e72972d1dc14203d83520f9ae6395d94 to yggdrasil:/var/lib/one//datastores/0/15/disk.1 [TM][I]: ExitCode: 1 [TM][E]: ERROR IN CLONE So, we can follow the execution path until the error: 1. the “qcow2 clone” is called with specified arguments 2. the “clone” script informs what it does 3. in debug mode[1] we see the command that will be executed: an SSH to yggdrasil running “cd” and “qemu-img” 4. an error occurs and we see the command and it's output Actually, the point 4 is already OK for me, but 1 to 3 are missing. The point 2 is already done by the script[2] but is not present in the logs and I don't know why. The point 3 can be done by common wrappers[3]. The point 1 is from core, I think. This could be generalized as: 1. the core log state changes 2. for each step to setup a VM, the core log what need to be done, like “setting up network”, “setting up storage”, etc. 3. the core log which script it will call with its arguments 4. the script log informative messages of what it does 5. before a command is executed, a log of what will be executed and where 6. after the command is executed, log the result Looking at my log files, not everything is logged, mostly the “what ONE wants to/will do”. I hope it will help understand what I would like. Regards. Footnotes: [1] or info I don't know what's the best [2] http://dev.opennebula.org/projects/opennebula/repository/revisions/release-4.2/entry/src/tm_mad/qcow2/clone#L69 [3] http://dev.opennebula.org/projects/opennebula/repository/revisions/release-4.2/entry/src/mad/sh/scripts_common.sh -- Daniel Dehennin Récupérer ma clef GPG: gpg --keyserver pgp.mit.edu --recv-keys 0x7A6FE2DF pgpxzNFDu5fMe.pgp Description: PGP signature ___ Users mailing list Users@lists.opennebula.org http://lists.opennebula.org/listinfo.cgi/users-opennebula.org
[one-users] Missing log when no error happens
Hello, I made some search without success, I'm asking here before creating any redmine issue. Using OpenNebula 4.2, I just see that there is no TM log on successful VMs setup: oneadmin@yggdrasil:~$ cut -d' ' -f6- /var/log/one/16.log [DiM][I]: New VM state is ACTIVE. [LCM][I]: New VM state is PROLOG. [LCM][I]: New VM state is BOOT [VMM][I]: Generating deployment file: /var/lib/one/vms/16/deployment.0 [VMM][I]: ExitCode: 0 [VMM][I]: Successfully execute network driver operation: pre. [VMM][I]: ExitCode: 0 [VMM][I]: Successfully execute virtualization driver operation: deploy. [VMM][I]: post: Executed sudo /usr/bin/ovs-vsctl set Port vnet0 tag=20. [VMM][I]: post: Executed sudo /usr/bin/ovs-ofctl add-flow baby-gnu in_port=5,dl_src=02:00:0a:02:00:00,priority=4,actions=normal. [VMM][I]: post: Executed sudo /usr/bin/ovs-ofctl add-flow baby-gnu in_port=5,priority=39000,actions=drop. [VMM][I]: ExitCode: 0 [VMM][I]: Successfully execute network driver operation: post. [LCM][I]: New VM state is RUNNING I generate an error to see what's comming in the logs: oneadmin@yggdrasil:~$ cut -d' ' -f6- /var/log/one/15.log [DiM][I]: New VM state is ACTIVE. [LCM][I]: New VM state is PROLOG. [TM][I]: Command execution fail: /var/lib/one/remotes/tm/qcow2/clone yggdrasil:/var/lib/one/datastores/1/e72972d1dc14203d83520f9ae6395d94 yggdrasil:/var/lib/one//datastores/0/15/disk.1 15 1 [TM][I]: clone: Cloning /var/lib/one/datastores/1/e72972d1dc14203d83520f9ae6395d94 in yggdrasil:/var/lib/one//datastores/0/15/disk.1 [TM][E]: clone: Command cd /var/lib/one/datastores/0/15; qemu-img create -b /var/lib/one/datastores/1/e72972d1dc14203d83520f9ae6395d94 -f qcow2 /var/lib/one/datastores/0/15/disk.1 failed: qemu-img: Could not open '/var/lib/one/datastores/1/e72972d1dc14203d83520f9ae6395d94' [TM][E]: Error copying yggdrasil:/var/lib/one/datastores/1/e72972d1dc14203d83520f9ae6395d94 to yggdrasil:/var/lib/one//datastores/0/15/disk.1 [TM][I]: ExitCode: 1 [TM][E]: Error executing image transfer script: Error copying yggdrasil:/var/lib/one/datastores/1/e72972d1dc14203d83520f9ae6395d94 to yggdrasil:/var/lib/one//datastores/0/15/disk.1 [DiM][I]: New VM state is FAILED [DiM][I]: New VM state is DONE. [TM][W]: Ignored: TRANSFER SUCCESS 15 - My log configuration is: LOG = [ system = file, debug_level = 3 ] I think it's a problem, each step should be clearly identified with the command it runs. It can be hard to reproduce an error, specially when the user delete the VM in “FAILED” state and try 2 or 3 times before asking the ONE admin. For the logs above, I don't know why the clone log[1] are not reported. Regards. Footnotes: [1] http://dev.opennebula.org/projects/opennebula/repository/revisions/release-4.2/entry/src/tm_mad/qcow2/clone#L69 -- Daniel Dehennin Récupérer ma clef GPG: gpg --keyserver pgp.mit.edu --recv-keys 0x7A6FE2DF pgpX5Upy1CEOP.pgp Description: PGP signature ___ Users mailing list Users@lists.opennebula.org http://lists.opennebula.org/listinfo.cgi/users-opennebula.org
[one-users] OpenNebula at Fosdem?
Hello, I wonder if the OpenNebula community will be present at FOSDEM 2014[1]? I just looks at dates and the deadline for stands and lightning talks is not passed. Regards. Footnotes: [1] https://fosdem.org/2014/news/2013-09-17-call-for-participation-part-two/ -- Daniel Dehennin Récupérer ma clef GPG: gpg --keyserver pgp.mit.edu --recv-keys 0x7A6FE2DF pgpftC_QffSZp.pgp Description: PGP signature ___ Users mailing list Users@lists.opennebula.org http://lists.opennebula.org/listinfo.cgi/users-opennebula.org
[one-users] OneFlow and “infrastructure libraries”
Hello, I would like to manage some kind of “infrastructure libraries” with VMs templates and networks. For example, an “infrastructure” could declare: - 1 network called “internet”, shared between all infrastructures to have network access with outside - 1 network called “DMZ” - 1 network called “LAN” - 1 “router” VMs with: * its first interface on “internet”, configured as DHCP or contextualized * its second interface on “DMZ” * its third interface on “LAN” - X “server” VMs with one interface on “DMZ” - Y “workstation” VMs with one interface on “LAN” When a user instanciate this “infrastructure”: - 2 new networks are created: * DMZ-UNIQ NUMBER * LAN-UNIQ NUMBER - a temporary clone of “router” is created with: * its first interface network untouched * its second interface network set to “DMZ-UNIQ NUMBER” * its third interface network set to “LAN-UNIQ NUMBER” - 1 VMs is started from that temporary VM template - a temporary clone of “server” is created with its interface network set to “DMZ-UNIQ NUMBER” - X VMs are started from that temporary VM template - a temporary clone of “workstation” is created with its interface network set to “LAN-UNIQ NUMBER” - Y VMs are started from that temporary VM template Do you think OneFlow could be extended for such use case? Or maybe it's already possible to do this but I do not find how? Regards. -- Daniel Dehennin Récupérer ma clef GPG: gpg --keyserver pgp.mit.edu --recv-keys 0x7A6FE2DF pgpi5WDl5xTyF.pgp Description: PGP signature ___ Users mailing list Users@lists.opennebula.org http://lists.opennebula.org/listinfo.cgi/users-opennebula.org
Re: [one-users] Lecture about OpenNebula
Daniel Dehennin daniel.dehen...@baby-gnu.org writes: I'm making a 20mn slideshow about OpenNebula at my work. [...] Hello, I would like to thank you very much, the presentation of our feedback on the few months with OpenNebula 3.8 was a success. Everybody was very interested. The most asked question was “How to migrate our vSphere to OpenNebula?” As we do not manage the proprietary solution and missing ressources[1], we could not give much response but “It looks to one target of OpenNebula, we need to make some tests”. I'll build a handout of our slides[2] to give more documentation than just slides and give you URLs of them if anybody is interested. I'm sorry! I was a “little” in the hurry with the surprise of ONE presentation and failed netiquette as a newbie on this mailing-list by not introducing myself. Here is the correction of my miss-behaviour: Quick and dirty personal curriculum vitae: I started computing in around 1999, thanks to Free Software and I'm more a sysadmin than a coder. Well, I like coding my configurations[3] ;-) I started to work in computing in 2003 as sysadmin. I'm interested in ONE for two reasons: 1. personally to manage my home servers (single node). virt-manager was no more an option ;-) 2. professionally, I'm working for the french ministry of National Education[4] in a group providing Free as in Freedom solution for schools, named EOLE[5] At work, we investigate to install a 2 or 3 nodes ONE in schools, sometime to replace single nodes ESXi, and federate all the nebulas to make a single portal and deploy new VM remotely. As I said, we are just starting, we installed a two nodes 3.8 ONE few months ago to QA tests for our solutions replacing the “unusable for us” vCenter Lab Manager and get some feedbacks. Now the test is successful, we will switch to 4.2 in a week or two. Regards. Footnotes: [1] Financial and humans [2] in french [3] thanks CFEngine, Puppet and the last I'm interested in but did not get the time to learn: SaltStack [4] https://en.wikipedia.org/wiki/Ministry_of_National_Education_%28France%29 [5] https://fr.wikipedia.org/wiki/EOLE_%28Linux%29 -- Daniel Dehennin Récupérer ma clef GPG: gpg --keyserver pgp.mit.edu --recv-keys 0x7A6FE2DF pgpQzo6mWx3FR.pgp Description: PGP signature ___ Users mailing list Users@lists.opennebula.org http://lists.opennebula.org/listinfo.cgi/users-opennebula.org
Re: [one-users] Booting VM from CD Image
Christoph Pleger christoph.ple...@cs.tu-dortmund.de writes: Hello, Hello, [...] After registering the cdrom image, I created a VM Template and included in it a section for the cdrom. That section looks like this: DISK=[ DRIVER=raw, IMAGE=Ubuntu 9.10 CD, IMAGE_UNAME=linux, TARGET=sr0 ] This prevents the VM from starting successfully (without that section, the VM goes to state RUNNING). The error log is: Fri Oct 18 16:31:30 2013 [VMM][I]: Command execution fail: cat EOT | /var/tmp/one/vmm/kvm/deploy /var/lib/one//datastores/0/1319/deployment.0 cloudhost05 1319 cloudhost05 Fri Oct 18 16:31:30 2013 [VMM][I]: error: Failed to create domain from /var/lib/one//datastores/0/1319/deployment.0 Fri Oct 18 16:31:30 2013 [VMM][I]: error: An error occurred, but the cause is unknown I got issue with ISO too, when I set my datastore with TM_MAD=qcow2, the VM could not boot, even when they are in raw mode. I use a dedicated TM_MAD=shared datastore for ISO and everything is fine. IIRC, the TM_MAD=qcow2 make use qemu-img backingstore option which is incompatible with raw images, right? Regards. -- Daniel Dehennin Récupérer ma clef GPG: gpg --keyserver pgp.mit.edu --recv-keys 0x7A6FE2DF pgptGqeZVlRgY.pgp Description: PGP signature ___ Users mailing list Users@lists.opennebula.org http://lists.opennebula.org/listinfo.cgi/users-opennebula.org
Re: [one-users] [Discuss] Fwd: Lecture about OpenNebula
Carlo Daffara carlo.daff...@cloudweavers.eu writes: I address some of the points on openstack/cloudstack/eucalyptus vs opennebula in my presentation at the latest conference; (video here: http://www.youtube.com/watch?feature=player_embeddedv=3C1heF0vwhc ) basically, it's a problem of what cloud architecture is a better fit for your needs, the amount of tinkering necessary to have a working product, and the much smaller codebase of OpenNebula that however provides most of the features of the competing platforms. Of course, this is my personal view. Thanks a lot, I really liked it, it's a very interesting presentation. Now I have another problem, I would like to share it, with the others. I will give the link in my slides but during our seminary we will setup a piratebox: - because there is a presentation about it - it permit to share our slides directly and differently We would like to put OpenNebulaConf 2013 on it, to let people retreive them. All the slides are on slideshare and I can't download them[1], I created an account, and click the “save” button: - it's cumbersome: validating each “save” with an email - it does not permit to download them, it just add a bookmark in my profile Is it possible to download PDF of the slides to share them during our seminary? Regards. Footnotes: [1] The youtube videos are OK, thanks to VideoDownloadHelper ;-) -- Daniel Dehennin Récupérer ma clef GPG: gpg --keyserver pgp.mit.edu --recv-keys 0x7A6FE2DF ___ Users mailing list Users@lists.opennebula.org http://lists.opennebula.org/listinfo.cgi/users-opennebula.org
Re: [one-users] [Discuss] Fwd: Lecture about OpenNebula
Daniel Dehennin daniel.dehen...@baby-gnu.org writes: [...] All the slides are on slideshare and I can't download them[1], I created an account, and click the “save” button: - it's cumbersome: validating each “save” with an email - it does not permit to download them, it just add a bookmark in my profile Sorry for the noises, blocking addons prevent the download by blocking some amazone S3 url ;-) Regards. -- Daniel Dehennin Récupérer ma clef GPG: gpg --keyserver pgp.mit.edu --recv-keys 0x7A6FE2DF ___ Users mailing list Users@lists.opennebula.org http://lists.opennebula.org/listinfo.cgi/users-opennebula.org