Re: [one-users] Info on ONE Instance for testing purpose
Hello, Do you know if such a thing exists ? If no it could be a great feature to be able to simulate ONE answers to API commands instead of having a full working infrastructure. Do you use a Continuous Integration tool in ONE development ? Do you test all features and builds ? Thanks in advance Cyrille Mercredi 02/01/2013 à 21:04 Duverne, Cyrille a écrit: Hello, We are creating a new scalable middleware product, and we need to have a fake RPC server for test purpose to launch commands from our continuous integration server. Is there any available on your side ? Thanks in advance Cyrille ___ Users mailing list Users@lists.opennebula.org http://lists.opennebula.org/listinfo.cgi/users-opennebula.org
[one-users] Cgroup blkio with Opennebula
Hi! We have some problems to integrate cgroup blkio inside Opennebula to manage I/O inputs between VM's and storage with templates. Has someone tried this configuration? Thanks! ___ Users mailing list Users@lists.opennebula.org http://lists.opennebula.org/listinfo.cgi/users-opennebula.org
Re: [one-users] a port named vnet0 already exists
Hi, looks like ovs problem. If you see vnet0 in ovs-vsctl show and no one is using it, you should delete it. regards BB On pon 07 jan 2013 19:37:17 CET, christopher barry wrote: Greetings, I have a brand new test setup of ONE compiled from master. I'm running OpenVPN on the hypervisors, and ovpn is using tap0 interface, subsequently ovpn is also using vmnet0 as it's ovs switch port. Jan 7 13:27:13 hyper-02 ovs-vsctl: 1|vsctl|INFO|Called as /usr/local/bin/ovs-vsctl --timeout=5 -vconsole:warn -- add-port br0 vnet0 -- comment ovs-brcompatd: addif br0 vnet0 Jan 7 13:27:13 hyper-02 ovs-vsctl: 2|vsctl|ERR|cannot create a port named vnet0 because a port named vnet0 already exists on bridge br0 When I try to instantiate a vm, (immediately after booting the hypervisor) it gives the messages above in /var/log/messages, and fails. Am I doing something obviously wrong here? Thanks, -C ___ Users mailing list Users@lists.opennebula.org http://lists.opennebula.org/listinfo.cgi/users-opennebula.org ___ Users mailing list Users@lists.opennebula.org http://lists.opennebula.org/listinfo.cgi/users-opennebula.org
Re: [one-users] unable to revert VMs settings to boot from HD
Sir, I owe you a glass of cold Zlaty Bazant! It worked just as described, many thanks! All the best, Andrzej On 08.01.2013 06:49, Benadik, Jan wrote: Hi, I have had the similar problem with virtio. If I remember, the issue was DEV_PREFIX=sd. Try to change it to DEV_PREFIX=hd, or remove it. Jan Dňa 08.01.2013 01:40, Andrzej wrote / napísal(a): OK, should be a simple but this one bothers me for a couple of days now. Slowly loosing mu mind :-) Basically I was experimenting with Opennebula VMs booting from network and it worked OK. But now regardless of what I do my machines want to boot from network. Tried to change context, no luck. I've wiped out old template, created and imported new disk image, crate new template from the scratch using new image but still, instead of booting from hd VM keeps trying iPXE boot. My context in template definition: CONTEXT=[ SSH_PUBLIC_KEY=ssh-rsa sniped ] CPU=3 DISK=[ BUS=virtio, DEV_PREFIX=sd, DRIVER=raw, IMAGE_ID=11, IMAGE_UNAME=oneadmin, SIZE=11000 ] GRAPHICS=[ LISTEN=0.0.0.0, TYPE=vnc ] INPUT=[ BUS=usb, TYPE=mouse ] MEMORY=15000 NAME=Banana001 NIC=[ MODEL=virtio, NETWORK=BridgeHost10 ] OS=[ ARCH=x86_64, BOOT=hd ] RAW=[ TYPE=kvm ] TEMPLATE_ID=4 VCPU=12 And when I'm checking kvm process on host here are its parameters: /usr/bin/kvm -S -M pc-1.0 -enable-kvm -m 15000 -smp 12,sockets=12,cores=1,threads=1 -name one-53 -uuid cd194922-7c1a-6abd-327f-f37d3e21fc23 -nodefconfig -nodefaults -chardev socket,id=charmonitor,path=/var/lib/libvirt/qemu/one-53.monitor,server,nowait -mon chardev=charmonitor,id=monitor,mode=control -rtc base=utc -no-shutdown -device lsi,id=scsi0,bus=pci.0,addr=0x4 -drive file=/var/lib/one//datastores/0/53/disk.0,if=none,id=drive-scsi0-0-0,format=raw -device scsi-disk,bus=scsi0.0,scsi-id=0,drive=drive-scsi0-0-0,id=scsi0-0-0 -drive file=/var/lib/one//datastores/0/53/disk.1,if=none,media=cdrom,id=drive-ide0-0-0,readonly=on,format=raw -device ide-drive,bus=ide.0,unit=0,drive=drive-ide0-0-0,id=ide0-0-0 -netdev tap,fd=20,id=hostnet0 -device virtio-net-pci,netdev=hostnet0,id=net0,mac=02:00:c0:76:43:09,bus=pci.0,addr=0x3 -usb -device usb-mouse,id=input0 -vnc 0.0.0.0:53 -vga cirrus -device virtio-balloon-pci,id=balloon0,bus=pci.0,addr=0x5 I think I've been through every single kvm, qemu, libvirt and oned config file more that 10 times. I need someone with a fresh look now and would really appreciate any hint :-) Where should I look? What I messed up? Let me know if you need more info. Andrzej ___ Users mailing list Users@lists.opennebula.org http://lists.opennebula.org/listinfo.cgi/users-opennebula.org -- *Ján Beňadik* Managed Services - Solution Design Architect mobile: +421 903 691 634 fax: +421 2 6852 8380 jan.bena...@atos.net mailto://jan.bena...@atos.net Vinohradnicka 6, 971 01 Prievidza www.sk.atos.net http://www.sk.atos.net __ ___ Users mailing list Users@lists.opennebula.org http://lists.opennebula.org/listinfo.cgi/users-opennebula.org
Re: [one-users] User Impersonation with API
Hi, This is done with the server_* drivers, it is explained in detail here [1] The ruby client side is implemented in src/cloud/common [2] Regards [1] http://opennebula.org/documentation:rel3.8:cloud_auth [2] http://dev.opennebula.org/projects/opennebula/repository/show/src/cloud/common?rev=one-3.8 -- Carlos Martín, MSc Project Engineer OpenNebula - The Open-source Solution for Data Center Virtualization www.OpenNebula.org | cmar...@opennebula.org | @OpenNebulahttp://twitter.com/opennebulacmar...@opennebula.org On Mon, Jan 7, 2013 at 3:58 PM, Ricardo Duarte rjt...@hotmail.com wrote: Hi there, I'm currently using the Ruby API to implement a custom cloud frontend. The end user will not log with the OpenNebula credentials, so my frontend will not be able to login with the user credentials. But it does have the oneadmin credentials. What I want to do is to impersonate the user, every time I create a virtual machine. Is there a way to do it? Thanks, Ricardo ___ Users mailing list Users@lists.opennebula.org http://lists.opennebula.org/listinfo.cgi/users-opennebula.org ___ Users mailing list Users@lists.opennebula.org http://lists.opennebula.org/listinfo.cgi/users-opennebula.org
Re: [one-users] Info on ONE Instance for testing purpose
Hi, When we don't need to check the interaction with real infrastructure we use the dummy drivers for hosts (im, vmm, network) and datastores (ds tm) $ onehost create dummyhost -i im_dummy -v vmm_dummy -n dummy $ onedatastore show 0 ... DATASTORE TEMPLATE TM_MAD=dummy $ onedatastore show 1 ... DATASTORE TEMPLATE DS_MAD=dummy TM_MAD=dummy You may need to uncomment some of them in oned.conf. The drivers can be found in /var/lib/one/remotes Regards -- Carlos Martín, MSc Project Engineer OpenNebula - The Open-source Solution for Data Center Virtualization www.OpenNebula.org | cmar...@opennebula.org | @OpenNebulahttp://twitter.com/opennebulacmar...@opennebula.org On Tue, Jan 8, 2013 at 9:36 AM, Duverne, Cyrille cyrille.duve...@euranova.eu wrote: Hello, Do you know if such a thing exists ? If no it could be a great feature to be able to simulate ONE answers to API commands instead of having a full working infrastructure. Do you use a Continuous Integration tool in ONE development ? Do you test all features and builds ? Thanks in advance Cyrille Mercredi 02/01/2013 à 21:04 Duverne, Cyrille a écrit: Hello, We are creating a new scalable middleware product, and we need to have a fake RPC server for test purpose to launch commands from our continuous integration server. Is there any available on your side ? Thanks in advance Cyrille ___ Users mailing list Users@lists.opennebula.org http://lists.opennebula.org/listinfo.cgi/users-opennebula.org ___ Users mailing list Users@lists.opennebula.org http://lists.opennebula.org/listinfo.cgi/users-opennebula.org
[one-users] LVM volume doesn't get deleted after VM is nuked
Dear Community, Yes, it's me again. As per a subject field, I have a weird problem when LVM volume doesn't get deleted after VM is being nuked. It seems that `sudo lvremove` is failing as LVM volume is still open/active. Or maybe I'm barking on the wrong tree... I've got two KVM hosts and both seems affected. Effectively machine gets deleted OK but it leaves LVM volume behind so I have to login and lvremove it manually to free up some space. Not a show-stopper but a bit of inconvenience really. Has anyone else experienced this issue? Here is related excerpt from the logs: Tue Jan 8 11:32:42 2013 [TM][D]: Message received: TRANSFER FAILURE 33 Error deleting /var/lib/one/datastores/0/33/disk.0 Tue Jan 8 11:32:42 2013 [TM][D]: Message received: LOG I 34 Command execution fail: /var/lib/one/remotes/tm/lvm/delete superbanana01:/ var/lib/one//datastores/0/34/disk.0 34 101 Tue Jan 8 11:32:42 2013 [TM][D]: Message received: LOG E 34 delete: Command DEV=$(readlink /var/lib/one/datastores/0/34/disk.0) Tue Jan 8 11:32:42 2013 [TM][D]: Message received: LOG I 34 sudo lvremove -f $DEV failed: Can't remove open logical volume lv-one-4-3 4-0 Tue Jan 8 11:32:42 2013 [TM][D]: Message received: LOG E 34 Error deleting /var/lib/one/datastores/0/34/disk.0 Tue Jan 8 11:32:42 2013 [TM][D]: Message received: LOG I 34 ExitCode: 5 Tue Jan 8 11:32:42 2013 [TM][D]: Message received: TRANSFER FAILURE 34 Error deleting /var/lib/one/datastores/0/34/disk.0 Let me know if you need more info. Many thanks as always for your help. Andrzej ___ Users mailing list Users@lists.opennebula.org http://lists.opennebula.org/listinfo.cgi/users-opennebula.org
Re: [one-users] Info on ONE Instance for testing purpose
Hello, Thanks a lot Carlos for this. So I still need to install OpenNebula somewhere to be able to simulate the API.. Kind regards Cyrille Mardi 08/01/2013 à 12:36 Carlos Martín Sánchez a écrit: Hi, When we don't need to check the interaction with real infrastructure we use the dummy drivers for hosts (im, vmm, network) and datastores (ds tm) $ onehost create dummyhost -i im_dummy -v vmm_dummy -n dummy $ onedatastore show 0 DATASTORE TEMPLATE TM_MAD=dummy $ onedatastore show 1 DATASTORE TEMPLATE DS_MAD=dummy TM_MAD=dummy You may need to uncomment some of them in oned.conf. The drivers can be found in /var/lib/one/remotes Regards -- Carlos Martín, MSc Project Engineer OpenNebula - The Open-source Solution for Data Center Virtualization www.OpenNebula.org [1] | cmar...@opennebula.org | @OpenNebula [2] [3] On Tue, Jan 8, 2013 at 9:36 AM, Duverne, Cyrille wrote: Hello, Do you know if such a thing exists ? If no it could be a great feature to be able to simulate ONE answers to API commands instead of having a full working infrastructure.. Do you use a Continuous Integration tool in ONE development ? Do you test all features and builds ? Thanks in advance Cyrille Mercredi 02/01/2013 à 21:04 Duverne, Cyrille a écrit: Hello, We are creating a new scalable middleware product, and we need to have a fake RPC server for test purpose to launch commands from our continuous integration server.. Is there any available on your side ? Thanks in advance Cyrille ___ Users mailing list Users@lists.opennebula.org http://lists.opennebula.org/listinfo.cgi/users-opennebula.org Links: -- [1] http://www.OpenNebula.org [2] http://twitter.com/opennebula [3] mailto:cmar...@opennebula.org ___ Users mailing list Users@lists.opennebula.org http://lists.opennebula.org/listinfo.cgi/users-opennebula.org
Re: [one-users] ganglia integration and base64 errors
Since version 3.8 the output of poll pushed to ganglia is compressed using zlib before being encoded in base 64. On Fri, Dec 28, 2012 at 12:27 PM, samuel sam...@gmail.com wrote: Hi folks, We've started playing in the integration of opennebula (3.8) and ganglia and we faced an issue in the decoding section. when we launch the poll command, we got the following output: -bash-4.1$ $HOME/var/remotes/vmm/kvm/poll --kvm eJx1jUsKwzAMBfc+hfDewZKMJfs2pvXSSUkcaG/fJG2hH7p8b2DGOQdmGqvDkMEA5HWp51bbNN8ykE8pIe33WPt8zcDeK6vfn6WXXjOUJ+0bJRbVY5e2oYf2JT1d1gwWwxDsEVT+DWIUEcL4VsRNyiiCf5ohKQX8qip/VkmHaM0do/I9ug== and when we try to decode it, there's no human-readable information: -bash-4.1$ $HOME/var/remotes/vmm/kvm/poll --kvm | base64 -d x�u� �0 ��}�b�%��eo 4ˤҦ�oo[ ,���p�5�� m] �֥\k���H�F� �1�Gޗ��^ �w�[E�|^mM0X7���~A b�h×h ŋ#�c:�Dg���PO��=�-bash-4.1$ Can anyone point where the error can be located? Thanks in advance, Samuel. ___ Users mailing list Users@lists.opennebula.org http://lists.opennebula.org/listinfo.cgi/users-opennebula.org -- Javier Fontán Muiños Project Engineer OpenNebula - The Open Source Toolkit for Data Center Virtualization www.OpenNebula.org | jfon...@opennebula.org | @OpenNebula ___ Users mailing list Users@lists.opennebula.org http://lists.opennebula.org/listinfo.cgi/users-opennebula.org
[one-users] backup script for SQLite DB
Hi guys,
Just so my presence on this list is not only marked by begging for help
but also some contributing :-)
Here is a backup script for Opennebula SQlite database in case someone
finds it useful. I run it from cron like that
01 5 * * * /root/bin/onedb.sh 21 |logger
And here goes the script itself, it's based on mysql dump script I found
somewhere on the web:
#!/bin/bash
#--
# This is a simple script to create a snapshot of a Opennebula SQLite DB
# Send bugreports, fixes, enhancements, t-shirts, money, beer pizza
# to andr...@mielnet.pl
#--
# Directory to store backups in
DST=/backup/opennebula
# Any backups older than this will be deleted first
KEEPDAYS=10
DATE=$(date +%Y-%m-%d)
# code
/usr/bin/logger Starting ONEDB Dump.
find ${DST} -type f -mtime +${KEEPDAYS} -exec rm -f {} \;
mkdir -p ${DST}
onedb backup --sqlite /var/lib/one/one.db ${DST}/onedb-${DATE}.backup
tar -czf ${DST}/onedb-${DATE}.tar.gz ${DST}/onedb-${DATE}.backup
rm ${DST}/onedb-${DATE}.backup
/usr/bin/logger OK, all ONEDB dumps done in $DST
#--
Not that I needed a backup but someday, who knows.
Any comments or improvements are welcomed :-)
All the best,
Andrzej
___
Users mailing list
Users@lists.opennebula.org
http://lists.opennebula.org/listinfo.cgi/users-opennebula.org
Re: [one-users] oned failed to start (raspberry pi - arch armv6, opennebula 3.8.1, selfcontained)
Hi, Try executing oned directly instead of one start, it may output more information. Anyway, if it fails before the log is created, you only need to debug what's happening in oned.cc main(), and Nebula.cc start() Regards -- Carlos Martín, MSc Project Engineer OpenNebula - The Open-source Solution for Data Center Virtualization www.OpenNebula.org | cmar...@opennebula.org | @OpenNebulahttp://twitter.com/opennebulacmar...@opennebula.org On Tue, Jan 8, 2013 at 8:23 AM, Paul Mayer equilibriu...@gmail.com wrote: Hi, I have just built opennebula 3.8.1. from source on my raspberry pi (running arch). It's a self-contained install and my .bashrc contains ONE_LOCATION, ONE_AUTH and ONE_XMLRPC set. Also, PATH contains the bin folder of the install. However, if I try to run one start (as oneadmin), I just get: usage: oned [-h] [-v] [-f] oned failed to start That's it, there is no oned.log for me to check under var (only sched.log). The sched.log says: Tue Jan 8 08:21:00 2013 [SCHED][I]: Init Scheduler Log system Tue Jan 8 08:21:00 2013 [SCHED][I]: Starting Scheduler Daemon Scheduler Configuration File DEFAULT_SCHED=POLICY=1 HYPERVISOR_MEM=0.1 LIVE_RESCHEDS=0 MAX_DISPATCH=30 MAX_HOST=1 MAX_VM=300 ONED_PORT=2633 SCHED_INTERVAL=30 Tue Jan 8 08:21:00 2013 [SCHED][I]: Starting scheduler loop... Tue Jan 8 08:21:00 2013 [SCHED][I]: Scheduler loop started. Tue Jan 8 08:21:02 2013 [SCHED][I]: Stopping the scheduler... Tue Jan 8 08:21:02 2013 [SCHED][I]: Scheduler loop stopped. Is there anything apparently wrong in there? I have no idea where to start looking right now. Any help would be much appreciated! Thanks Regards, Paul ___ Users mailing list Users@lists.opennebula.org http://lists.opennebula.org/listinfo.cgi/users-opennebula.org ___ Users mailing list Users@lists.opennebula.org http://lists.opennebula.org/listinfo.cgi/users-opennebula.org
Re: [one-users] Wrong source file path for contextualization in deployment.0
Dear thaibinhtt, It looks like you are using a 5.1 ESX version, which is currently not supported by OpenNebula. In any case, you can install the latest version of libvirt, which includes support for ESX 5.1, and allegedly will make OpenNebula work with it. We would LOVE to hear feedback on OpenNebula + ESX 5.1 ;) Regards, -Tino -- Constantino Vázquez Blanco, PhD, MSc Project Engineer OpenNebula - The Open-Source Solution for Data Center Virtualization www.OpenNebula.org | @tinova79 | @OpenNebula ___ Users mailing list Users@lists.opennebula.org http://lists.opennebula.org/listinfo.cgi/users-opennebula.org
Re: [one-users] a port named vnet0 already exists
Thanks for reply. I've disabled OpenVPN for the moment to try to isolate the issue. I manually removed vnet0 from the br0 switch, but still the vm will not start. I can start the vm manually from the fe using virt-manager though, so I'm not sure if it's a template issue. I'll try some more tests and reply with findings. Also, re: ovs: is there an easy way to 'see' who is using a specific switch port? Thanks, -C On Tue, 2013-01-08 at 10:05 +0100, Blaž Balon wrote: Hi, looks like ovs problem. If you see vnet0 in ovs-vsctl show and no one is using it, you should delete it. regards BB On pon 07 jan 2013 19:37:17 CET, christopher barry wrote: Greetings, I have a brand new test setup of ONE compiled from master. I'm running OpenVPN on the hypervisors, and ovpn is using tap0 interface, subsequently ovpn is also using vmnet0 as it's ovs switch port. Jan 7 13:27:13 hyper-02 ovs-vsctl: 1|vsctl|INFO|Called as /usr/local/bin/ovs-vsctl --timeout=5 -vconsole:warn -- add-port br0 vnet0 -- comment ovs-brcompatd: addif br0 vnet0 Jan 7 13:27:13 hyper-02 ovs-vsctl: 2|vsctl|ERR|cannot create a port named vnet0 because a port named vnet0 already exists on bridge br0 When I try to instantiate a vm, (immediately after booting the hypervisor) it gives the messages above in /var/log/messages, and fails. Am I doing something obviously wrong here? Thanks, -C ___ Users mailing list Users@lists.opennebula.org http://lists.opennebula.org/listinfo.cgi/users-opennebula.org ___ Users mailing list Users@lists.opennebula.org http://lists.opennebula.org/listinfo.cgi/users-opennebula.org
Re: [one-users] Cgroup blkio with Opennebula
Emilio, I am gearing up right now myself for testing this very thing. the iotune libvirt domain xml setting looks interesting. If that's not doable, then the hooks infrastructure may be the way to go. keep me in the loop on your findings and I will as well. I'm going to be hitting central storage, so I'll need to do a bit more integration beyond host-based guest throttling. Any and all ideas on this welcome. -C On Tue, 2013-01-08 at 09:41 +0100, Emilio Moreno wrote: Hi! We have some problems to integrate cgroup blkio inside Opennebula to manage I/O inputs between VM's and storage with templates. Has someone tried this configuration? Thanks! ___ Users mailing list Users@lists.opennebula.org http://lists.opennebula.org/listinfo.cgi/users-opennebula.org ___ Users mailing list Users@lists.opennebula.org http://lists.opennebula.org/listinfo.cgi/users-opennebula.org
Re: [one-users] Cgroup blkio with Opennebula
Here's a post I put to the tgt list about it: http://www.spinics.net/lists/linux-stgt/msg03561.html -C On Tue, 2013-01-08 at 12:48 -0500, christopher barry wrote: Emilio, I am gearing up right now myself for testing this very thing. the iotune libvirt domain xml setting looks interesting. If that's not doable, then the hooks infrastructure may be the way to go. keep me in the loop on your findings and I will as well. I'm going to be hitting central storage, so I'll need to do a bit more integration beyond host-based guest throttling. Any and all ideas on this welcome. -C On Tue, 2013-01-08 at 09:41 +0100, Emilio Moreno wrote: Hi! We have some problems to integrate cgroup blkio inside Opennebula to manage I/O inputs between VM's and storage with templates. Has someone tried this configuration? Thanks! ___ Users mailing list Users@lists.opennebula.org http://lists.opennebula.org/listinfo.cgi/users-opennebula.org ___ Users mailing list Users@lists.opennebula.org http://lists.opennebula.org/listinfo.cgi/users-opennebula.org
[one-users] Changes in the OpenNebula release cycle and QA
We are very happy to inform you that the release processes from OpenNebula and the professional version released by C12G Labs are being merged. This is very good news for the community as it adds a new level of quality testing to the released packages and the bugs found in that process are patched faster. You can get more information in this link: http://blog.opennebula.org/?p=3966 Cheers -- Javier Fontán Muiños Project Engineer OpenNebula - The Open Source Toolkit for Data Center Virtualization www.OpenNebula.org | jfon...@opennebula.org | @OpenNebula ___ Users mailing list Users@lists.opennebula.org http://lists.opennebula.org/listinfo.cgi/users-opennebula.org
Re: [one-users] Info on ONE Instance for testing purpose
Yes but dont forget that you can make a self-contained installation on a single folder (-d option http://opennebula.org/documentation:rel3.8:compile). So you can have a selfcontained install with sqlite and dummy drivers withing your project, that you can use for testing purposes. Or your tests can install a fresh instance everytime, then you provide the right oned.conf and sqlite file with the initial testing stateHectorOn Tue, 08 Jan 2013 13:02:55 +0100, Duverne, Cyrille cyrille.duve...@euranova.eu wrote:Hello,Thanks a lot Carlos for this.So I still need to install OpenNebula somewhere to be able to simulate the API.Kind regardsCyrilleMardi 08/01/2013 à 12:36 Carlos Martín Sánchez a écrit:Hi,When we don't need to check the interaction with real infrastructure we use the dummy drivers for hosts (im, vmm, network) and datastores (ds tm)$ onehost create dummyhost -i im_dummy -v vmm_dummy -n dummy $ onedatastore show 0...DATASTORE TEMPLATE TM_MAD="dummy" $ onedatastore show 1...DATASTORE TEMPLATE DS_MAD="dummy"TM_MAD="dummy" You may need to uncomment some of them in oned.conf.The drivers can be found in/var/lib/one/remotesRegards--Carlos Martín, MScProject EngineerOpenNebula -The Open-source Solution for Data Center Virtualization www.OpenNebula.org | cmar...@opennebula.org | @OpenNebula On Tue, Jan 8, 2013 at 9:36 AM, Duverne, Cyrille cyrille.duve...@euranova.eu wrote: Hello,Do you know if such a thing exists ? If no it could be a great feature to be able to simulate ONE answers to API commands instead of having a full working infrastructure. Do you use a Continuous Integration tool in ONE development ? Do you test all features and builds ?Thanks in advanceCyrilleMercredi 02/01/2013 à 21:04 Duverne, Cyrille a écrit: Hello,We are creating a new scalable middleware product, and we need to have a fake RPC server for test purpose to launch commands from our continuous integration server. Is there any available on your side ?Thanks in advanceCyrille ___ Users mailing list Users@lists.opennebula.org http://lists.opennebula.org/listinfo.cgi/users-opennebula.org -- Hector Sanjuan@hecsanjuan___ Users mailing list Users@lists.opennebula.org http://lists.opennebula.org/listinfo.cgi/users-opennebula.org
Re: [one-users] LVM volume doesn't get deleted after VM is nuked
Hi Andrzej I think your problem is related to the delete action problem that was discussed a few weeks ago: http://lists.opennebula.org/pipermail/users-opennebula.org/2012-December/021237.html It is scheduled to be fixed in Release 4.0 : http://dev.opennebula.org/issues/1685 You should try using the VM Cancel action instead of the Delete action when possible. Simon On Tue, Jan 8, 2013 at 6:57 AM, Andrzej andr...@mielnet.pl wrote: Dear Community, Yes, it's me again. As per a subject field, I have a weird problem when LVM volume doesn't get deleted after VM is being nuked. It seems that `sudo lvremove` is failing as LVM volume is still open/active. Or maybe I'm barking on the wrong tree... I've got two KVM hosts and both seems affected. Effectively machine gets deleted OK but it leaves LVM volume behind so I have to login and lvremove it manually to free up some space. Not a show-stopper but a bit of inconvenience really. Has anyone else experienced this issue? Here is related excerpt from the logs: Tue Jan 8 11:32:42 2013 [TM][D]: Message received: TRANSFER FAILURE 33 Error deleting /var/lib/one/datastores/0/33/disk.0 Tue Jan 8 11:32:42 2013 [TM][D]: Message received: LOG I 34 Command execution fail: /var/lib/one/remotes/tm/lvm/delete superbanana01:/ var/lib/one//datastores/0/34/disk.0 34 101 Tue Jan 8 11:32:42 2013 [TM][D]: Message received: LOG E 34 delete: Command DEV=$(readlink /var/lib/one/datastores/0/34/disk.0) Tue Jan 8 11:32:42 2013 [TM][D]: Message received: LOG I 34 sudo lvremove -f $DEV failed: Can't remove open logical volume lv-one-4-3 4-0 Tue Jan 8 11:32:42 2013 [TM][D]: Message received: LOG E 34 Error deleting /var/lib/one/datastores/0/34/disk.0 Tue Jan 8 11:32:42 2013 [TM][D]: Message received: LOG I 34 ExitCode: 5 Tue Jan 8 11:32:42 2013 [TM][D]: Message received: TRANSFER FAILURE 34 Error deleting /var/lib/one/datastores/0/34/disk.0 Let me know if you need more info. Many thanks as always for your help. Andrzej ___ Users mailing list Users@lists.opennebula.org http://lists.opennebula.org/listinfo.cgi/users-opennebula.org ___ Users mailing list Users@lists.opennebula.org http://lists.opennebula.org/listinfo.cgi/users-opennebula.org
Re: [one-users] LVM volume doesn't get deleted after VM is nuked
I swear that I searched list before posting :-) Somehow I missed it, apologies. Many thanks for taking a time and responding to my message. I'll refrain from using delete action for the time being and use cancel method instead. All the best. Andrzej Simon Boulet si...@nostalgeek.com wrote: Hi Andrzej I think your problem is related to the delete action problem that was discussed a few weeks ago: http://lists.opennebula.org/pipermail/users-opennebula.org/2012-December/021237.html It is scheduled to be fixed in Release 4.0 : http://dev.opennebula.org/issues/1685 You should try using the VM Cancel action instead of the Delete action when possible. Simon On Tue, Jan 8, 2013 at 6:57 AM, Andrzej andr...@mielnet.pl wrote: Dear Community, Yes, it's me again. As per a subject field, I have a weird problem when LVM volume doesn't get deleted after VM is being nuked. It seems that `sudo lvremove` is failing as LVM volume is still open/active. Or maybe I'm barking on the wrong tree... I've got two KVM hosts and both seems affected. Effectively machine gets deleted OK but it leaves LVM volume behind so I have to login and lvremove it manually to free up some space. Not a show-stopper but a bit of inconvenience really. Has anyone else experienced this issue? Here is related excerpt from the logs: Tue Jan 8 11:32:42 2013 [TM][D]: Message received: TRANSFER FAILURE 33 Error deleting /var/lib/one/datastores/0/33/disk.0 Tue Jan 8 11:32:42 2013 [TM][D]: Message received: LOG I 34 Command execution fail: /var/lib/one/remotes/tm/lvm/delete superbanana01:/ var/lib/one//datastores/0/34/disk.0 34 101 Tue Jan 8 11:32:42 2013 [TM][D]: Message received: LOG E 34 delete: Command DEV=$(readlink /var/lib/one/datastores/0/34/disk.0) Tue Jan 8 11:32:42 2013 [TM][D]: Message received: LOG I 34 sudo lvremove -f $DEV failed: Can't remove open logical volume lv-one-4-3 4-0 Tue Jan 8 11:32:42 2013 [TM][D]: Message received: LOG E 34 Error deleting /var/lib/one/datastores/0/34/disk.0 Tue Jan 8 11:32:42 2013 [TM][D]: Message received: LOG I 34 ExitCode: 5 Tue Jan 8 11:32:42 2013 [TM][D]: Message received: TRANSFER FAILURE 34 Error deleting /var/lib/one/datastores/0/34/disk.0 Let me know if you need more info. Many thanks as always for your help. Andrzej ___ Users mailing list Users@lists.opennebula.org http://lists.opennebula.org/listinfo.cgi/users-opennebula.org -- Sent from my Android phone with K-9 Mail.___ Users mailing list Users@lists.opennebula.org http://lists.opennebula.org/listinfo.cgi/users-opennebula.org
Re: [one-users] Cgroup blkio with Opennebula
Hi, If you need to test the blkio you can add something like this: RAW = [ TYPE = kvm, DATA = blkiotuneweight100/weight/blkiotune ] to the VM templates. Just in case you missed it, OpenNebula generates cputuneshares for cpu cgroups.This value is based on the CPU template attribute so a VM with CPU=1 will received twice the shares of other VM with CPU=0.5. More info here [1]. We may think to add an specific template attribute for the blkio cgroups driver... Please keep us updated with the results of your tests. Cheers Ruben [1] http://opennebula.org/documentation:rel3.8:kvmg#working_with_cgroups_optional On Tue, Jan 8, 2013 at 9:41 AM, Emilio Moreno emilio.mor...@adam.es wrote: cgroup blkio -- Ruben S. Montero, PhD Project co-Lead and Chief Architect OpenNebula - The Open Source Solution for Data Center Virtualization www.OpenNebula.org | rsmont...@opennebula.org | @OpenNebula ___ Users mailing list Users@lists.opennebula.org http://lists.opennebula.org/listinfo.cgi/users-opennebula.org
Re: [one-users] Instantiate to Persistent
Hi Ricardo, We are missing something here, I think that a persistent image is what you are looking for [1]. If not could you elaborate a bit what are you missing? Cheers Ruben [1] http://opennebula.org/documentation:rel3.8:img_guide#making_images_persistent On Mon, Jan 7, 2013 at 12:32 PM, Ricardo Duarte rjt...@hotmail.com wrote: Hi there, I'm trying to make all my virtual machines persistent. So, I would like disks for vms instantiated from templates to be created as persistent clones, instead of volatile disks on system DS. Is there a way to do it? Thanks, Ricardo ___ Users mailing list Users@lists.opennebula.org http://lists.opennebula.org/listinfo.cgi/users-opennebula.org -- Ruben S. Montero, PhD Project co-Lead and Chief Architect OpenNebula - The Open Source Solution for Data Center Virtualization www.OpenNebula.org | rsmont...@opennebula.org | @OpenNebula ___ Users mailing list Users@lists.opennebula.org http://lists.opennebula.org/listinfo.cgi/users-opennebula.org
Re: [one-users] User Impersonation with API
Hi Carlos, It works if I use the CloudAuth class, thanks. But I kind of hacked this into my code, by copying all the files in CloudAuth and also server_cipher_auth.rb. As far as I can see, there is nothing simpler on the oca API. Is there any chance of implementing the impersonation as a method into OpenNebula::Client? Regards, Ricardo From: cmar...@opennebula.org Date: Tue, 8 Jan 2013 12:24:42 +0100 Subject: Re: [one-users] User Impersonation with API To: rjt...@hotmail.com CC: users@lists.opennebula.org Hi,This is done with the server_* drivers, it is explained in detail here [1]The ruby client side is implemented in src/cloud/common [2] Regards [1] http://opennebula.org/documentation:rel3.8:cloud_auth[2] http://dev.opennebula.org/projects/opennebula/repository/show/src/cloud/common?rev=one-3.8 -- Carlos Martín, MSc Project Engineer OpenNebula - The Open-source Solution for Data Center Virtualizationwww.OpenNebula.org | cmar...@opennebula.org | @OpenNebula On Mon, Jan 7, 2013 at 3:58 PM, Ricardo Duarte rjt...@hotmail.com wrote: Hi there, I'm currently using the Ruby API to implement a custom cloud frontend. The end user will not log with the OpenNebula credentials, so my frontend will not be able to login with the user credentials. But it does have the oneadmin credentials. What I want to do is to impersonate the user, every time I create a virtual machine. Is there a way to do it? Thanks, Ricardo ___ Users mailing list Users@lists.opennebula.org http://lists.opennebula.org/listinfo.cgi/users-opennebula.org ___ Users mailing list Users@lists.opennebula.org http://lists.opennebula.org/listinfo.cgi/users-opennebula.org
