Re: [one-users] DataTables warning message Sunstone
Quoting Daniel Molina (dmol...@opennebula.org): > I have uploaded the following patch and it will be included in one-4.12, > preventing this issue to happen again. > http://dev.opennebula.org/issues/3586 Thanks for the quick response and the fix! Gr. Stefan -- | BIT BV http://www.bit.nl/Kamer van Koophandel 09090351 | GPG: 0xD14839C6 +31 318 648 688 / i...@bit.nl ___ Users mailing list Users@lists.opennebula.org http://lists.opennebula.org/listinfo.cgi/users-opennebula.org
Re: [one-users] DataTables warning message Sunstone
Andreas Calvo schreef op 11 februari 2015 22:36:44 CET: >I'm experimenting something similar. > >Which version and base operating system are you using? Ubuntu Trusty 14.04 LTS, ONE 4.10.1 Gr. Stefan -- BIT BV http://www.bit.nl/Kamer van Koophandel 09090351 | GPG: 0xD14839C6 +31 318 648 688 / i...@bit.nl ___ Users mailing list Users@lists.opennebula.org http://lists.opennebula.org/listinfo.cgi/users-opennebula.org
[one-users] DataTables warning message Sunstone
Hi List, If I log into sunstone (as oneadmin) I receive the following message: DataTables warning: table id=datatable_marketplace - Requested unknown parameter 'files.0.os-arch' for row 27. For more information about this error, please see http://datatables.net/tn/4 Is this (public) "marketplace" related or is something else messed up? Anyone has ever seen this? Gr. Stefan -- | BIT BV http://www.bit.nl/Kamer van Koophandel 09090351 | GPG: 0xD14839C6 +31 318 648 688 / i...@bit.nl pgpoYcY5Z6mqi.pgp Description: PGP signature ___ Users mailing list Users@lists.opennebula.org http://lists.opennebula.org/listinfo.cgi/users-opennebula.org
Re: [one-users] reg sunstone - VNC
Quoting Rajendar K (k.rajen...@gmail.com): > Hi All, >I am using opennebula 4.6 version. My setup has > (i) ONE - Machine A ( sunstone server is also running) > - two nodes z1-nc04, z1-nc03 > > (ii) Sunstone Sever - Machine B, (using the same credential of Machine A) > >the issue i am facing is that i can access the VNC of any VM > using > the sunstone [machine A], but when i access through machine B, it fails. > Both machines are in the same subnet. (using telnet i can see its being > connected) > > The novnc log is as follows, > > WebSocket server settings: > - Listen on :29876 > - Flash security policy server > - No SSL/TLS support (no cert file) > - proxying from :29876 to targets in > /srv/cloud/one/var/sunstone_vnc_tokens > > 1: 10.1.26.51: Plain non-SSL (ws://) WebSocket connection > 1: 10.1.26.51: Version hybi-13, base64: 'False' > 1: 10.1.26.51: Path: '/?token=npfbnzreud4ozaiasiwq' > 1: connecting to: z1-nc04:5995 > 1: handler exception: [Errno -3] Temporary failure in name resolution ^^ This indicates a resolving issue. Do both servers have the same resolvers configured (/etc/resvolv.conf) and or /etc/hosts file with static entries? Gr. Stefan -- | BIT BV http://www.bit.nl/Kamer van Koophandel 09090351 | GPG: 0xD14839C6 +31 318 648 688 / i...@bit.nl ___ Users mailing list Users@lists.opennebula.org http://lists.opennebula.org/listinfo.cgi/users-opennebula.org
Re: [one-users] Cloud View with Persistent Image
Quoting Ruben S. Montero (rsmont...@opennebula.org): > We have the sheduled actions, as you know this feature triggers actions on > an specific date. We could extend these actions to be triggered on states > (i.e. triggered by the scheduler). This, compared with hooks (you can > program actions on given states), allow you to set the actions per VM. Also > users could access this feature, and no modification of oned.conf is > requires. > > Does it make sense? Totally, this would be really nice, and add a lot of flexibility. Gr. Stefan -- | BIT BV http://www.bit.nl/Kamer van Koophandel 09090351 | GPG: 0xD14839C6 +31 318 648 688 / i...@bit.nl ___ Users mailing list Users@lists.opennebula.org http://lists.opennebula.org/listinfo.cgi/users-opennebula.org
[one-users] Refresh button in Sunstone makes graphs disappear in ONE 4.10
Hi, Hitting the refresh button in Sunstone in either the "Capacity" or "Network" view makes the graphs disappear (instead of refreshing them with new data). They return if you click another view in between. Tested with Chromium 38, Apache Passenger 4.0.53. Gr. Stefan -- | BIT BV http://www.bit.nl/Kamer van Koophandel 09090351 | GPG: 0xD14839C6 +31 318 648 688 / i...@bit.nl ___ Users mailing list Users@lists.opennebula.org http://lists.opennebula.org/listinfo.cgi/users-opennebula.org
Re: [one-users] vCenter support prevents seperate Sunstone installations in ONE 4.10
Quoting Daniel Molina (dmol...@opennebula.org): > Hi Stefan, > > Indeed it is a bug, a file is missing in the sunstone package. You can fix > it including the following file in /usr/lib/one/ruby > https://github.com/OpenNebula/one/blob/master/src/vmm_mad/remotes/vcenter/vcenter_driver.rb This is not sufficient, new error is thrown: [ 2014-11-05 11:30:54.4794 21003/7f59add87700 Pool2/Implementation.cpp:287 ]: Could not spawn process for application /usr/lib/one/sunstone: An error occured while starting up the preloader. Error ID: a6f5c45b Error details saved to: /tmp/passenger-error-W31QXw.html Message from application: cannot load such file -- rbvmomi (LoadError) /usr/lib/ruby/1.9.1/rubygems/custom_require.rb:36:in `require' /usr/lib/ruby/1.9.1/rubygems/custom_require.rb:36:in `require' /usr/lib/one/ruby/vcenter_driver.rb:39:in `' /usr/lib/ruby/1.9.1/rubygems/custom_require.rb:36:in `require' /usr/lib/ruby/1.9.1/rubygems/custom_require.rb:36:in `require' /usr/lib/one/sunstone/routes/vcenter.rb:26:in `' /usr/lib/ruby/1.9.1/rubygems/custom_require.rb:36:in `require' /usr/lib/ruby/1.9.1/rubygems/custom_require.rb:36:in `require' /usr/lib/one/sunstone/sunstone-server.rb:323:in `block in ' /usr/lib/one/sunstone/sunstone-server.rb:322:in `each' /usr/lib/one/sunstone/sunstone-server.rb:322:in `' /usr/lib/ruby/1.9.1/rubygems/custom_require.rb:36:in `require' /usr/lib/ruby/1.9.1/rubygems/custom_require.rb:36:in `require' config.ru:10:in `block in ' /usr/lib/ruby/vendor_ruby/rack/builder.rb:55:in `instance_eval' /usr/lib/ruby/vendor_ruby/rack/builder.rb:55:in `initialize' config.ru:1:in `new' config.ru:1:in `' /usr/share/passenger/helper-scripts/rack-preloader.rb:112:in `eval' /usr/share/passenger/helper-scripts/rack-preloader.rb:112:in `preload_app' /usr/share/passenger/helper-scripts/rack-preloader.rb:158:in `' /usr/share/passenger/helper-scripts/rack-preloader.rb:29:in `' /usr/share/passenger/helper-scripts/rack-preloader.rb:28:in `' Message from application: cannot load such file -- rbvmomi (LoadError). It turns out that "/usr/lib/one/ruby/vendors/rbvmomi" is also missing. > > Sorry for the inconvenience Sorry for not having tested this earlier ... Gr. Stefan -- | BIT BV http://www.bit.nl/Kamer van Koophandel 09090351 | GPG: 0xD14839C6 +31 318 648 688 / i...@bit.nl ___ Users mailing list Users@lists.opennebula.org http://lists.opennebula.org/listinfo.cgi/users-opennebula.org
[one-users] vCenter support prevents seperate Sunstone installations in ONE 4.10
Hi, I upgraded two test installs to ONE 4.10. All went fine on the VM with a "all in one" setup (ONE / SUNSTONE on the same server). But on the test installation with opennebula and sunstone in seperate virtual machines, Sunstone throws this error: [ 2014-11-05 09:26:35.1733 6489/7fd61076a700 Pool2/Implementation.cpp:287 ]: Could not spawn process for application /usr/lib/one/sunstone: An error occured while starting up the preloader. Error ID: bf1abc0e Error details saved to: /tmp/passenger-error-Cd1rIQ.html Message from application: cannot load such file -- vcenter_driver (LoadError) /usr/lib/ruby/1.9.1/rubygems/custom_require.rb:36:in `require' /usr/lib/ruby/1.9.1/rubygems/custom_require.rb:36:in `require' /usr/lib/one/sunstone/routes/vcenter.rb:26:in `' /usr/lib/ruby/1.9.1/rubygems/custom_require.rb:36:in `require' /usr/lib/ruby/1.9.1/rubygems/custom_require.rb:36:in `require' /usr/lib/one/sunstone/sunstone-server.rb:323:in `block in ' /usr/lib/one/sunstone/sunstone-server.rb:322:in `each' /usr/lib/one/sunstone/sunstone-server.rb:322:in `' /usr/lib/ruby/1.9.1/rubygems/custom_require.rb:36:in `require' /usr/lib/ruby/1.9.1/rubygems/custom_require.rb:36:in `require' config.ru:9:in `block in ' /usr/lib/ruby/vendor_ruby/rack/builder.rb:55:in `instance_eval' /usr/lib/ruby/vendor_ruby/rack/builder.rb:55:in `initialize' config.ru:1:in `new' config.ru:1:in `' /usr/share/passenger/helper-scripts/rack-preloader.rb:112:in `eval' /usr/share/passenger/helper-scripts/rack-preloader.rb:112:in `preload_app' /usr/share/passenger/helper-scripts/rack-preloader.rb:158:in `' /usr/share/passenger/helper-scripts/rack-preloader.rb:29:in `' /usr/share/passenger/helper-scripts/rack-preloader.rb:28:in `' Most notably "Message from application: cannot load such file -- vcenter_driver (LoadError)". It turns out that if I move "/var/lib/one/remotes/im/vcenter.d, /var/lib/one/remotes/vmm/vcenter" out of the way, Sunstone throws this error. If I comment "- vcenter" in sunstone-server.conf, Sunstone works fine. Is this a (new) dependency from Sunstone for vcenter support? Or a bug? Or did I miss something? Thanks, Stefan -- | BIT BV http://www.bit.nl/Kamer van Koophandel 09090351 | GPG: 0xD14839C6 +31 318 648 688 / i...@bit.nl pgpkVhnY86MXK.pgp Description: PGP signature ___ Users mailing list Users@lists.opennebula.org http://lists.opennebula.org/listinfo.cgi/users-opennebula.org
Re: [one-users] Problem Boot Order Image
Quoting Anandharaj Subramaniam (anandharaj.man...@mimos.my): > Hi, > > How do we set / enable the bootmenu? Below is my current setup in vmm_kvm.conf > > OS = [ boot = "hd", arch = "x86_64" ] That's a RAW (KVM) attribute: RAW = "" By including it in vmm_kvm.conf it's enabled for all (newly) created VM's. Gr. Stefan -- | BIT BV http://www.bit.nl/Kamer van Koophandel 09090351 | GPG: 0xD14839C6 +31 318 648 688 / i...@bit.nl ___ Users mailing list Users@lists.opennebula.org http://lists.opennebula.org/listinfo.cgi/users-opennebula.org
Re: [one-users] Opennebula + OpenVSwitch + FLoodlight with multiple hosts
Jhon Masschelein schreef op 20 oktober 2014 17:26:56 CEST: >Hi, > >I am trying to get OpenNebula to work using OVS + VLAN networking, but >I cannot get VMs on different hosts to talk to each other. > >If I just configure the OVS bridge on the host, set up the OpenNebula >virtual network as an OVS network and add the hosts using the OVS >driver, everything works within a host. Any VMs running on that host >can talk to each other. > >Also, the two hosts themselves can talk to each other using the ip >address assigned to the OVS bridge device. > >But a vm on one host cannot ping a VM on the other host. > >A number of internet resources like >http://openvswitch.org/support/config-cookbooks/vlan-configuration-cookbook/ >would seem to indicate that this should work, but it does not, at least >not for me. > >As a next step, I set up a Floodlight openflow controller. >With the above setup, I assign the controller to the OVSwitches and >almost immediately, all running VMs are able to talk to each other. > >However, any VM's I start after the Floodlight controller has started, >are again confined to their respective hosts. >A restart of the Floodlght controller fixes it again: all VM's (old and >new) are able to talk to each other. > > >Of course, restarting floodlight every time does not seem to be the >right way of doing this. I also reconfigured floodlight to "learning >switch mode" but that did not change anything. >I do see the ports added to the switches when the new VMs come up, but >nothing more. > >Has anybody been able to get this setup to work? Is an openflow >controller like floodlight required to get this to work reliably? >FYI; the floodlight controller is not controlling the switch that is >sitting between the hosts. I am not sure if this is even relevant since >the web page I mentioned above indicates it should all work without a >controller. If it can be made to work without the need for an external >controller like floodlight, that would be totally acceptable. > >If anyone has been able to make a OVS setup work, I would be very >grateful for any information you can pass my way. Sure, this indeed should be working out of the box. Can you send the output of "ovs-ofctl dump-flows bridge-name" of the hypervisors? And "ovs-vsctl show"? Do you make use of the "mac-spoofing" and "IP hijacking prevention" features of ONE /openvswitch? Gr. Stefan ___ Users mailing list Users@lists.opennebula.org http://lists.opennebula.org/listinfo.cgi/users-opennebula.org
Re: [one-users] Problem Boot Order Image
Dian Sukmono Aji schreef op 22 oktober 2014 06:15:56 CEST: >Hai, > >I've Succeed install opennebula on single machine. Next step i create >VM on sunstone by uploading Ubuntu ISO cd source and create empty RAW >Datablock. Second step is trying to install Ubuntu 14.04 on the >datablock. After strugeling with boot order, i finnaly manage to boot >on ISO cd Ubuntu image by forcing the cd to target as hda and force to >boot with : >RAW=[ >DATA=" > > > >", >TYPE="kvm" ] > >had to add above lines on context. > >Now after the install finish i want to create VM with that datablock >but after boot, it said no boot faile : no a bootable disk. > >This is template from VM where i installed the ubuntu to the datablock >: >AUTOMATIC_REQUIREMENTS !(PUBLIC_CLOUD = YES) >CONTEXT >DISK_ID2 >ETH0_DNS 8.8.8.8 >ETH0_GATEWAY 10.1.20.1 >ETH0_IP10.1.20.59 >ETH0_MAC 02:00:0a:01:14:3b >ETH0_MASK 255.255.255.0 >ETH0_NETWORK 10.1.20.0 >NETWORKYES >TARGET hdb >CPU1 >DISK >0 >CLONE NO >CLONE_TARGET SYSTEM >DATASTORE default >DATASTORE_ID 1 >DEV_PREFIX hd >DISK_ID0 >IMAGE ISO Ubuntu Server 14.04.1 LTS >IMAGE_ID 10 >IMAGE_UNAMEonengadmin >LN_TARGET NONE >READONLY YES >SAVE NO >SIZE 572 >SOURCE /var/lib/one//datastores/1/f061e7368909905227e0c02df28d8533 >TARGET hda >TM_MAD shared >TYPE CDROM >1 >CLONE YES >CLONE_TARGET SYSTEM >DATASTORE default >DATASTORE_ID 1 >DEV_PREFIX hd >DISK_ID1 >IMAGE Disk RAW 10GB >IMAGE_ID 14 >IMAGE_UNAMEonengadmin >LN_TARGET NONE >READONLY NO >SAVE NO >SIZE 10240 >SOURCE /var/lib/one//datastores/1/e9c405962e0a842a448c64f550354cc7 >TARGET hdc >TM_MAD shared >TYPE FILE >GRAPHICS >LISTEN 0.0.0.0 >PORT 5934 >TYPE VNC >MEMORY 1024 >NIC >AR_ID 0 >BRIDGE br0 >IP 10.1.20.59 >MAC02:00:0a:01:14:3b >NETWORKpriv-net >NETWORK_ID 0 >NETWORK_UNAME oneadmin >NIC_ID 0 >VLAN NO >RAW >DATA >TYPE kvm >TEMPLATE_ID11 >VMID 34 > >and this is the template from i'm trying to boot the datablock : >AUTOMATIC_REQUIREMENTS !(PUBLIC_CLOUD = YES) >CONTEXT >DISK_ID1 >ETH0_DNS 8.8.8.8 >ETH0_GATEWAY 10.1.20.1 >ETH0_IP10.1.20.60 >ETH0_MAC 02:00:0a:01:14:3c >ETH0_MASK 255.255.255.0 >ETH0_NETWORK 10.1.20.0 >NETWORKYES >TARGET hda >CPU1 >DISK >CLONE YES >CLONE_TARGET SYSTEM >DATASTORE default >DATASTORE_ID 1 >DEV_PREFIX hd >DISK_ID0 >IMAGE Ubuntu 14.04.1 >IMAGE_ID 15 >IMAGE_UNAMEonengadmin >LN_TARGET NONE >READONLY NO >SAVE NO >SIZE 10240 >SOURCE /var/lib/one//datastores/1/dca998a00ad0109e93e23a0588070017 >TARGET hdb >TM_MAD shared >TYPE FILE >GRAPHICS >LISTEN 0.0.0.0 >PORT 5941 >TYPE VNC >MEMORY 1024 >NIC >AR_ID 0 >BRIDGE br0 >IP 10.1.20.60 >MAC02:00:0a:01:14:3c >NETWORKpriv-net >NETWORK_ID 0 >NETWORK_UNAME oneadmin >NIC_ID 0 >VLAN NO >TEMPLATE_ID15 >VMID 41 > >Please inlight me. There is always a contextualization CD-ROM / ISO attached to the VM (for configuring networking, hostname, ssh-keys, etc) which might be "hda" now, and can't be boot from of course. Adjust the template so you boot from disk and re-instantiate. This might be fixed in ONE 4.10: http://dev.opennebula.org/issues/3158 Instead of boot dev=cdrom (RAW) you could set it to bootmenu so you can press F12 during boot and choose which media to boot from. You can make that a global default as well by adding this to "vmm_kvm.conf" (and restart ONE after that). Gr. Stefan ___ Users mailing list Users@lists.opennebula.org http://lists.opennebula.org/listinfo.cgi/users-opennebula.org
Re: [one-users] Cloud View with Persistent Image
Quoting Ruben S. Montero (rsmont...@opennebula.org): > Hi > > These two are on our radar and probably scheduled for the next release. I > totally agree that we need to simplify the provision process even more. > > About feature 1. > > We have two issues: > > [1] Clone a running VM. http://dev.opennebula.org/issues/2637 > [2] Clone a template . http://dev.opennebula.org/issues/2051 > > Last one includes images (pretty much the procedure outlined by Anandharaj > under "New VM request") and I think this will cover exactly your request. > > About feature 2. > > We want to add the ability to update any part of a template for VM, most of > it can be done if the VM is running (eg. attach a new disk or nic). Others > like memory can be resized when the VM is powered off. You would not need > to update and recreate the template. > > [3] http://dev.opennebula.org/issues/2065 It would be nice if some actions could be "suspended" until the desired state has been reached. This would give you the possibility to perform for example resize actions (change (v)CPU, RAM, etc.) and have them be applied later, after the VM changes state: running -> poweroff -> running. Similar to the "Some changes may require a guest shutdown to take effect." message of "virt-manager". You can then schedule a power cycle to have the VM resized when convenient (midnight for example). Sure you could script this as (one)admin but for an end user this is more convenient. Gr. Stefan -- | BIT BV http://www.bit.nl/Kamer van Koophandel 09090351 | GPG: 0xD14839C6 +31 318 648 688 / i...@bit.nl ___ Users mailing list Users@lists.opennebula.org http://lists.opennebula.org/listinfo.cgi/users-opennebula.org
Re: [one-users] existing vmware (free edition) image to kvm image
Quoting Amier Anis (myma...@gmail.com): > Dear Team, > > I need some advise how do i cater this. We have plan to migrate all vmware > vm to kvm vm as our opennebula implementation using kvm. Are these only linux VM's or also Windows? Windows doesn't like changes in hardware. You might have to boot Windows in a special "new hardware, check if you've drivers for them" kinda mode to be able to use the qemu hardware. Make sure you define a qemu machine type with a fixed version. If you upgrade qemu/kvm later on, you will get similar issues you might encounter while migrating from VMware to KVM. Linux doesn't care. At least I haven't run into trouble migrating even older vm's to new "hardware". > > I would like to know how the best i can move this vmware to kvm. > > I my mind, I make the converting from vmware to qcow2 image and then upload > via sunstone as image and run as persistent. Is that the correct way. You can even run VMDK images with qemu but you don't have snapshot possibilty in the future. Uploading them through sunstone will almost certainly not work. I've not been able to upload images > 4 GB. I would scp / cp them to a nfs share that the frontend (opennebula daemon is running on) has access too. Then create an image through sunstone with "path" options, where the path is where your image is located. But, the image will be copied to the final datastore location. I would recommend the following: Create all the images / templates in Sunstone beforehand with correct names, sizes and types. They will be created as qcow2 images and won't occupy any storage. While migrating / converting to qcow2 set the path of the conversion to the path of the image. After the conversion has finished (make sure permissions are those of oneadmin) you will be able to instantiate your vm. Good luck! Gr. Stefan -- | BIT BV http://www.bit.nl/Kamer van Koophandel 09090351 | GPG: 0xD14839C6 +31 318 648 688 / i...@bit.nl ___ Users mailing list Users@lists.opennebula.org http://lists.opennebula.org/listinfo.cgi/users-opennebula.org
[one-users] Password rollover oneamdin
Hi, If you ever do a password rollover for the "oneadmin" user don't forget to restart the scheduler (mm_sched) process. It's using oneadmin credentials to authenticate apparently as I found this in my logs after changing the password: Oct 8 14:39:01 oned mm_sched[2242]: [Z0][POOL][E]: ONE returned error while retrieving pool info: Oct 8 14:39:01 oned mm_sched[2242]: [Z0][POOL][E]: [VirtualMachinePoolInfo] User couldn't be authenticated, aborting call. After the restart all was fine again. Gr. Stefan -- | BIT BV http://www.bit.nl/Kamer van Koophandel 09090351 | GPG: 0xD14839C6 +31 318 648 688 / i...@bit.nl pgpPk_LM7JMOJ.pgp Description: PGP signature ___ Users mailing list Users@lists.opennebula.org http://lists.opennebula.org/listinfo.cgi/users-opennebula.org
Re: [one-users] BOOTPROTO, DHCP_HOSTNAME, and vm IPs in Sunstone
Quoting kerryhall . (kerryh...@gmail.com): > I would definitely not prefer that the vm manage the IP. What I would like > to do upon new vm creation: > > 1. VM gets spun up with BOOTPROTO=dhcp > 2. VM gets spun up with DHCP_HOSTNAME=$NAME > 3. IP given to the VM by the dhcp server on my network is then the IP that > is listed in sunstone for that VM. > > AFAIK: > 1. There is no way to achieve item 1 in ONE out of the box, I have to > either write a bash script that gets run on VM spin up, or create an image > that does this. You don't need to make a VM contextualized to achieve that. Just configure networking in the image to "DHCP". > 2. I can set the _hostname_ via template, but in order to set DHCP_HOSTNAME > again I have to run a bash script on spin up. Why the DHCP_HOSTNAME? If you have a pre-configured list (all MAC's statically configured in the DHCP server) it will get the correct ip-address. You can set the correct HOSTNAME based on the VM template if you want. > 3. There is no way to accomplish item 3, outside of writing a bash script > that runs on my head node that grabs IP data and dumps it into the > database. Certainly this is not preferable. At this point you've already accomplished point 3: VM has the IP-address listed in OpenNebula. The only thing that you don't have is the VM name in the "host" specificiation in the DHCP config. You can however, set the hostname in the DHCP configuration (option host-name "your-vm-name-here") and configure the hostname (point 2) based on this. This might even be a setting in dhclient / dhcpcd to do that automatically. Then you just have to make sure you give the VM the right IP and the right name (in Sunstone). > > These three items taken all together suggest that including the option to > easily support DHCP out of the box would be a nice feature to have. > > I'm working on a bash script currently for 1 and 2, but for item 3 I am out > of luck. > > I hope that makes sense. Let me know if it doesn't and I would be happy to > clarify further. Thank you for your help!! I might completely misunderstand you, sorry if that's the case. Gr. Stefan -- | BIT BV http://www.bit.nl/Kamer van Koophandel 09090351 | GPG: 0xD14839C6 +31 318 648 688 / i...@bit.nl ___ Users mailing list Users@lists.opennebula.org http://lists.opennebula.org/listinfo.cgi/users-opennebula.org
[one-users] # RVMs not correct during live migrate
Hi, I'm currently upgrading to ONE-4.8.0 on a cloud. I'm live migrating a bunch of vm's. While doing that I see that the RVM counter for the Hosts is not correct. All the scheduled VM's are already counted on the hypervisor depicted to run the VM's. After the migration has finished the amount of RVMs is correct. Note that this is not a big issue whatsoever, more an artefact. FYI, Gr. Stefan -- | BIT BV http://www.bit.nl/Kamer van Koophandel 09090351 | GPG: 0xD14839C6 +31 318 648 688 / i...@bit.nl pgpe3A91Pcukk.pgp Description: PGP signature ___ Users mailing list Users@lists.opennebula.org http://lists.opennebula.org/listinfo.cgi/users-opennebula.org
Re: [one-users] Extra IP's not reachable from outside worl
Hi Johan, Quoting Johan Kooijman (m...@johankooijman.com): > Hi, > > I have a VM with 6 IP's. Contextualization configured them for me, it > works. However: all IP's on interfaced other than eth0, are not > reachable from the outside world. > > When I down the ethX interface, add the IP's as an alias on eth0 and > send out an arping from eth0, the IP becomes available. Sending the > arp from the eth1 interface doesn't help. Do you use OpenvSwitch? Do you have arp cache poisoning prevention enabled? See /var/lib/one/remotes/vnm/OpenNebulaNetwork.conf. If so, OpenvSwitch prevents ARP on interfaces other than the one it has OpenFlow rules configured for. See "ovs-ofctl add-flow" logging in oned log to check what ip's have been configured on what interface(s) and/or check "ovs-ofctl dump-flows bridge_name" on the hypervisor. Gr. Stefan -- | BIT BV http://www.bit.nl/Kamer van Koophandel 09090351 | GPG: 0xD14839C6 +31 318 648 688 / i...@bit.nl ___ Users mailing list Users@lists.opennebula.org http://lists.opennebula.org/listinfo.cgi/users-opennebula.org
Re: [one-users] Multiple VLAN and Bridge
Hi, Quoting Amier Anis (myma...@gmail.com): > Dear Stefan, > > It's great ... yes I would like you to share here. I hv tried before but no > luck. > My server have at least 2 physical nic. 1 is management and another one is > storage. I use this little script to setup networking on our hosts (openvswitch should already be installed and running): #!/bin/sh echo "adding bridges" ovs-vsctl add-br storage ovs-vsctl add-br uplink ovs-vsctl add-br mgmt-migr echo "adding bonds" ovs-vsctl add-bond storage bond0 eth0 eth5 bond_mode=balance-tcp lacp=active other_config:lacp-time=fast trunks=list-of-vlan-ids-here ovs-vsctl add-bond uplink bond1 eth1 eth4 bond_mode=balance-tcp lacp=active other_config:lacp-time=fast trunks=list-of-vlan-ids-here ovs-vsctl add-bond mgmt-migr bond2 eth2 eth3 bond_mode=balance-tcp lacp=active other_config:lacp-time=fast trunks=list-of-vlan-ids-here We have all vlans "tagged/trunked" on the port(s)(channels). In order to have vlan interface(s) on the hypervisor itself we create some openvswitch "internal" ports: echo "adding vlans" ovs-vsctl add-port storage vlanVID tag=VID -- set interface vlanID type=internal ovs-vsctl add-port mgmt-migr vlanVID tag=VID -- set interface vlanVID type=internal ovs-vsctl add-port mgmt-migr vlanVID tag=VID -- set interface vlanVID type=internal That way we can configure networking for hypervisor management, vm-migration and storage (nfs) on the hypervisors themselves. It will just show up as a regular interface (ip list). If you don't want / can't make portchannels (bonds/trunkports/teams, whatever you want to call them: 802.3ad) you can add ports to bridges like this: ovs-vsctl add-br bridge-name At this point you got two options. Create a trunk port (tag all vlans: 802.1Q), or provide a list of vlans. If you define a "trunk port" you don't have to add newly created vlans on the openvswitch port if you add new vlans on the physical switch port. trunk port: ovs-vsctl add-port eth0 bridge-name vlan_mode=trunk or Explicitly specify vlans: ovs-vsctl add-port eth0 bridge-name trunks=vlan_id1,vlan_id2 In /etc/network/interfaces (if using Ubuntu/Debian) you have to make sure the physical devices are marked up/down: auto eth0 iface eth0 inet manual pre-up ip link set $IFACE up post-down ip link set $IFACE down auto eth1 iface eth1 inet manual pre-up ip link set $IFACE up post-down ip link set $IFACE down ... for every interface you want to add to openvswitch ... And of course configure networking for the vlan (internal) interface(s). ovs-vsctl show should give you a nice overview of what has been configured. Just add the openvswitch bridge(s) to OpenNebula and off you go. Let me know if I can be of any further assistance. Gr. Stefan -- | BIT BV http://www.bit.nl/Kamer van Koophandel 09090351 | GPG: 0xD14839C6 +31 318 648 688 / i...@bit.nl pgpZx2l7kjQYK.pgp Description: PGP signature ___ Users mailing list Users@lists.opennebula.org http://lists.opennebula.org/listinfo.cgi/users-opennebula.org
Re: [one-users] ONE with MariaDB
Quoting Carlos Martín Sánchez (cmar...@opennebula.org): > Hi, > > On Thu, Sep 11, 2014 at 3:49 AM, Anandharaj Subramaniam < > anandharaj.man...@mimos.my> wrote: > > > Hi, > > > > > > > > Does OpenNebula support MariaDB? > > > > No, we only test it with MySQL. > > My understanding is that it should be compatible. It would be interesting > to know if anyone has deployed it with MariaDB. Just converted a little test setup (sqlite) to mariadb, based on this guide [1]. It works just fine. mariadb on Ubuntu Trusty: ii libmariadbclient18:amd645.5.39-0ubuntu0.14.04.1 amd64MariaDB database client library ii mariadb-client-5.5 5.5.39-0ubuntu0.14.04.1 amd64MariaDB database client binaries ii mariadb-client-core-5.5 5.5.39-0ubuntu0.14.04.1 amd64MariaDB database core client binaries ii mariadb-common 5.5.39-0ubuntu0.14.04.1 all MariaDB common metapackage ii mariadb-server-5.5 5.5.39-0ubuntu0.14.04.1 amd64MariaDB database server binaries ii mariadb-server-core-5.5 5.5.39-0ubuntu0.14.04.1 amd64MariaDB database core server files Gr. Stefan [1]: http://vadikgo.tumblr.com/post/34325489321/convert-an-opennebula-db-from-sqlite-to-mysql -- | BIT BV http://www.bit.nl/Kamer van Koophandel 09090351 | GPG: 0xD14839C6 +31 318 648 688 / i...@bit.nl ___ Users mailing list Users@lists.opennebula.org http://lists.opennebula.org/listinfo.cgi/users-opennebula.org
Re: [one-users] Multiple VLAN and Bridge
Quoting Amier Anis (myma...@gmail.com): > *Dear Team,* > > What is the best configuration my network host to be setup I would go with OpenvSwitch. It's easy to setup and you can do stuff (vlan trunking) that is very difficult to set up the "traditional" way. If at all possible, use different nic's to seperate "storage", "uplink" and "management/migration" traffic. To add redundancy/capacity you can use bonds (port channels). I can post some openvswitch setup commands if you want. Gr. Stefan -- | BIT BV http://www.bit.nl/Kamer van Koophandel 09090351 | GPG: 0xD14839C6 +31 318 648 688 / i...@bit.nl ___ Users mailing list Users@lists.opennebula.org http://lists.opennebula.org/listinfo.cgi/users-opennebula.org
Re: [one-users] Problem with Virtual Router 4.8
Quoting Hüning, Christian (christian.huen...@haw-hamburg.de): > Hi folks, > > > > I figured it out myself. It had to do with the automatically generated VLAN > ids from OpenvSwitch and the VMs not being on the same physical host. As it > seems my Cisco router would not allow these high(?) VLAN IDs. I tried > setting the VLAN Id to 5 manually and it worked! > > A quick view into Ciscos manual showed that in normal VLAN mode VLAN Ids > are only allowed to range between 1 and 1002. One can of course activate > Extended VLAN mode and then you may provide VLAN IDs between 1006 and > 4094. Is there a possibility to set a range for the automatically generated > vlan ids? There's at least the possibility to configure with which vlan to start, config file located in "../one/remotes/vnm/OpenNebulaNetwork.conf" # General Options # Configure the initial VLAN ID tag (corresponds to vnet ID = 0) :start_vlan: 2 I just configure the VLAN_ID's in the vnet template (VLAN = "YES", VLAN_ID = "VID") so you have control over which VLAN gets used. Gr. Stefan p.s. Don't forget to run "onehost sync --force" to update host configs -- | BIT BV http://www.bit.nl/Kamer van Koophandel 09090351 | GPG: 0xD14839C6 +31 318 648 688 / i...@bit.nl pgpNyPfFn6Ieh.pgp Description: PGP signature ___ Users mailing list Users@lists.opennebula.org http://lists.opennebula.org/listinfo.cgi/users-opennebula.org
Re: [one-users] BOOTPROTO, DHCP_HOSTNAME, and vm IPs in Sunstone
Quoting kerryhall . (kerryh...@gmail.com): > It sounds like it's not really possible for a DHCP client to suggest an IP > to the DHCP server. > > I think I would like to make a feature request: easy DHCP integration. > > Does my use case make sense? I can clarify further if needed. Why do you want to manage IP's from the VM itself? If you want predictable IP's for your VM's you can use static assignments in DHCP servers. Just prefill an address range with MAC's for future VM's and you're done. Just make sure that the VM gets the right IP in ONE (and therefore the right MAC-address) and off you go. Together with MAC/ARP spoofing protection [1] you can be assured that a VM is only able to use the IP you've configured. Oh, and there already is an integration possible, see [2]. If you really want to manage the DHCP server from the VM itself there are more possibilities (discussed on this list some time ago): [3], and [4,5] including examples. Gr. Stefan [1]: http://docs.opennebula.org/4.8/administration/networking/openvswitch.html#openflow-rules [2]: http://docs.opennebula.org/4.8/administration/networking/router.html [3]: http://lists.opennebula.org/pipermail/users-opennebula.org/2014-May/044932.html [4]: http://mperedim.wordpress.com/2011/02/17/opennebula-dhcpd-contextualization-magic/ [5]: http://blog.nominet.org.uk/tech/2005/12/21/using-omapi-object-management-application-programming-interface/ -- | BIT BV http://www.bit.nl/Kamer van Koophandel 09090351 | GPG: 0xD14839C6 +31 318 648 688 / i...@bit.nl ___ Users mailing list Users@lists.opennebula.org http://lists.opennebula.org/listinfo.cgi/users-opennebula.org
Re: [one-users] Default Attributes
Quoting Javier Fontan (jfon...@opennebula.org): > I've just tried the RAW parameter in the driver conf and works OK. > Make sure you restart oned as the default values are read by oned and > not by the driver itself. > > Also I had to change the line to this to make it work: > > RAW = "" Thanks, that did the trick. Obviously I didn't try all combinations ;). Gr. Stefan -- | BIT BV http://www.bit.nl/Kamer van Koophandel 09090351 | GPG: 0xD14839C6 +31 318 648 688 / i...@bit.nl ___ Users mailing list Users@lists.opennebula.org http://lists.opennebula.org/listinfo.cgi/users-opennebula.org
Re: [one-users] IP Addressing
Quoting Robert Foote (rfo...@bpsnode.com): > Hi All: > > > > I have some VMs in my opennebula cloud that need to use about 5 static IP > addresses, that will not be leased to them through OpenNebula, they will be > in use manually due to the need for VRRP on them. > > > > However, those IP addresses, are part of a VNET that I've already > established and have leases coming out of, but those specific IPs have not > yet been leased out by opennebula. Is there a way I can tell opennebula not > to use those specific IP addresses when future VMs are created on that > specific VNET? Jep. In ONE you cane put leases on HOLD. Just go to "leases" of the specific VNET, add an IP and click "Hold IP". On the cli you would do something like this: onevnet hold . In ONE 4.8 onwards you can also make a reservation: onevnet reserve . Gr. Stefan -- | BIT BV http://www.bit.nl/Kamer van Koophandel 09090351 | GPG: 0xD14839C6 +31 318 648 688 / i...@bit.nl ___ Users mailing list Users@lists.opennebula.org http://lists.opennebula.org/listinfo.cgi/users-opennebula.org
[one-users] Default Attributes
Hi, I'm trying to get some Default Attributes for kvm [1], i.e. default bootmenu in vm's: RAW = "" entry in vmm_exec_kvm.conf. I've tried old and new syntax, see [2] but it's not working for newly created vm's. I'm running ONE 4.8. I've synced all the hosts and restarted opennebula (if at all needed). This extra bit of XML is not included in the domain XML's. Any hints? Gr. Stefan [1]: http://docs.opennebula.org/4.8/administration/virtualization/kvmg.html?highlight=kvm#default-attributes [2]: http://dev.opennebula.org/issues/2951 -- | BIT BV http://www.bit.nl/Kamer van Koophandel 09090351 | GPG: 0xD14839C6 +31 318 648 688 / i...@bit.nl ___ Users mailing list Users@lists.opennebula.org http://lists.opennebula.org/listinfo.cgi/users-opennebula.org
Re: [one-users] user authentication failed in LDAP configuration
Quoting Sudeep Narayan Banerjee (snbaner...@iitgn.ac.in): > Dear Sir, > > I am trying to configure ldap in opennebula. > > I have added string *default* under AUTH_MAD in /etc/one/oned.conf file. > > AUTH_MAD = [ > executable = "one_auth_mad", > #authn = "ssh,x509,ldap,server_cipher,server_x509" > authn = "ssh,x509,ldap,*default*,server_cipher,server_x509" There's another part that goes with this [1]: "To be able to use this driver for users that are still not in the user database you must set it to the default driver. To do this go to the auth drivers directory and copy the directory ldap to default. In system-wide installations you can do this using this command: $ cp -R /var/lib/one/remotes/auth/ldap /var/lib/one/remotes/auth/default > > > Also I have set up the ldap.conf file with below configurations. > --- > :auth_method: :simple > > # Ldap server > #:host: localhost > :host: 192.168.22.12 // also can provide the hostname > :port: 389 > > # Uncomment this line for tsl conections > #:encryption: :simple_tls > > # base hierarchy where to search for users and groups > #:base: 'dc=domain' > :base: 'dc=iitgn,dc=ac,dc=in' > > # group the users need to belong to. If not set any user will do > #:group: 'cn=cloud,ou=groups,dc=domain' > > # field that holds the user name, if not set 'cn' will be used > :user_field: 'cn' > > # for Active Directory use this user_field instead > #:user_field: 'sAMAccountName' > > # field name for group membership, by default it is 'member' > #:group_field: 'member' > :group_field: 'eduPersonAffiliation' > > # user field that that is in in the group group_field, if not set 'dn' > will be used > #:user_group_field: 'dn' > -- > > Do I need to provide anywhere the password of the ldap server to test its > connectivity? Well, basic connectivity checking won't hurt: telnet 192.168.22.12 389 -> do you get a connection? Any logging on the ldap server itself? It's virtually impossible to debug ldap authentication issues only by logging in through opennebula. If you don't connect with the right user, don't have search permissions on the LDAP tree or whatever you will get an error but LDAP is not going to tell you what happened. This might be a time consuming process, be patient, be creative. I would suggest you install ldapsearch (ldap-utils on ubuntu) and do the following test to see you get the right output: ldapsearch -h 192.168.22.12 -w your-password-here -x -D binddn -b 'dc=iitgn,dc=ac,dc=in' -s sub '(you specific sub here)' It depends on your LDAP installation/configuration how you have to bind, what password you need (if at all), etc. You might find the info in [2] useful. Good luck, Gr. Stefan [1]: http://docs.opennebula.org/4.6/administration/authentication/ldap.html?highlight=ldap [2]: http://n40lab.wordpress.com/2013/11/22/centos-6-4-connecting-opennebula-sunstone-to-ldap-server/ -- | BIT BV http://www.bit.nl/Kamer van Koophandel 09090351 | GPG: 0xD14839C6 +31 318 648 688 / i...@bit.nl ___ Users mailing list Users@lists.opennebula.org http://lists.opennebula.org/listinfo.cgi/users-opennebula.org
Re: [one-users] state and lcmstate codes
Quoting Maxim Terletskiy (terlets...@emu.ru): > Hi! > > Is there any doc where STATE and LCM_STATE codes listed? I'm tryind > to find which numbers failed vm returns. I guess you're looking for this [1]. Gr. Stefan [1]: http://docs.opennebula.org/4.6/integration/system_interfaces/api.html?highlight=rpc -- | BIT BV http://www.bit.nl/Kamer van Koophandel 09090351 | GPG: 0xD14839C6 +31 318 648 688 / i...@bit.nl ___ Users mailing list Users@lists.opennebula.org http://lists.opennebula.org/listinfo.cgi/users-opennebula.org
Re: [one-users] one-context 4.6 RPM on Fedora 20 VM
Quoting Javier Fontan (jfon...@opennebula.org): > I am going to move the context packages to an addon repository [1]. We > think this will make easier to contribute to it as right now the code > is buried in the main repository, new collaborators can be added to > the repository and experimental branches can be created. It will allow > us to do separate releases from OpenNebula so new features and fixes > can be addressed in the middle of a release. > > What do you think about it? I think it's a good idea. Context packages are proably one of the most "hacked upon" scripts in ONE anyway (to help local use cases). I like to move it, move it ;). Gr. Stefan -- | BIT BV http://www.bit.nl/Kamer van Koophandel 09090351 | GPG: 0xD14839C6 +31 318 648 688 / i...@bit.nl ___ Users mailing list Users@lists.opennebula.org http://lists.opennebula.org/listinfo.cgi/users-opennebula.org
Re: [one-users] Error Migration
Quoting Rhesa Mahendra (rh...@lintasmediadanawa.com): > Hello, > > I get error when i live migrate vm, this is error ini oned.log : > > Message received: LOG E 710 migrate: Command "virsh --connect > qemu:///system migrate --live one-710 qemu+ssh://node128/system" > failed: error: internal error process exited while connecting to > monitor: Supported machines are: Are you trying to live migrate between different CentOS versions? According to [1] that's not supported. Gr. Stefan [1]: https://access.redhat.com/site/documentation/en-US/Red_Hat_Enterprise_Linux/5/html/Virtualization/chap-Virtualization-KVM_live_migration.html -- | BIT BV http://www.bit.nl/Kamer van Koophandel 09090351 | GPG: 0xD14839C6 +31 318 648 688 / i...@bit.nl ___ Users mailing list Users@lists.opennebula.org http://lists.opennebula.org/listinfo.cgi/users-opennebula.org
Re: [one-users] VM in opennebula failing
Quoting Sudeep Narayan Banerjee (snbaner...@iitgn.ac.in): > Wed Jun 25 02:13:29 2014 [LCM][I]: New VM state is RUNNING > *Wed Jun 25 02:13:30 2014 [LCM][I]: New VM state is UNKNOWN* > Wed Jun 25 02:13:35 2014 [VMM][I]: VM found again, state is RUNNING > > Why the state should be Running & Unknown at the same time? Any clue? The state is not running & Unknown at the same time. Those are different messages logged 1 second apart. Gr. Stefan -- | BIT BV http://www.bit.nl/Kamer van Koophandel 09090351 | GPG: 0xD14839C6 +31 318 648 688 / i...@bit.nl ___ Users mailing list Users@lists.opennebula.org http://lists.opennebula.org/listinfo.cgi/users-opennebula.org
Re: [one-users] VM in opennebula failing
Quoting Sudeep Narayan Banerjee (snbaner...@iitgn.ac.in): > Dear Sir, > > This is in response to the subject that I posted on June-21, but still no > response. > > I have tried to improve little bit from there but still no the VM is > failing; but now throwing new error. > > I am posting some output of the configuration files, please have a look and > kindly8 respond. > [root@nc1 ~]# grep -vE '^($|#)' /etc/default/libvirt-bin > grep: /etc/default/libvirt-bin: No such file or directory > > [oneadmin@front ~]$ onehost list > ID NAMECLUSTER RVM ALLOCATED_CPU ALLOCATED_MEM > STAT >3 nc1 - 0 0 / 800 (0%) 0K / 3.6G (0%) on > > [oneadmin@front ~]$ oneimage list > ID USER GROUP NAMEDATASTORE SIZE TYPE PER STAT > RVMS >0 oneadmin oneadmin CentOS-6.5_x86_ default 267M OSNo > rdy 0 > > [oneadmin@front ~]$ onetemplate list > ID USERGROUP NAME > REGTIME >0 oneadminoneadminCentOS-6.5 06/21 > 23:09:56 > > Output of one .log file > - > [oneadmin@front ~]$ cat /var/log/one/15.log > Tue Jun 24 13:19:31 2014 [DiM][I]: New VM state is ACTIVE. > Tue Jun 24 13:19:31 2014 [LCM][I]: New VM state is PROLOG. > Tue Jun 24 13:20:48 2014 [LCM][I]: New VM state is BOOT > Tue Jun 24 13:20:48 2014 [VMM][I]: Generating deployment file: > /var/lib/one/vms/15/deployment.0 > Tue Jun 24 13:20:49 2014 [VMM][I]: ExitCode: 0 > Tue Jun 24 13:20:49 2014 [VMM][I]: Successfully execute network driver > operation: pre. > Tue Jun 24 13:20:49 2014 [VMM][I]: Command execution fail: cat << EOT | > /var/tmp/one/vmm/kvm/deploy '/var/lib/one/datastores/0/15/deployment.0' > 'nc1' 15 nc1 > Tue Jun 24 13:20:49 2014 [VMM][I]: error: Failed to create domain from > /var/lib/one/datastores/0/15/deployment.0 > Tue Jun 24 13:20:49 2014 [VMM][I]: error: unable to set security context > 'unconfined_u:object_r:svirt_image_t:s0:c514,c893' on > '/var/lib/one/datastores/0/15/disk.0': Permission denied > Tue Jun 24 13:20:49 2014 [VMM][E]: Could not create domain from > /var/lib/one/datastores/0/15/deployment.0 It looks like a permission issue. As oneadmin user (su - oneadmin), can you "touch" anything in /var/lib/one/datastores/0? What are the permissions on /var/lib/one (ls -larth)? Are you using nfsv4 (grep nfs4 /proc/mounts)? Are the permissions the same on the frontend and the hosts? Gr. Stefan -- | BIT BV http://www.bit.nl/Kamer van Koophandel 09090351 | GPG: 0xD14839C6 +31 318 648 688 / i...@bit.nl ___ Users mailing list Users@lists.opennebula.org http://lists.opennebula.org/listinfo.cgi/users-opennebula.org
Re: [one-users] one-context 4.6 RPM on Fedora 20 VM
Quoting Javier Fontan (jfon...@opennebula.org): > I've been checking the scripts and there's not an easy fix for to make > it work in fedora. I'll install a fedora and take a look at the > problem. Maybe using "ip" command can be the solution. I'm busy working on replacing ifconfig with "ip" in the context scripts. Context scripts also assume that interface names start with "eth" which is no longer the case with systemd (arch linux systems might end up with "ens" interfaces for example"). As soon as I got it working I'll send a pull request. Gr. Stefan -- | BIT BV http://www.bit.nl/Kamer van Koophandel 09090351 | GPG: 0xD14839C6 +31 318 648 688 / i...@bit.nl ___ Users mailing list Users@lists.opennebula.org http://lists.opennebula.org/listinfo.cgi/users-opennebula.org
Re: [one-users] greetings
Quoting Galimba (gali...@gmail.com): > Hello everyone. > My name is Sebastian. I'm new to this list and tho I've been a sysadmin for > several years now, I've only recently dived into Cloud Computing. > I have successfully installed OpenNebula 4.4 on a local computer behind a > firewall at my university. I set up two nodes and another dedicated > computer as a NFS datastore. > The plan is to provide my research group with the IAAS that OpenNebula > brings to the table. > At the moment, I'm dealing with an issue I haven't been able to solve, and > perhaps some of you could throw me a hint. > My university assigned me over 100 public ip addresses to provide each VM. > If I were to plug the cable directly to the OpenNebula box, then I know I > could create my templates with public ip addresses and then everything > should be fine. The problem is that I have a firewall in the middle, > managing all the public ips, and my OpenNebula box is on a LAN behind that > firewall. Question: Do you want to filter the traffic for your vm's on the "firewall in the middle"? If the answer is yes than you might want to use the vm-hook like Valentin suggested. If not then a vlan with public IP's is probably the easiest way to go. Another possibility is to use the "Public Cloud" interface from ONE, specifically: EC2 [1]. It makes use of Elastic IPs. It uses scripting to handle the mapping of public to private ips. Especially the scripts that interact with the OpenFlow seem promising [2]. Yet another way of doing this is to route the block of 100 ip's to a router/firewall (possible running on ONE) (through a little ip interconnection block). In that case you don't have to filter on the "firewall in the middle" and or do NAT (which I think is very ugly). So like this: public ip -> interconnect-ip - router/firwall - router-ip-routed-ips -> vm's with public ip. This will also work for IPv6. Natting IPv6 is possible, but even more ugly ;). You still have the possibility to do some filtering on the firewall while leaving the rest of the ports open. If you like GUI's, pfSense is a very nice and capable firewall (based on OpenBSD's pf) [3]. If you would like to use pfSense on KVM -> don't use virtio network drivers, broken on KVM (at least that is our experience, intel e1000 works fine). Good luck, and have a fun and bright cloudy day ;), Gr. Stefan [1]: http://docs.opennebula.org/4.6/advanced_administration/public_cloud/ec2qug.html [2]: http://community.opennebula.org/ecosystem:onenox [3]: https://www.pfsense.org/ -- | BIT BV http://www.bit.nl/Kamer van Koophandel 09090351 | GPG: 0xD14839C6 +31 318 648 688 / i...@bit.nl signature.asc Description: Digital signature ___ Users mailing list Users@lists.opennebula.org http://lists.opennebula.org/listinfo.cgi/users-opennebula.org
Re: [one-users] Generate a noVNC session token outside Sunstone
Quoting Rodrigue Chakode (rodrigue.chak...@gmail.com): > Hi, > Reading the OpenNebula documentation [1], it appears that noVNC session > tokens are generated only when user clicks on the appropriate icon from > Sunstone. > > That's OK. > > However, I would like to know if it's possible, subject to appropriate > tunings or not, to generate and activate a noVNC session without logging > into Sunstone? > My concern is to be able to connect a VM through an external noVNC client. Currently you cannot. You can mimic a user session to hit sunstone on /vm//startvnc ... you will get a json-response with a key "token". With that token you can connect to vnc proxy. It would be convenient if a token could be retrieved through a xml-rpc call. We created feature request #2655 for that [1]. Gr. Stefan [1]: http://dev.opennebula.org/issues/2655 -- | BIT BV http://www.bit.nl/Kamer van Koophandel 09090351 | GPG: 0xD14839C6 +31 318 648 688 / i...@bit.nl ___ Users mailing list Users@lists.opennebula.org http://lists.opennebula.org/listinfo.cgi/users-opennebula.org
Re: [one-users] NFS hard or soft mount for datastores?
Quoting ML mail (mlnos...@yahoo.com): > Hello, > > I am using a Linux NFS server to store and run my VM images and would > like to know what type of NFS mount does OpenNebula recommend for > mounting the datastore NFS shares on my ONE hosts. Would you rather > recommend hard or soft NFS mount options? If you care about your data, never ever use soft mounts. You risk corrupting all your vm images if something goes wrong with your nfs server and/or network connecting all of it. Gr. Stefan -- | BIT BV http://www.bit.nl/Kamer van Koophandel 09090351 | GPG: 0xD14839C6 +31 318 648 688 / i...@bit.nl ___ Users mailing list Users@lists.opennebula.org http://lists.opennebula.org/listinfo.cgi/users-opennebula.org
Re: [one-users] Custom network for VM Live-migration
Quoting Javier Fontan (jfon...@opennebula.org):
> I haven't tested this but it may work.
>
> When you do a migration the destination host is specified. The host
> name is the same as the one configured in OpenNebula so it will use
> the same network as ssh connections. To use another network you can
> modify the script so the destination host is changed. For example,
> let's asume that the machines have 192.168.1.x for the standard
> service network (ssh) and another network for migrations at 10.0.1.x.
> The node names for these both networks can be the same but adding a
> suffix to the migration network, something like this:
>
> 192.168.1.1 node01
> 10.0.1.1 node01-migration
> 192.168.1.2 node02
> 10.0.1.2 node02-migration
>
> Now you can change the migration script to use the migration network
> as destination host, for example for kvm:
>
> exec_and_log "virsh --connect $LIBVIRT_URI migrate --live $deploy_id
> $QEMU_PROTOCOL://${dest_host}-migration/system" \
> "Could not migrate $deploy_id to $dest_host"
${dest_host}-migration results in "FQDN-migration" which doesn't work.
However with a little hack I made it work (/var/tmp/one/vmm/kvm/migrate):
-8<---
HOSTNAME=$(cut -f1 -d. <<< $2)
DOMAIN=$(cut -f2- -d. <<< $2)
MIGSUF="migration"
DEST_MIGR_HOST=$HOSTNAME-$MIGSUF.$DOMAIN
exec_and_log "virsh --connect $LIBVIRT_URI migrate --live $deploy_id
$QEMU_PROTOCOL://$DEST_MIGR_HOST/system --migrateuri
tcp://$DEST_MIGR_HOST" \
"Could not migrate $deploy_id to $dest_host"
->8---
The "--migrateuri tcp://$DEST_MIGR_HOST" part is especially important,
because normally virsh would find out the hostname to connect to by
"virsh hostname" command, which isn't the one with the "-migration"
suffix, and only the initial communication would be done over the live-migrate
network.
Thanks,
Gr. Stefan
--
| BIT BV http://www.bit.nl/Kamer van Koophandel 09090351
| GPG: 0xD14839C6 +31 318 648 688 / i...@bit.nl
___
Users mailing list
Users@lists.opennebula.org
http://lists.opennebula.org/listinfo.cgi/users-opennebula.org
[one-users] onetemplate instantiate --hold option sunstone
Hi list, I would like to be able to instanatie a template and put it on "HOLD" state instead of "PENDING". On the cli there's the options of adding "--hold". Is there a sunstone equivalent I'm missing? Otherwise adding a checkbox with "STATE PENDING" would be a useful little feature IMHO. Thanks, Gr. Stefan -- | BIT BV http://www.bit.nl/Kamer van Koophandel 09090351 | GPG: 0xD14839C6 +31 318 648 688 / i...@bit.nl ___ Users mailing list Users@lists.opennebula.org http://lists.opennebula.org/listinfo.cgi/users-opennebula.org
Re: [one-users] one-context 4.6 RPM on Fedora 20 VM
Quoting ML mail (mlnos...@yahoo.com): > I tried with NETWORK=YES and Sunstone actually says "Template updated > correctly" but still this custom variable is not present or saved in that > template, this is really weird. Could you try on the CLI: 8< CONTEXT=[ NETWORK="YES" ] >8 Gr. Stefan -- | BIT BV http://www.bit.nl/Kamer van Koophandel 09090351 | GPG: 0xD14839C6 +31 318 648 688 / i...@bit.nl ___ Users mailing list Users@lists.opennebula.org http://lists.opennebula.org/listinfo.cgi/users-opennebula.org
Re: [one-users] one-context 4.6 RPM on Fedora 20 VM
Quoting ML mail (mlnos...@yahoo.com): > I see, then it would be great indeed if this could be adapted for the > new ifconfig output or as suggested by someone else on the list use > the new ip tool. > > Strangely enough I tried to add the NETWORK=yes to the custom vars of > the template context but even if I click "add" and "update" the next > time I view this template the custom variable is gone... Any idea why > this custom var is not being saved? I am using ONE 4.6.0. NETWORK=yes should be all caps if I'm not mistaken, i.e. NETWORK=YES. Could you try again? Gr. Stefan -- | BIT BV http://www.bit.nl/Kamer van Koophandel 09090351 | GPG: 0xD14839C6 +31 318 648 688 / i...@bit.nl ___ Users mailing list Users@lists.opennebula.org http://lists.opennebula.org/listinfo.cgi/users-opennebula.org
Re: [one-users] one-context 4.6 RPM on Fedora 20 VM
Quoting Javier Fontan (jfon...@opennebula.org): > The output changes a lot from the old ifconfig and can not be > correctly parser. We will have to devise a new system to get that > info. I would recommend "ip" for that (iproute2). It's faster, more powerful, and consistent across linux distros. Gr. Stefan -- | BIT BV http://www.bit.nl/Kamer van Koophandel 09090351 | GPG: 0xD14839C6 +31 318 648 688 / i...@bit.nl ___ Users mailing list Users@lists.opennebula.org http://lists.opennebula.org/listinfo.cgi/users-opennebula.org
Re: [one-users] ACL rules - vDCADMIN
Quoting Daniel Molina (dmol...@opennebula.org): > Hi Stefan, > > Sunstone does not check ACLs to display the different actions, instead you > have to use the "sunstone-views/*.yaml" files to tweak the layout. Ah, check. Would it be an (awfull) lot of work to make (a) sunstone view(s) dynamic based on permissions (ACLs), instead of hard coding into views? Advantage would be that a user/(vDC)administrator only sees "buttons"/"tabs" he/she is allowd to see/use and a change in permissions would be reflected automatically (i.e. re-login / clear broser cache). Gr. Stefan -- | BIT BV http://www.bit.nl/Kamer van Koophandel 09090351 | GPG: 0xD14839C6 +31 318 648 688 / i...@bit.nl ___ Users mailing list Users@lists.opennebula.org http://lists.opennebula.org/listinfo.cgi/users-opennebula.org
Re: [one-users] Custom network for VM Live-migration
Quoting Javier Fontan (jfon...@opennebula.org):
> I haven't tested this but it may work.
>
> When you do a migration the destination host is specified. The host
> name is the same as the one configured in OpenNebula so it will use
> the same network as ssh connections. To use another network you can
> modify the script so the destination host is changed. For example,
> let's asume that the machines have 192.168.1.x for the standard
> service network (ssh) and another network for migrations at 10.0.1.x.
> The node names for these both networks can be the same but adding a
> suffix to the migration network, something like this:
>
> 192.168.1.1 node01
> 10.0.1.1 node01-migration
> 192.168.1.2 node02
> 10.0.1.2 node02-migration
>
> Now you can change the migration script to use the migration network
> as destination host, for example for kvm:
>
> exec_and_log "virsh --connect $LIBVIRT_URI migrate --live $deploy_id
> $QEMU_PROTOCOL://${dest_host}-migration/system" \
> "Could not migrate $deploy_id to $dest_host"
>
> libvirt will use ssh to create the connection to the remote host so
> make sure known_hosts is correctly populated.
Check, sounds good. I'm gonna try it out. I'll let you know if it works.
Gr. Stefan
--
| BIT BV http://www.bit.nl/Kamer van Koophandel 09090351
| GPG: 0xD14839C6 +31 318 648 688 / i...@bit.nl
___
Users mailing list
Users@lists.opennebula.org
http://lists.opennebula.org/listinfo.cgi/users-opennebula.org
[one-users] Custom network for VM Live-migration
Hi List, Is it possible to choose/change the network ONE uses to live-migrate the VM's? We've got dedicted dark-fiber between switches/DCs for live-migration so to not disrupt other traffic (with "live-migrate" VLAN on it). I guess it's not currently possible, but could be made to work if "migration" command (/var/lib/one/remotes/vmm/$hypervisor/migrate) would accept an extra argument "migrate-network" specifying the network to use for live-migration. This could be made a property of the cluster a host is a member of (every host in the cluster should have this live migrate network configured). What changes in ONE would be needed (besides the drivers) to make this work? Gr. Stefan -- | BIT BV http://www.bit.nl/Kamer van Koophandel 09090351 | GPG: 0xD14839C6 +31 318 648 688 / i...@bit.nl ___ Users mailing list Users@lists.opennebula.org http://lists.opennebula.org/listinfo.cgi/users-opennebula.org
[one-users] ACL rules - vDCADMIN
Hi, I'm playing with vDC's and vDCADMIN accounts. I've created a vDC (group, admin user for group and added a resource provider): onegroup create --name UNDEF --admin_user VDCADMIN_UNDEF --admin_password notdisclosed --resources TEMPLATE+VM --admin_resources TEMPLATE+VM+IMAGE+NET onegroup add_provider UNDEF 0 103 ID USER RES_VHNIUTGDCOZ RID OPE_UMAC ZONE 32 @104 VT- * ---c * 33 #7 U-- @104 umac * 34 #7 V-NI-T- @104 um-c * 35 @104 -H- %103 -m--#0 36 @104 --ND--- %103 u---#0 According to the ACL's the vDC Admin "VDCADMIN_UNDEF" should have permissions to create templates and images. The "VDCADMIN_UNDEF" however does not have the ability to create a new template / image. There's no green "plus" symbol that allows you to create a new template / image. Am I missing something here? I've explicitly added an ACL to grant all rights for resource "Templates" but to no avail (#7 TEMPLATE/* USE+MANAGE+ADMIN+CREATE *). Thanks, Stefan -- | BIT BV http://www.bit.nl/Kamer van Koophandel 09090351 | GPG: 0xD14839C6 +31 318 648 688 / i...@bit.nl ___ Users mailing list Users@lists.opennebula.org http://lists.opennebula.org/listinfo.cgi/users-opennebula.org
Re: [one-users] Gratuitous ARP replies: VM live-migration without loss of IP connectivity
Quoting Javier Fontan (jfon...@opennebula.org): > Thanks Stefan! > > Do you think it could be do this in the network scripts? Yes, actually that would be the best place. The arp reply can be send right away, before the "mac_spoofing" scripts kick in. That way there's also no need for an extra temporary rule to circumvent the "mac_spoofing". The script now creates a temporary "NORMAL" rule with priority "5" to allow the arp reply to be send. It also means the arp reply is send earlier in the process, which is even better. Another benefit of including this in the network driver scripts is that they can be optimized per driver (openvswitch / legacy bridging, 802.1Q, etc.). > are executed in the node to prepare the node for a new machine or a > migrated one. >Does it make sense for newly created VMs? They take more to get the IP >but it may also be good to tell the switch beforehand. Yes, it does make sense to do it for newly created VM's. Actually, the hook is triggered for every VM in state "RUNNING", thus also newly created VM's. Normally a gratuitous arp reply is send by the VM itself as soon as network interface is up and running. Sending a few extra won't hurt at all :). > > Even if this does not make sense it is a nice candidate for an addon. I've taken a look at the Ruby script for OpenvSwitch (OpenvSwitch.rb). I'll see if I can make it work for OpenvSwitch. Gr. Stefan -- | BIT BV http://www.bit.nl/Kamer van Koophandel 09090351 | GPG: 0xD14839C6 +31 318 648 688 / i...@bit.nl ___ Users mailing list Users@lists.opennebula.org http://lists.opennebula.org/listinfo.cgi/users-opennebula.org
[one-users] Gratuitous ARP replies: VM live-migration without loss of IP connectivity
Hi List, In some larger cloud / network environments IPv4 connectivity issues may arise when VM's get (live-)migrated from one HOST to the other. The MAC address(es) of the Virtual Machine (VM) might still be mapped to the wrong port on switches/routers upstream from the switch the HOST is patched on. This might lead to IPv4 connectivity issues for the VM being migrated. This issue becomes escpecially noticable when the VM doesn't generate a lot of network traffic and/or the MAC address table timeout on the switch(es) is large. To overcome this issue I made a "VM_HOOK" and script(s) that run on the HOST when the hook is triggered [1]. It will execute a script that sends Gratuitous ARP replies (unsollicited ARP reply) on the virtual network interface(s) on behalf of the VM. These VM_HOOK and script(s) will work on a KVM/QEMU environment with OpenvSwitch. It should be fairly easy to adjust the scripts to support other enviroments (i.e. Xen with "legacy" bridging). Environments based on VMware with VM's that have "vmware-tools" installed are covered by VMware ESX. VMware will have the VM send gratuitous ARP replies as soon as the live-migration finishes. It does _not_ work for VM's without vmware-tools installed though. I hope "one-grarp" will be useful for some of you out there suffering from this issue. Gr. Stefan [1]: https://github.com/hydro-b/one-grarp P.s. In so called "switched fabric" networks the issue above might not occur because the "fabric" operates as one (logical) switch (like Brocade's Ethernet Fabric). -- | BIT BV http://www.bit.nl/Kamer van Koophandel 09090351 | GPG: 0xD14839C6 +31 318 648 688 / i...@bit.nl ___ 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
Quoting Daniel Dehennin (daniel.dehen...@baby-gnu.org): > Daniel Dehennin writes: > > > [...] > > > I found a solution: > > #+begin_src > > OS=[ARCH="x86_64",BOOT="hd"] > > RAW=[DATA=" > > > > > > > > ",TYPE="kvm"] > > #+end_src > > I tried to move this RAW section from the template to > /etc/one/vmm_exec/vmm_exec_kvm.conf but it's not added to the > deployement.0. > > Is this issue specific to my installation or does someone reproduce it? Did you do a "onehost sync (--force)" after this change? Gr. Stefan -- | BIT BV http://www.bit.nl/Kamer van Koophandel 09090351 | GPG: 0xD14839C6 +31 318 648 688 / i...@bit.nl signature.asc Description: Digital 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
Quoting Daniel Dehennin (daniel.dehen...@baby-gnu.org): > Daniel Dehennin 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. Gr. Stefan -- | BIT BV http://www.bit.nl/Kamer van Koophandel 09090351 | GPG: 0xD14839C6 +31 318 648 688 / i...@bit.nl signature.asc Description: Digital signature ___ Users mailing list Users@lists.opennebula.org http://lists.opennebula.org/listinfo.cgi/users-opennebula.org
Re: [one-users] gateone integration
Quoting Thomas Stein (himbe...@meine-oma.de): > Hello. > > I saw a youtube video the other day. And the guy in it talked about > gateone and opennebula beeing a good team. Could someone post a link to > a howto or something? http://docs.opennebula.org/4.6/advanced_administration/application_insight/onegate_overview.html?highlight=onegate In section "Application Monitoring" there is a sample script. Gr. Stefan -- | BIT BV http://www.bit.nl/Kamer van Koophandel 09090351 | GPG: 0xD14839C6 +31 318 648 688 / i...@bit.nl ___ Users mailing list Users@lists.opennebula.org http://lists.opennebula.org/listinfo.cgi/users-opennebula.org
[one-users] Print "Name" in (confirmation) dialog
Hi, Features #1028 and #2708 (duplicate) are closed in 4.6. A confirmation dialog now returns the ID of the object. I would also like to have the "Name" printed of the Object as well. Just the "ID" doesn't provide me with enough any information. A "Name" normally does have a special meaning and would provide much more information. Is there a special reason why "Name" is left out? Gr. Stefan -- | BIT BV http://www.bit.nl/Kamer van Koophandel 09090351 | GPG: 0xD14839C6 +31 318 648 688 / i...@bit.nl ___ Users mailing list Users@lists.opennebula.org http://lists.opennebula.org/listinfo.cgi/users-opennebula.org
[one-users] Sunstone Live Migrate
Hi, If I select an individual VM in Sunstone and click "Live Migrate" a new window pops up and displays the Hosts to choose from. It does not print the where the VM [ID] is currently running on. However, if the checkbox in front of the VM is selected and then the VM itself, it will display the info where on which host the VM [ID] is currently running on. I guess the "checkbox" always needs to be checked if (only) an individual VM is selected. Gr. Stefan -- | BIT BV http://www.bit.nl/Kamer van Koophandel 09090351 | GPG: 0xD14839C6 +31 318 648 688 / i...@bit.nl ___ Users mailing list Users@lists.opennebula.org http://lists.opennebula.org/listinfo.cgi/users-opennebula.org
Re: [one-users] Sunstone Dashboard "Real Capcacity Usage"
Quoting Stefan Kooman (ste...@bit.nl): > Hi, > > I'm wondering what the math is behind the "Real Capcacity Usage" in > Sunstone for "Virtual Machines". I'm seeing a 29% CPU and 102% Memory as > Real Capacity Usage for VM's. The hosts have for CPU: > > 1790/6400 (28%) Allocated 611/6400 (28%) Real > > And for Memory: > > 217.5 GB / 755.8 GB (29%) Allocated 144.9 GB / 755.8 GB (19%) Real > > I don't understand how the Real Capcacity Usage for Hosts is calculated. I ment to say "I don't understand how the Real Capcacity Usage for Virtual Machines" is calculated. Gr. Stefan -- | BIT BV http://www.bit.nl/Kamer van Koophandel 09090351 | GPG: 0xD14839C6 +31 318 648 688 / i...@bit.nl ___ Users mailing list Users@lists.opennebula.org http://lists.opennebula.org/listinfo.cgi/users-opennebula.org
[one-users] Sunstone Dashboard "Real Capcacity Usage"
Hi, I'm wondering what the math is behind the "Real Capcacity Usage" in Sunstone for "Virtual Machines". I'm seeing a 29% CPU and 102% Memory as Real Capacity Usage for VM's. The hosts have for CPU: 1790/6400 (28%) Allocated 611/6400 (28%) Real And for Memory: 217.5 GB / 755.8 GB (29%) Allocated 144.9 GB / 755.8 GB (19%) Real I don't understand how the Real Capcacity Usage for Hosts is calculated. Graphical representation of above in the attachment. Can you please explain? Gr. Stefan -- | BIT BV http://www.bit.nl/Kamer van Koophandel 09090351 | GPG: 0xD14839C6 +31 318 648 688 / i...@bit.nl ___ Users mailing list Users@lists.opennebula.org http://lists.opennebula.org/listinfo.cgi/users-opennebula.org
Re: [one-users] New graphic accounting interface in Sunstone - OpenNebula 4.6.1
Quoting Tino Vazquez (cvazq...@c12g.com): > Dear users, > > Just a quick note to highlight the graphical accounting information > tab in Sunstone. It is accesible for the end user through the Cloud > View to check their consumption, and for the cloud administrator > through the Admin View, where she can check accounting information for > a particular virtual data center. Nice! Are you planning on adding extra metrics to the equation (Network TX/RX, DISK IOPS)? Gr. Stefan -- | BIT BV http://www.bit.nl/Kamer van Koophandel 09090351 | GPG: 0xD14839C6 +31 318 648 688 / i...@bit.nl ___ Users mailing list Users@lists.opennebula.org http://lists.opennebula.org/listinfo.cgi/users-opennebula.org
Re: [one-users] VM life cycle diagram
Quoting Hamada, Ondrej (ondrej.ham...@acision.com): > Hi, you can find it here: > http://docs.opennebula.org/4.6/user/virtual_resource_management/vm_guide_2.html#virtual-machine-life-cycle Ah, thanks. In that doc I found a reference to [1]. It gives a complete VM Life Cyle. However it's a .PNG with a too low resolution to be able to read the labels :(. It would be nice if it could be replaced with a SVG version (or a high res. PNG). Thanks. Gr. Stefan [1]: http://docs.opennebula.org/4.6/user/virtual_resource_management/vm_guide_2.html#virtual-machine-life-cycle -- | BIT BV http://www.bit.nl/Kamer van Koophandel 09090351 | GPG: 0xD14839C6 +31 318 648 688 / i...@bit.nl ___ Users mailing list Users@lists.opennebula.org http://lists.opennebula.org/listinfo.cgi/users-opennebula.org
[one-users] VM life cycle diagram
Hi, I'm looking for the "VM Life Cycle Diagram" (graphviz output of different states of a VM during it's life cycle). This was linked to [1] in ONE 4.4. Is this diagram still available somewhere? Thanks, Gr. Stefan [1]: http://opennebula.org/_media/documentation:rel4.4:states-complete.png -- | BIT BV http://www.bit.nl/Kamer van Koophandel 09090351 | GPG: 0xD14839C6 +31 318 648 688 / i...@bit.nl ___ Users mailing list Users@lists.opennebula.org http://lists.opennebula.org/listinfo.cgi/users-opennebula.org
Re: [one-users] Sunstone 403 at login (only with chromium / chrome)
Quoting Daniel Molina (dmol...@opennebula.org): > Hi Stefan, > > > On 12 February 2014 17:55, Stefan Kooman wrote: > > > Hi, > > > > We have trouble logging in to sunstone with Chromium as a browser. After > > provoding credentials we end up with a "403" forbidden. Chromium debug > > console show this: > > > > When the default login page gets loaded it gets a "401 Unauthorized" > > trying to GET /jquery.min.map/vendor/4.0. After providing credentials I > > get a "Unexpected error. Status 403. Check the server logs.". Chromium > > console logs it's getting a 403 during a "POST" for "login" > > (jquery-1.9.1.min.js:5 Script). In firefox everything works as > > expected. > > It doesn't matter if I provide correct credentials or not. It always > > gives a 403. > > > > > The 401 error with jquery.min.map should not be a problem, it is not > required is just a debuggin library. Sunstone only returns a 403 error if > the resource requested is not a valid one (i.e: GET immage instead of > image). > > Could you check > * if there is any message in the 403 error body. > * what is the path that is been used in the POST login request > * maybe the problem is in the apache configuration This issue disappeared after upgrading to ONE 4.6. Gr. Stefan -- | BIT BV http://www.bit.nl/Kamer van Koophandel 09090351 | GPG: 0xD14839C6 +31 318 648 688 / i...@bit.nl ___ Users mailing list Users@lists.opennebula.org http://lists.opennebula.org/listinfo.cgi/users-opennebula.org
Re: [one-users] VM monitoring information do not get deleted
Quoting Wilma Hermann (wilma.herm...@gmail.com): > Hi, > > I observed a problem with two OpenNebula setups, that I set up with version > 4.4 and which I upgraded to 4.6 some weeks ago: The VM monitoring > information does not seem to be deleted from the database (MySQL) after > VM_MONITORING_EXPIRATION_TIME has expired. > > I have a sandbox for testing issues: A single machine (both frontend and > host) with a single virtual machine, that runs 24/7. When I upgraded > OpenNebula 4.4 to 4.6, the SQL-Dump created by "onedb upgrade" was 3.6 MB > big (perfectly okay for such a small setup). Today, when I dumped the DB, > the backup file is 176 MB in size. Wondering about the size, I inspected > the database and found ~77k rows in the "vm_monitoring" table. Obviously, > OpenNebula writes rows into this table every few seconds without ever > deleting anything. > > I didn't change VM_MONITORING_EXPIRATION_TIME in oned.conf (it was > commented out), so it should delete old values after 4h. I manually set > VM_MONITORING_EXPIRATION_TIME to 14400 as well as other values: No effect, > the DB continues to inflate. > > Meanwhile, Sunstone begins to become unresponsible when I open the details > of a VM. I believe this is due to generating the CPU and memory graphs > which has to process several ten thousands of rows. > > Did I miss some setting or is this a bug? After reading the above things start to make sense. We're using a MySQL master-master replication setup, with one oned server as primary master. The amount of network traffic, InnoDB activity, disk throughput, etc have gone up tremendously. See attached images to get an impression. For newly created vm's opening "capacity" or "network" tab this isn't a problem, yet. But for vm's that are already running for month's this is a problem. I see the sunstone instance that is serving me dropping out of the load-balancer for not replying to health-checks in time. Just by clicking the "network" tab of a long running vm. If this is a bug I need a workaround soon before running out of disk space ;). Gr. Stefan P.s Thanks for Wilma for spotting this, haven't had time to look into this issue: too busy with reverting back from trusty -> saucy on hypervisors, more on that later. -- | BIT BV http://www.bit.nl/Kamer van Koophandel 09090351 | GPG: 0xD14839C6 +31 318 648 688 / i...@bit.nl ___ Users mailing list Users@lists.opennebula.org http://lists.opennebula.org/listinfo.cgi/users-opennebula.org
[one-users] Sunstone template update - Description not updated
Hi, There seems to be a bug in Sunstone. If I update the DESCRIPTION in a TEMPLATE in sunstone it responds with "template updated correctly", but the change is not saved. onetemplate update $TEMPLATE_ID does work as expected. Gr. Stefan P.s. I've cleared my cache and tried both Chromium / Firefox (both latest) -- | BIT BV http://www.bit.nl/Kamer van Koophandel 09090351 | GPG: 0xD14839C6 +31 318 648 688 / i...@bit.nl ___ Users mailing list Users@lists.opennebula.org http://lists.opennebula.org/listinfo.cgi/users-opennebula.org
Re: [one-users] system datastores 0 not show capacity hdd
Quoting Amier Anis (myma...@gmail.com): > Dear List, > > I have an issue that system datastores 0 was not able to show the capacity > hdd. > This is fresh installation with internal hdd. normal partition scheme, > centos 6.5, opennebula 4.4 Do you have a host in your setup? The "system" datastore is checked on the host, not on the frontend. Gr. Stefan -- | BIT BV http://www.bit.nl/Kamer van Koophandel 09090351 | GPG: 0xD14839C6 +31 318 648 688 / i...@bit.nl ___ Users mailing list Users@lists.opennebula.org http://lists.opennebula.org/listinfo.cgi/users-opennebula.org
Re: [one-users] can user's RAM quota acount only running/active VMs?
Quoting Carlos Martín Sánchez (cmar...@opennebula.org): > Hi, > > On Fri, May 2, 2014 at 10:51 PM, Liu, Guang Jun (Gene) < > gene@alcatel-lucent.com> wrote: > > > Hi there, > > > > Is there a way to config or twist OpenNebula so that quota management > > (e.g. RAM) counts resources of active VMs only? > > > > I found the quota management (RAM per user) counts all the VMs which > > created by a user. We found if the quota only counts the resources from the > > active VMs, that means more sense. E.g., I have 16RAM quota. I create one > > VM w/ 16G RAM. Then I save/stop my VM. The system should allow me to create > > another VM w/ 16G RAM if quota management counts only active VMs. > > > > Thanks! > > Gene > > > > There isn't a way to configure that. It would require to modify the core. > > But I think it makes sense. So I return the question to the community: > ¿Should we consider this change in the quota behaviour? I vote for this change. Your not actually consuming resources on the hypervisor(s) for not having running vm's ... Gr. Stefan -- | BIT BV http://www.bit.nl/Kamer van Koophandel 09090351 | GPG: 0xD14839C6 +31 318 648 688 / i...@bit.nl ___ Users mailing list Users@lists.opennebula.org http://lists.opennebula.org/listinfo.cgi/users-opennebula.org
Re: [one-users] Limit Storage Bandwidth
Quoting Rhesa Mahendra (rh...@lintasmediadanawa.com): > Guys, > > Could we limit bandwidth from VM read write into Datastores per > second? for avoid network outage, thanks and please help There is already a feature request for that, plannen Target version: 4.8, see [1]. Gr. Stefan [1] http://dev.opennebula.org/issues/2530 -- | BIT BV http://www.bit.nl/Kamer van Koophandel 09090351 | GPG: 0xD14839C6 +31 318 648 688 / i...@bit.nl ___ Users mailing list Users@lists.opennebula.org http://lists.opennebula.org/listinfo.cgi/users-opennebula.org
Re: [one-users] creating system datastore
Quoting Christophe Duez (christophe.d...@student.uantwerpen.be): > hello, is it normal that when you create a new system datastore opennebula > does not make a directory for it? As soon as you deploy a VM in that datastore it should create the directory on the host. > however it creates a directory for new image and file directories > Is this because of security reasons, bug reasons, configuration, . Gr. Stefan -- | BIT BV http://www.bit.nl/Kamer van Koophandel 09090351 | GPG: 0xD14839C6 +31 318 648 688 / i...@bit.nl ___ Users mailing list Users@lists.opennebula.org http://lists.opennebula.org/listinfo.cgi/users-opennebula.org
Re: [one-users] sunstone create or update template leads to browser hang
Quoting Thomas Stein (himbe...@meine-oma.de): > Hello. > > Whenever i try to create or update a template in sunstone i have to wait > 2min until the wizzard shows up. It's opennebula 4.6 and Firefox. This > happens with Macosx and Linux. Is this a known problem? Any workarounds? I can confirm this with latest firefox (29) on linux. I get the following popup from firefox: A script on this page may be busy, or it may have stopped responding. You can stop the script now, or you can continue to see if the script will complete. Script: https://cust-sunstone.bit.nl/vendor/4.0/jquery-1.11.0.min.js?v=4.6.0:2 Gr. Stefan -- | BIT BV http://www.bit.nl/Kamer van Koophandel 09090351 | GPG: 0xD14839C6 +31 318 648 688 / i...@bit.nl ___ 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
Quoting Thomas Stein (himbe...@meine-oma.de): > On Friday 09 May 2014 10:19:49 Neil Schneider wrote: > > > OS=[ > > ARCH="x86_64", > > BOOT="cdrom", > > GUESTOS="rhel6_64Guest", > > ROOT="hda" ] > > I see. I wasn't aware of the fact i could use hd*a*. I just used hd and sd > and > vd. Stupid me. Well, actually I _don't_ think it's you being stupid. I stumbled upon this a long time ago. It would be nice if OpenNebula would pick "hda" for the ISO (CDROM) (not context CDROM) if "CDROM" is choosen as boot device (instead of hd). Or make it clear which device will be used to boot from (and/or make it a checkbox). Gr. Stefan -- | BIT BV http://www.bit.nl/Kamer van Koophandel 09090351 | GPG: 0xD14839C6 +31 318 648 688 / i...@bit.nl ___ Users mailing list Users@lists.opennebula.org http://lists.opennebula.org/listinfo.cgi/users-opennebula.org
Re: [one-users] vCPU and /proc/cpuinfo
Quoting Vassilis Vatikiotis (vatik...@iit.demokritos.gr): > Hello all, > > I provision a template where a number of vCPUs are specified, in my > case 2 vCPUs. However, in /proc/cpuinfo only one CPU shows up. What > plays? The template tab of the VM shows, correctly, 2 CPUs. What OS do you use? You know for sure it's a "SMP" kernel? What does a "virsh dumpxml one-id" on the hypervisor tell you about the # of CPU's? Gr. Stefan -- | BIT BV http://www.bit.nl/Kamer van Koophandel 09090351 | GPG: 0xD14839C6 +31 318 648 688 / i...@bit.nl ___ Users mailing list Users@lists.opennebula.org http://lists.opennebula.org/listinfo.cgi/users-opennebula.org
Re: [one-users] ip spoofing
Quoting Maxim Terletskiy (terlets...@emu.ru): > Hi! > > No we're using vlans with simple bridging. I've used ovswitch in > past and do not remember anything about "ip hijacking" prevention in > it. How can ovswitch know what ip/mac must be on vm interface? Will > it be usefull if VMs living on different virtualization hosts? "openvswitch" by itself doesn't do anything to prevent "mac spoofing" or "ip hijacking". That's done by ONE based on OpenFlow rules. From the docs [1]: Mac-spoofing These rules prevent any traffic to come out of the port the MAC address has changed. in_port=,dl_src=,priority=4,actions=normal in_port=,priority=39000,actions=normal IP hijacking These rules prevent any traffic to come out of the port for IPv4 IP’s not configured for a VM in_port=,arp,dl_src=priority=45000,actions=drop in_port=,arp,dl_src=,nw_src=,priority=46000,actions=normal See /var/lib/one/remotes/vnm/ovswitch/OpenvSwitch.rb as well. Using openvswitch gives you exactly what you asked for, without the need for hacking ebtables/iptables script. In a much cleaner way IMHO. Gr. Stefan -- | BIT BV http://www.bit.nl/Kamer van Koophandel 09090351 | GPG: 0xD14839C6 +31 318 648 688 / i...@bit.nl ___ Users mailing list Users@lists.opennebula.org http://lists.opennebula.org/listinfo.cgi/users-opennebula.org
Re: [one-users] opennebula 4.6 and ubuntu 14.04
Quoting Kenneth (kenn...@apolloglobal.net): > > > The latest OpenNebula is 4.6 right? And latest ubuntu is 14.04. > > Is > it safe to install opennebula 4.6 in ubuntu 14.04 now for production > systems? Well, I'm going to upgrade our production systems, probably next week, to Ubuntu 14.04 and ONE 4.6 (13.10 -> 14.04, ONE 4.4.0 -> 4.6). I Already tested some stuff on beta / RC releases of Ubuntu and ONE and everything is looking good. I know one of our customers is running this combination succesfully. Gr. Stefan -- | BIT BV http://www.bit.nl/Kamer van Koophandel 09090351 | GPG: 0xD14839C6 +31 318 648 688 / i...@bit.nl ___ Users mailing list Users@lists.opennebula.org http://lists.opennebula.org/listinfo.cgi/users-opennebula.org
Re: [one-users] ip spoofing
Quoting Maxim Terletskiy (terlets...@emu.ru): > Thanks for answer. > > We're giving white ips from several networks, dhcp not very well fit > our needs. So using context for network settings setup. > > Now we're looking for solution which prevent vm A with ip X from use > of ip Y which belongs to vm B. X and Y are ips from one network in > one vlan. Are you using openvswitch? In that case you already have "mac spoofing" and "IP Hijacking" prevention [1]. OpenFlow rules are used to set that up. Note that it only works for IPv4 addresses. IPv6 is not covered yet. Gr. Stefan [1]: http://docs.opennebula.org/4.6/administration/networking/openvswitch.html -- | BIT BV http://www.bit.nl/Kamer van Koophandel 09090351 | GPG: 0xD14839C6 +31 318 648 688 / i...@bit.nl ___ Users mailing list Users@lists.opennebula.org http://lists.opennebula.org/listinfo.cgi/users-opennebula.org
Re: [one-users] ubuntu 14.04 as image - no network on first boot
Quoting Javier Fontan (jfon...@opennebula.org): > We have applied the changes made by Michael. That seems to do the > trick. Thank you! > > You can find the package at > http://dev.opennebula.org/attachments/download/779/one-context_4.6.0.deb I would add an "ip addr flush dev $i" to make sure ip's are removed, ifdown doesn't take care of that :(. Gr. Stefan -- | BIT BV http://www.bit.nl/Kamer van Koophandel 09090351 | GPG: 0xD14839C6 +31 318 648 688 / i...@bit.nl ___ Users mailing list Users@lists.opennebula.org http://lists.opennebula.org/listinfo.cgi/users-opennebula.org
Re: [one-users] System volumes created with no space
Quoting Ruben S. Montero (rsmont...@opennebula.org): > That's right Stefan > > System DS are monitored through the host (so we can accommodate ssh system > ds and other types that cannot be accessed from the front-end). If no host > is reporting space for the System DS size will be 0. Check, thanks for confirming. @MariusRex, so then the question is: why aren't your host monitoring the system datastore? Gr. Stefan -- | BIT BV http://www.bit.nl/Kamer van Koophandel 09090351 | GPG: 0xD14839C6 +31 318 648 688 / i...@bit.nl ___ Users mailing list Users@lists.opennebula.org http://lists.opennebula.org/listinfo.cgi/users-opennebula.org
Re: [one-users] System volumes created with no space
Quoting Marius Rex (marius...@venda.com): > > I am using Opennebula 4.5.80 on Centos6 with KVM. > > /var/lib/one is a NFS mount, and has over a terabyte of free space available. > I have it shared between my administrative host and my KVM servers, but I > cannot launch a VM as there is never any space seen as available in the > systems datastore. > > [oneadmin$] onedatastore list > ID NAMESIZE AVAIL CLUSTER IMAGES TYPE DS TM >0 system0M - snowman 0 sys -shared >1 default 1.4T 100% snowman 1 img fs shared >2 files 1.4T 100% snowman 0 fil fs ssh > > All volumes are actually located on the same NFS volume: > > [oneadmin]$ df -h /var/lib/one > FilesystemSize Used Avail Use% Mounted on > snowman03.of-1.uk.venda.com:/tank/systems/one > 1.4T 946M 1.4T 1% /var/lib/one > > Yet: > > [oneadmin]$ onedatastore show 0 > DATASTORE 0 INFORMATION > > ID : 0 > NAME : system > USER : oneadmin > GROUP : oneadmin > CLUSTER: snowman > TYPE : SYSTEM > DS_MAD : - > TM_MAD : shared > BASE PATH : /var/lib/one//datastores/0 > DISK_TYPE : FILE > > DATASTORE CAPACITY > > TOTAL: : 0M > FREE: : 0M > USED: : 0M > LIMIT: : - > > PERMISSIONS > > OWNER : um- > GROUP : u-- > OTHER : --- > > DATASTORE TEMPLATE > > BASE_PATH="/var/lib/one//datastores/" > SHARED="YES" > TM_MAD="shared" > TYPE="SYSTEM_DS" > > IMAGES > [oneadmin]$ onedatastore show 1 > DATASTORE 1 INFORMATION > > ID : 1 > NAME : default > USER : oneadmin > GROUP : oneadmin > CLUSTER: snowman > TYPE : IMAGE > DS_MAD : fs > TM_MAD : shared > BASE PATH : /var/lib/one//datastores/1 > DISK_TYPE : FILE > > DATASTORE CAPACITY > > TOTAL: : 1.4T > FREE: : 1.4T > USED: : 945M > LIMIT: : - > > PERMISSIONS > > OWNER : um- > GROUP : u-- > OTHER : --- > > DATASTORE TEMPLATE > > BASE_PATH="/var/lib/one//datastores/" > CLONE_TARGET="SYSTEM" > DISK_TYPE="FILE" > DS_MAD="fs" > LN_TARGET="NONE" > TM_MAD="shared" > TYPE="IMAGE_DS" > > IMAGES > 5 > > > So I tried creating a new systems datastore on separate NFS mount as a test: > > [oneadmin]$ cat template.ds > NAME= systems_nfs > TM_MAD = shared > TYPE= SYSTEM_DS > BASE_PATH = /mnt/systems/one > > [oneadmin]$ onedatastore create template.ds -c snowman > > And I get another system volume with no space: > > [oneadmin@snow-opennebula01 ~]$ onedatastore show 102 > DATASTORE 102 INFORMATION > > ID : 102 > NAME : systems_nfs > USER : oneadmin > GROUP : oneadmin > CLUSTER: snowman > TYPE : SYSTEM > DS_MAD : - > TM_MAD : shared > BASE PATH : /mnt/systems/one/102 > DISK_TYPE : FILE > > DATASTORE CAPACITY > > TOTAL: : 0M > FREE: : 0M > USED: : 0M > LIMIT: : - > > PERMISSIONS > > OWNER : um- > GROUP : u-- > OTHER : --- > > DATASTORE TEMPLATE > > BASE_PATH="/mnt/systems/one/" > SHARED="YES" > TM_MAD="shared" > TYPE="SYSTEM_DS" > > IMAGES > > [oneadmin]$ df -h /mnt/systems/one > Filesystem
Re: [one-users] Virtual machine vlan tagging
Quoting Neil Schneider (n...@ifxonline.com): > I'm running OpenNebula in a virtual machine using openvswitch to connect > the machines. I've tried creating a virtual network in libvirt attached to > the bridge device then attached to it. I tried configuring with the > network interfaces connected directly to the bridge. I keep running into a > problem, where the virtual machine wants to add it's own tag to the > interface. I've been googling 'till I'm blue in the face and I can't seem > to find why vlan tags are getting attached to the devices without me > configuring them and it's ignoring any vlan tags I try to impose. Here's > what ovs-vsctl shows me. > > ovs-vsctl show > c6def17d-2cc6-499e-a461-af4fe9aab78a > Bridge management > Port "veth1_one" > tag: 1 > Interface "veth1_one" > Port management > Interface management > type: internal > Port "vlan10" > tag: 10 > Interface "vlan10" > type: internal > Bridge storage > Port storage > Interface storage > type: internal > Port "vlan20" > tag: 20 > Interface "vlan20" > type: internal > Port "veth2_one" > tag: 2 > Interface "veth2_one" > Bridge public > Port "vnet0" > Interface "vnet0" > Port "veth3_one" > Interface "veth3_one" > Port "em1" > Interface "em1" > Port public > Interface public > type: internal > ovs_version: "2.1.0" > > veth1_one and veth2_one are interfaces on the virtual machine. Notice the > tags. I have that configured no where and I can't seem to find where it's > coming from. Of course it keeps the virtual machines from communicating. > The interface that has no vlan associated with it works as expected. It's > the two other interfaces designated management and storage that I'm having > issues with. > > I sure could use some help finding why this is happening. I don't mind > having vlan tags associated, but I want to determine what they are. I would do this a little bit different. Instead of handling virtual networking yourselve, let libvirt do it for you. I attached "management.xml" and "storage.xml" livirt virtual networking definitions. You can enable them like this: "virsh net-define /path/to/definition.xml". management is in vlan 10, storage in vlan 20. Start the networks: virsh net-start management, virsh net-start storage. And let's make them auto-started: virsh net-autostart management, virsh net-autostart storage. Know in your vm definition do the following (virsh edit vm_name): Example for managment interface: For storage interface: Example for storage interface: You should know have your virtual interfaces in the right network with the right vlan tag. Gr. Stefan -- | BIT BV http://www.bit.nl/Kamer van Koophandel 09090351 | GPG: 0xD14839C6 +31 318 648 688 / i...@bit.nl management.xml Description: XML document storage.xml Description: XML document ___ Users mailing list Users@lists.opennebula.org http://lists.opennebula.org/listinfo.cgi/users-opennebula.org
[one-users] OpenNebula 4.5.85-1 on Ubuntu Trusty Tahr
Hi list, I've build OpenNebula 4.6 RC (4.5.85) for Ubuntu Trusty Tahr (14.04) [1]. The package libxmlrpc-c3-dev does not exist any more in Ubuntu Trusty Tahr (virtual package). I've replaced the requirement in "control" with "libxmlrpc-core-c3-dev". Besides that "libxmlrpc-c++8-dev" is also required for building. With those packages in Build-Depends it builds fine. Besides that I've removed "rubygems" as a dependency for "ruby-opennebula". Rubygems is included in "ruby" package in Trusty. I'm testing this build in a vm. So far oned, and sunstone work. The rest is, as of yet, untested. @DEVS Are you planning on supporting/releasing OpenNebula 4.6 for Trusty at OpenNebula 4.6 release time? Gr. Stefan [1]: http://obit.bit.nl/ubuntu/trusty/opennebula/ -- | BIT BV http://www.bit.nl/Kamer van Koophandel 09090351 | GPG: 0xD14839C6 +31 318 648 688 / i...@bit.nl ___ Users mailing list Users@lists.opennebula.org http://lists.opennebula.org/listinfo.cgi/users-opennebula.org
Re: [one-users] Sunstone Datastores Capacity unreadable
Quoting Stefan Kooman (ste...@bit.nl): > Hi, > > I'm doing some testing on Sunstone 4.6 RC. I found a little cosmetic > thing in Datastores overview. Capacity is unreadable when multiple > datastores are available. See attachment. Somehow these attachments always get lost ;). Gr. Stefan -- | BIT BV http://www.bit.nl/Kamer van Koophandel 09090351 | GPG: 0xD14839C6 +31 318 648 688 / i...@bit.nl ___ Users mailing list Users@lists.opennebula.org http://lists.opennebula.org/listinfo.cgi/users-opennebula.org
[one-users] Sunstone Datastores Capacity unreadable
Hi, I'm doing some testing on Sunstone 4.6 RC. I found a little cosmetic thing in Datastores overview. Capacity is unreadable when multiple datastores are available. See attachment. Gr. Stefan -- | BIT BV http://www.bit.nl/Kamer van Koophandel 09090351 | GPG: 0xD14839C6 +31 318 648 688 / i...@bit.nl ___ Users mailing list Users@lists.opennebula.org http://lists.opennebula.org/listinfo.cgi/users-opennebula.org
Re: [one-users] Help with OpenNebula Setup
Quoting Grant Tailor (therealwebg...@gmail.com): > I have been trying to setup OpenNebula > http://docs.opennebula.org/4.6/desig...entos_kvm.html for about 2 days now. > I went through hell setting it up to this point. I am hoping ports that > need to be opened will be mentioned on the documentation page. If i am > given permission i can add many things to the current documentation. The documentation is available on github [1]. You can "fork" the wiki, make your changes / additions and send a "pull request" to OpenNebula so they can review and merge your changes. Gr. Stefan [1]: https://github.com/OpenNebula/docs -- | BIT BV http://www.bit.nl/Kamer van Koophandel 09090351 | GPG: 0xD14839C6 +31 318 648 688 / i...@bit.nl ___ Users mailing list Users@lists.opennebula.org http://lists.opennebula.org/listinfo.cgi/users-opennebula.org
Re: [one-users] Call for translations OpenNebula 4.6 (DL 25th April)
Quoting Daniel Molina (dmol...@opennebula.org): > Dear community, > > This is a reminder of the call for translations for our web interface: > Sunstone. We kindly ask all translators to complete the translations as > much as possible before tomorrow EOB. > > We would like to thank all the contributors for the great response and the > efforts carried out so far. It is great to count with so many languages :) > > https://www.transifex.com/projects/p/one 100% Done for Dutch translation! Might not be perfectly translated but at least we can improve from here onward. Gr. Stefan -- | BIT BV http://www.bit.nl/Kamer van Koophandel 09090351 | GPG: 0xD14839C6 +31 318 648 688 / i...@bit.nl ___ Users mailing list Users@lists.opennebula.org http://lists.opennebula.org/listinfo.cgi/users-opennebula.org
Re: [one-users] Share IP's
Quoting Stefan Kooman (ste...@bit.nl):
> Quoting Gareth Bult (gar...@linux.co.uk):
> > Hi,
> >
> > I'm looking to share a single IP address between two VM's - can anyone
> > point me in the right direction? (I'm using the 'clean-traffic'
> > network filter ...)
> >
> > My first idea was to add a second interface to each VM, with a shared
> > IP, but on reflection because IP's are linked to MAC's, this was never
> > going to work.
>
> Why not? I've got several VM's with extra (V)IP's. An interface
> mac-address is normally "linked" to the ip it gets assigned, not the
> other way around.
Not quite true those were IPv6 ip's ... for IPv4 it's different.
There's not only an "MAC spoofing" filter but also an "arp" filter.
Extra ipv4 ip's do not work by default.
> > Is there a way of doing this, or is the clean traffic filter
> > incompatible with shared IP's ...
>
> clean traffic filter does prevent Mac-spoofing, but you're allowed to add
> ip aliases.
You can disable "arp" filtering quite easily:
(/var/lib/)one/remotes/vnm/ovswitch/OpenvSwitch.rb
8< -
def mac_spoofing
add_flow("in_port=#{port},arp,dl_src=#{@nic[:mac]}",:drop,45000)
#
add_flow("in_port=#{port},arp,dl_src=#{@nic[:mac]},nw_src=#{@nic[:ip]}",:normal,46000)
add_flow("in_port=#{port},arp,dl_src=#{@nic[:mac]}",:normal,46000)
add_flow("in_port=#{port},dl_src=#{@nic[:mac]}",:normal,4)
add_flow("in_port=#{port}",:drop,39000)
end
>8 -
Gr. Stefan
--
| BIT BV http://www.bit.nl/Kamer van Koophandel 09090351
| GPG: 0xD14839C6 +31 318 648 688 / i...@bit.nl
___
Users mailing list
Users@lists.opennebula.org
http://lists.opennebula.org/listinfo.cgi/users-opennebula.org
Re: [one-users] (anti-) affinity groups support for scheduler
Quoting Carlos Martín Sánchez (cmar...@opennebula.org): > Hi Stefan, > > How do you plan to implement it? I was thinking that instead of making the > scheduler go through all the VMs running on a Host, the core could have > support for this. It could add/remove the value of VM's AFFINITY_GROUP to > the host on deployment/shutdown. > > This way the host will have AFFINITY_GROUP="nameA, nameB, nameC" > automatically populated, making the scheduling faster. Support in core would be even better. At first I was thinking about a similar approach: make something that runs on the hypervisors and create "AFFINITY_GROUPS" based on some info. But it's hard to make it "dynamically" as there is no coupling between the scheduler and the hosts a priori. Having support for this in core would be awesome, and would speed op scheduling. Proably scales better too: instead of checking for a (number of) vm's on all hosts it's reduced to checking for an "attribute" at all hosts. Gr. Stefan -- | BIT BV http://www.bit.nl/Kamer van Koophandel 09090351 | GPG: 0xD14839C6 +31 318 648 688 / i...@bit.nl ___ Users mailing list Users@lists.opennebula.org http://lists.opennebula.org/listinfo.cgi/users-opennebula.org
Re: [one-users] Share IP's
Quoting Gareth Bult (gar...@linux.co.uk): > Hi, > > I'm looking to share a single IP address between two VM's - can anyone > point me in the right direction? (I'm using the 'clean-traffic' > network filter ...) > > My first idea was to add a second interface to each VM, with a shared > IP, but on reflection because IP's are linked to MAC's, this was never > going to work. Why not? I've got several VM's with extra (V)IP's. An interface mac-address is normally "linked" to the ip it gets assigned, not the other way around. > Is there a way of doing this, or is the clean traffic filter > incompatible with shared IP's ... clean traffic filter does prevent Mac-spoofing, but you're allowed to add ip aliases. Maybe I don't understand what you're trying to achieve. Gr. Stefan -- | BIT BV http://www.bit.nl/Kamer van Koophandel 09090351 | GPG: 0xD14839C6 +31 318 648 688 / i...@bit.nl ___ Users mailing list Users@lists.opennebula.org http://lists.opennebula.org/listinfo.cgi/users-opennebula.org
[one-users] (anti-) affinity groups support for scheduler
Hi List, I'm trying to accomplish the following. I want to have the "match making scheduler" schedule two or more vm's on the same hypervisor (webserver and database server, to reduce network traffic between HOSTS. I know there is a way of doing this with "CURRENT_VMS". CURRENT_VMS only seems to accept "VM ID's" and not the name of a VM. The drawback of having an ID hardcoded in a template is that if the VM gets recreated from a template somewhere in the future (because of some changes in the template) the REQUIREMENT will never be fullfilled and the vm never deployed. One way to get around this would be to create so called "(anti-)affinity groups. So in a VM TEMPLATE you would define "AFFINITY_GROUP=$AFFINITY_GROUP" and the scheduler would check if a VM with that particular AFFINITY_GROUP is running on a hypervisor. If so, it would place this VM on the same hypervisor. If not it deploys the VM on a hypervisor that has highest priority after filtering. You might wonder why I would not just select a HOST for these particular VM's. With HOST=$HOST I would not be able to bring those VM's up on a different hypervisor in case of a disaster without modifying a template or manually forcing a deploy. Not something you've got time for while battling a (major) outage. You can think of an AFFINITY_GROUP as a selective "black hole": Sucking up VM's it has affinity with. What do you think of this? Does it makes sense to you? Would you have use for this funtionality? Gr. Stefan -- | BIT BV http://www.bit.nl/Kamer van Koophandel 09090351 | GPG: 0xD14839C6 +31 318 648 688 / i...@bit.nl ___ Users mailing list Users@lists.opennebula.org http://lists.opennebula.org/listinfo.cgi/users-opennebula.org
Re: [one-users] 4.6 changes overview
Quoting Robert Schweikert (rjsch...@suse.com): > /usr/lib/systemd/system/econe.service > /usr/lib/systemd/system/occi.service > /usr/lib/systemd/system/oneflow.service > /usr/lib/systemd/system/onegate.service http://dev.opennebula.org/issues/2697 This feature request would suggest these systemd scripts have yet to be written ... Gr. Stefan -- | BIT BV http://www.bit.nl/Kamer van Koophandel 09090351 | GPG: 0xD14839C6 +31 318 648 688 / i...@bit.nl ___ Users mailing list Users@lists.opennebula.org http://lists.opennebula.org/listinfo.cgi/users-opennebula.org
Re: [one-users] 4.6 changes overview
Quoting Robert Schweikert (rjsch...@suse.com): > Hi, > > Trying to set up the 4.5.80 sources as beta packages in the > Cloud:OpenNebula:Testing in the openSUSE Build service. I had little > time to keep up with development since the 4.4. release and the > linked release notes [1] are still dead. I found these: http://docs.opennebula.org/4.6/release_notes/release_notes_46beta/index.html > > I could use a quick rundown on the structural changes to the code base. > > Looks like ozones has disappeared, what does that mean/imply? It's been superseded by "zones" and "federation". Same idea, new implementation. Federation is new: it's now possible to administer multiple "zones" with one interface [1]. > Previously I had an opennebula-ozones package for those that needed > to enable zones for their cloud. There are a bunch of new services: Econe and occi are not new, oneflow/onegate where introduced in ONE 4.2. > > /usr/lib/systemd/system/econe.service EC2 compatible API for OpenNebula [2]. > /usr/lib/systemd/system/occi.service Open Cloud Computing Interface for OpenNebula [3]. > /usr/lib/systemd/system/oneflow.service Automatic scale-up / scale-down of VM's / App's based on "elasticity" rules. [4]. > /usr/lib/systemd/system/onegate.service Let me quote the doc's here "OneGate allows Virtual Machine guests to push monitoring information to OpenNebula. Users and administrators can use it to gather metrics, detect problems in their applications, and trigger OneFlow auto-scaling rules." [5]. Gr. Stefan [1]: http://docs.opennebula.org/4.6/advanced_administration/data_center_federation/federationconfig.html#federationconfig [2]:http://docs.opennebula.org/4.6/advanced_administration/public_cloud/ec2qcg.html [3]: http://docs.opennebula.org/4.6/advanced_administration/public_cloud/occicg.html [4]: http://docs.opennebula.org/4.6/advanced_administration/application_flow_and_auto-scaling/index.html [5]: http://docs.opennebula.org/4.6/advanced_administration/application_insight/index.html Gr. Stefan -- | BIT BV http://www.bit.nl/Kamer van Koophandel 09090351 | GPG: 0xD14839C6 +31 318 648 688 / i...@bit.nl ___ Users mailing list Users@lists.opennebula.org http://lists.opennebula.org/listinfo.cgi/users-opennebula.org
Re: [one-users] Opennebula and openvswitch problem.
Quoting Leszek Master (keks...@gmail.com): > Ubuntu server 14.04 doesn't have opennebula in repo. So i decidec to do > Campbell way. I've installed clean Ubuntu Server 12.04 LTS, added cloud > repository and installed everything. I've got the same error: > > failed: error: Unable to add bridge br0 port vnet0: Operation not supported > > It still doesn't work without brcompat :(. Now i have to manually downgrade > openvswitch version from 1.10.2 to 1.4.6 to get brcompat working. You might want to do it another way: hypervisors: Trusty In a virtual machine: opennebula on precise/raring/quantal/saucy Gr. Stefan -- | BIT BV http://www.bit.nl/Kamer van Koophandel 09090351 | GPG: 0xD14839C6 +31 318 648 688 / i...@bit.nl ___ Users mailing list Users@lists.opennebula.org http://lists.opennebula.org/listinfo.cgi/users-opennebula.org
Re: [one-users] Opennebula and openvswitch problem.
Quoting James Dobson (jdob...@velocix.com): > You need to upgrade the version of libvirt, basically support for openvswitch > wasn't added until a later version if you don't have the later version you > have to use brcompat. > > http://libvirt.org/formatnetwork.html Ah, _that_ was it ... thanks James for reminding me. I tried backporting libvirt versions of saucy / trusty but that won't work out of the box (loads of dependencies that are not going to be installed). I would go for Trusty. Gr. Stefan -- | BIT BV http://www.bit.nl/Kamer van Koophandel 09090351 | GPG: 0xD14839C6 +31 318 648 688 / i...@bit.nl ___ Users mailing list Users@lists.opennebula.org http://lists.opennebula.org/listinfo.cgi/users-opennebula.org
Re: [one-users] Opennebula and openvswitch problem.
Quoting Leszek Master (keks...@gmail.com): > I'm waiting for official relese of next LTS version, i can use only LTS in > my production, so i was testing it on the 12.04. If there isn't any > official manual how ot solve this problem i'll upgrade my distro and try > then :) Thanks for your help. Precise/quantal suffers from this bug: https://bugs.launchpad.net/bugs/1084028 Fixed in newer releases: saucy / trusty (tested by me). This is apart from the legacy bridging stuff. With virtualization development happening so quickly I would recommend going for newer instead of older. Hopefully linux 3.14 and Qemu 2.0 will make it into Trusty ... Gr. Stefan -- | BIT BV http://www.bit.nl/Kamer van Koophandel 09090351 | GPG: 0xD14839C6 +31 318 648 688 / i...@bit.nl ___ Users mailing list Users@lists.opennebula.org http://lists.opennebula.org/listinfo.cgi/users-opennebula.org
Re: [one-users] Opennebula and openvswitch problem.
Quoting Leszek Master (keks...@gmail.com): > Still the same problem: > > failed: error: Unable to add bridge br0 port vnet0: Operation not supported > > sudo ovs-vsctl show > e2e80002-00ea-464d-9c62-5af5e0889d81 > Bridge "br0" > Port "eth1" > Interface "eth1" > Port "br0" > Interface "br0" > type: internal > ovs_version: "1.4.6" > leni@node-03:~$ sudo brctl show > bridge name bridge id STP enabled interfaces > leni@node-03:~$ > > > Tryied updating openvswitch from 1.4 to 1.9, the problem stays the same. > The only thing to get this working for now is to load brcompat module to > os, but it's problematic with new nodes. Any one have solution to this > problem? I believe I managed to get it going on Precise, but I had some issues, just like you. It's been some time ago though. Ubuntu Precise Trusty Tahr is due in a few weeks. Any particular reason you would want to stay on Precise?. Gr. Stefan -- | BIT BV http://www.bit.nl/Kamer van Koophandel 09090351 | GPG: 0xD14839C6 +31 318 648 688 / i...@bit.nl ___ Users mailing list Users@lists.opennebula.org http://lists.opennebula.org/listinfo.cgi/users-opennebula.org
Re: [one-users] Ceph image creation works ... but disk is empty (ONE 4.6 beta)
Quoting Jaime Melis (jme...@opennebula.org): > Hi Stefan, > > bug confirmed, and (hopefully) fixed. > > Can you try with this patch: > https://github.com/OpenNebula/one/commit/d44e5be354c8d7b3d8bf23a3875a3b08b49ee776 > > That should be applied on /var/lib/one/remotes/downloader.sh (and run > onehost sync). > > Please let me know if after applying this patch it works for you. Great > catch, btw :) I do the cathing, you do the fixing, deal?! ;). Works like a charm ... Just deployed my first vm on CEPH via OpenNebula \o/. Gr. Stefan -- | BIT BV http://www.bit.nl/Kamer van Koophandel 09090351 | GPG: 0xD14839C6 +31 318 648 688 / i...@bit.nl ___ Users mailing list Users@lists.opennebula.org http://lists.opennebula.org/listinfo.cgi/users-opennebula.org
Re: [one-users] Opennebula and openvswitch problem.
Quoting Leszek Master (keks...@gmail.com): > I've installed opennebula and got it working with flat network, but i need > to set up working openvswitch networking in my cloud. So i followed the > manual from opennebula docs but i've got error while creating VM: > > [VMM][I]: error: Unable to add bridge br0 port vnet0: Operation not > supported Do you have support for ovs-vsctl / ovs-ofctl in sudoers file on your hosts? Something like: %oneadminALL=(root) NOPASSWD: /usr/bin/ovs-vsctl %oneadminALL=(root) NOPASSWD: /usr/bin/ovs-ofctl What OpenNebula version are you using? Gr. Stefan -- | BIT BV http://www.bit.nl/Kamer van Koophandel 09090351 | GPG: 0xD14839C6 +31 318 648 688 / i...@bit.nl ___ Users mailing list Users@lists.opennebula.org http://lists.opennebula.org/listinfo.cgi/users-opennebula.org
Re: [one-users] EC2 driver error (ONE 4.6 beta)
Quoting Daniel Molina (dmol...@opennebula.org): > Could you try removing the m1.small|large|... empty fields or setting them > to 0? That helps :). I was able to reproduce the issue with a "KEYPAIR" with a "white space", i.e. "key space" in a EC2 VM template: Mon Mar 31 21:46:59 2014 [VMM][E]: Error parsing monitoring information: USEDMEMORY=0 USEDCPU=0 NETTX=0 NETRX=0 STATE=a AWS_DNS_NAME=ec2-54-72-149-220.eu-west-1.compute.amazonaws.com AWS_PRIVATE_DNS_NAME=ip-172-31-21-87.eu-west-1.compute.internal AWS_KEY_NAME=key space AWS_AVAILABILITY_ZONE=eu-west-1c AWS_VPC_ID=vpc-7ef7b615 AWS_PRIVATE_IP_ADDRESS=172.31.21.87 AWS_IP_ADDRESS=54.72.149.220 AWS_SUBNET_ID=subnet-7cf7b617 AWS_SECURITY_GROUPS=sg-1a5bb275 AWS_INSTANCE_TYPE=t1.micro If I use a KEYPAIR without space(s) it it able to monitor correctly: [VMM][D]: VM $ONE_ID successfully monitored: I created issues #2820 and #2821. Gr. Stefan -- | BIT BV http://www.bit.nl/Kamer van Koophandel 09090351 | GPG: 0xD14839C6 +31 318 648 688 / i...@bit.nl ___ Users mailing list Users@lists.opennebula.org http://lists.opennebula.org/listinfo.cgi/users-opennebula.org
[one-users] EC2 driver error (ONE 4.6 beta)
Hi, One of the TechDay attendee was so excited about cloud-bursting he tried the same evening in his own cloud. At first he didn't succeed. This turned out to be a template problem with a "KEYPAIR" containing a "space". I tried to reproduce that, but in the process I hit some errors: Add a EC2 host: onehost create eu-west-1 --im ec2 --vm ec2 --net dummy Resulting in: Mon Mar 31 14:56:06 2014 [InM][D]: Monitoring host eu-west-1 (4) Mon Mar 31 14:56:07 2014 [InM][I]: Command execution fail: /var/lib/one/remotes/im/run_probes ec2 /var/lib/one//datastores 4124 20 4 eu-west-1 Mon Mar 31 14:56:07 2014 [InM][I]: /var/lib/one/remotes/vmm/ec2/ec2_driver.rb:304:in `*': nil can't be coerced into Float (TypeError) Mon Mar 31 14:56:07 2014 [InM][I]: from /var/lib/one/remotes/vmm/ec2/ec2_driver.rb:304:in `block in monitor_all_vms' Mon Mar 31 14:56:07 2014 [InM][I]: from /var/lib/one/remotes/vmm/ec2/ec2_driver.rb:303:in `each' Mon Mar 31 14:56:07 2014 [InM][I]: from /var/lib/one/remotes/vmm/ec2/ec2_driver.rb:303:in `monitor_all_vms' Mon Mar 31 14:56:07 2014 [InM][I]: from ./poll:26:in `' Mon Mar 31 14:56:07 2014 [InM][E]: Error executing poll Relevant info in ec2_driver.conf: eu-west-1: region_name: eu-west-1 access_key_id: STRING secret_access_key: ANOTHER_STRING capacity: t1.micro: 1 m1.small: m1.large: m1.xlarge: I added the "t1.micro: 1". This is opennebula 4.6 beta. Any hints? Gr. Stefan -- | BIT BV http://www.bit.nl/Kamer van Koophandel 09090351 | GPG: 0xD14839C6 +31 318 648 688 / i...@bit.nl ___ Users mailing list Users@lists.opennebula.org http://lists.opennebula.org/listinfo.cgi/users-opennebula.org
Re: [one-users] Add `SSH_PUBLIC_KEY` to contextualization to set root credentials.
Quoting Christophe Duez (christophe.d...@student.uantwerpen.be): > Hello > In the marketplace you can download images. > However in the discription this is says: Add `SSH_PUBLIC_KEY` to > contextualization to set root credentials > what do they mean with this? In the template you create to use this image you have to provide your "SSH_PUBLIC_KEY" so you can log in with ssh and your private key. Root password is not set. So you can only log in through ssh. You might also "chroot" the image (using a linux live environment like GRML [1]), set a password there and log in through console or ssh with a password. Gr. Stefan [1]: http://grml.org/ P.s make sure the image is "persistent" if you set a password, otherwise your changes will be lost. -- | BIT BV http://www.bit.nl/Kamer van Koophandel 09090351 | GPG: 0xD14839C6 +31 318 648 688 / i...@bit.nl ___ Users mailing list Users@lists.opennebula.org http://lists.opennebula.org/listinfo.cgi/users-opennebula.org
Re: [one-users] Native GlusterFS support
Quoting Shankhadeep Shome (shank15...@gmail.com): > Can you share that ppa for the community please? I am assuming the qemu-img > ppa that has gluster support. FYI the fedora releases 19 and 20 do have > glusterfs support in qemu without extra packages. https://launchpad.net/~semiosis/+archive/ubuntu-qemu-glusterfs It's not my PPA but it works. I've run into issues (1 UNIT test failed) while enabeling support for glusterfs in ubuntu trusty deb. Gr. Stefan -- | BIT BV http://www.bit.nl/Kamer van Koophandel 09090351 | GPG: 0xD14839C6 +31 318 648 688 / i...@bit.nl ___ Users mailing list Users@lists.opennebula.org http://lists.opennebula.org/listinfo.cgi/users-opennebula.org
Re: [one-users] Native GlusterFS support
Quoting Javier Fontan (jfon...@opennebula.org): > It could be a permission problem. Try this command with both root and > oneadmin: > > $ qemu-img info gluster://gluster1:24007/gv0/53f35f5ca22540d558404caa3b9907d4 qemu-img info gluster://gluster1:24007/gv0/53f35f5ca22540d558404caa3b9907d4 qemu-img: Could not open 'gluster://gluster1:24007/gv0/53f35f5ca22540d558404caa3b9907d4': Unknown protocol It turns out that stock libvirt/qemu in Ubuntu Trusty do _not_ have gluster support. I've installed a PPA [1], with that PPA installed it does work: qemu-img info gluster://gluster1:24007/gv0/53f35f5ca22540d558404caa3b9907d4 image: gluster://gluster1:24007/gv0/53f35f5ca22540d558404caa3b9907d4 file format: raw virtual size: 40M (41943040 bytes) disk size: 40M (ttylinux image). I'll go ahead and install the PPA on the other node too. Thanks for helping me out. Gr. Stefan -- | BIT BV http://www.bit.nl/Kamer van Koophandel 09090351 | GPG: 0xD14839C6 +31 318 648 688 / i...@bit.nl ___ Users mailing list Users@lists.opennebula.org http://lists.opennebula.org/listinfo.cgi/users-opennebula.org
Re: [one-users] Opennebula 4.4 plus "dual stack IPv4/IPv6"
Quoting Steven Timm (t...@fnal.gov): > > Does anyone have a worked example of an OpenNebula virtual network > that has a so-called "dual stack" setup, i.e with both ipv4 and > ipv6 addresses created and assigned by OpenNebula in a "fixed" network? > > The documents appear to indicate that you can do only ipv4 or ipv6, > not both. You can do both, in a "FIXED" as well as a "RANGED" network. You do need to enable "CONTEXT_FORCE_IPV4", see here my template: VIRTUAL NETWORK TEMPLATE CLUSTER="kvm_cluster" CONTEXT_FORCE_IPV4="yes" DNS="" GATEWAY="" GATEWAY6="" NETWORK_ADDRESS="" NETWORK_MASK="" Gr. Stefan -- | BIT BV http://www.bit.nl/Kamer van Koophandel 09090351 | GPG: 0xD14839C6 +31 318 648 688 / i...@bit.nl ___ Users mailing list Users@lists.opennebula.org http://lists.opennebula.org/listinfo.cgi/users-opennebula.org
Re: [one-users] Native GlusterFS support
Quoting Javier Fontan (jfon...@opennebula.org): > Maybe I have a bug in the documentation. Set the ds mad to "fs" and > check if it works. I changed the setting and that helped :). Unfortunately qemu logs an error: qemu-system-x86_64: -drive file=gluster://gluster1:24007/gv0/53f35f5ca22540d558404caa3b9907d4,if=none,id=drive-virtio-disk0,format=raw,cache=none: Gluster connection failed for server=gluster1 port=24007 volume=gv0 image=53f35f5ca22540d558404caa3b9907d4 transport=tcp ... while the gluster volume is mounted correctly on this hypervisor I'll try to debug this later. Gr. Stefan -- | BIT BV http://www.bit.nl/Kamer van Koophandel 09090351 | GPG: 0xD14839C6 +31 318 648 688 / i...@bit.nl ___ Users mailing list Users@lists.opennebula.org http://lists.opennebula.org/listinfo.cgi/users-opennebula.org
[one-users] Ceph image creation works ... but disk is empty (ONE 4.6 beta)
Hi List, I've setup a 5 node ceph cluster and 2 kvm nodes. Everything is working fine. Ceph cluster is healthy, etc. I'm able to list images on the frontend, kvm nodes. Datastore is added succesfully: DATASTORE 100 INFORMATION ID : 100 NAME : ceph_one_ds USER : oneadmin GROUP : oneadmin CLUSTER: - TYPE : IMAGE DS_MAD : ceph TM_MAD : ceph BASE PATH : /var/lib/one//datastores/100 DISK_TYPE : RBD DATASTORE CAPACITY TOTAL: : 1T FREE: : 967.1G USED: : 42.1G LIMIT: : - PERMISSIONS OWNER : um- GROUP : u-- OTHER : --- DATASTORE TEMPLATE BASE_PATH="/var/lib/one//datastores/" BRIDGE_LIST="oned1" CEPH_HOST="ceph1 ceph2 ceph3" CEPH_SECRET="e6ba4d0a-0e20-4a3a-9053-0a7ef50a02b9" CEPH_USER="libvirt" CLONE_TARGET="SELF" DISK_TYPE="RBD" DS_MAD="ceph" LN_TARGET="NONE" POOL_NAME="one" TM_MAD="ceph" TYPE="IMAGE_DS" IMAGES 8 9 However, If I import an image from the marketplace it gets registered succesfully. But actually the image is empty: qemu-img info rbd:one/one-9 image: rbd:one/one-9 file format: raw virtual size: 0 (0 bytes) disk size: unavailable Image details: IMAGE 9 INFORMATION ID : 9 NAME : centos_ceph_test USER : oneadmin GROUP : oneadmin DATASTORE : ceph_one_ds TYPE : OS REGISTER TIME : 03/24 10:47:10 PERSISTENT : Yes SOURCE : one/one-9 PATH : http://marketplace.c12g.com/appliance/4fc76a938fb81d351702/download/0 SIZE : 10G STATE : rdy RUNNING_VMS: 0 PERMISSIONS OWNER : um- GROUP : --- OTHER : --- IMAGE TEMPLATE DEV_PREFIX="hd" MD5="94176fbba17f5efcabb0b05fc55b1a16" I've manually created a qemu-image (qemu-img convert -f qcow2 -O raw debian_squeeze.qcow2 rbd:one/debian2) and made a libvirt xml based on openenbula default. That image just works. Everything seems to be in order, no errors anywhere, but result in empty images nevertheless. Gr. Stefan -- | BIT BV http://www.bit.nl/Kamer van Koophandel 09090351 | GPG: 0xD14839C6 +31 318 648 688 / i...@bit.nl ___ Users mailing list Users@lists.opennebula.org http://lists.opennebula.org/listinfo.cgi/users-opennebula.org
Re: [one-users] Native GlusterFS support
Quoting Javier Fontan (jfon...@opennebula.org): > Now that the packages for OpenNebula 4.6 beta are ready is anyone > willing to give a shot to the gluster integration? Any feedback is > welcome. > > Post: http://opennebula.org/native-glusterfs-image-access-for-kvm-drivers/ > Packages: http://opennebula.org/software/ > Documentation: > http://docs.opennebula.org/4.6/administration/storage/gluster_ds.html I'm trying glusterfs on Ubuntu Saucy (frontend) and Ubuntu Trusty (nodes). I've followed the documentation but something is not working. Might be me missing something here. DATASTORE 101 INFORMATION ID : 101 NAME : gluster_gv0_ds USER : oneadmin GROUP : oneadmin CLUSTER: - TYPE : IMAGE DS_MAD : shared TM_MAD : shared BASE PATH : /var/lib/one//datastores/101 DISK_TYPE : DATASTORE CAPACITY TOTAL: : 0M FREE: : 0M USED: : 0M LIMIT: : - PERMISSIONS OWNER : um- GROUP : u-- OTHER : --- DATASTORE TEMPLATE BASE_PATH="/var/lib/one//datastores/" CLONE_TARGET="SYSTEM" DISK_TYPE="GLUSTER" DS_MAD="shared" GLUSTER_HOST="gluster1:24007" GLUSTER_VOLUME="gv0" LN_TARGET="NONE" TM_MAD="shared" TYPE="IMAGE_DS" oneadmin@oned1:~$ onedatastore list ID NAMESIZE AVAIL CLUSTER IMAGES TYPE DS TM 0 system215.9G 92% - 0 sys -shared 1 default 215.9G 92% - 1 img fs shared 2 files 17.6G 38% - 0 fil fs ssh 100 ceph_one_ds 1T 91% - 2 img ceph ceph 101 gluster_gv0_d 0M - - 0 img shared shared It reports "0M" - AVAIL. The "system" and "default" datastores are mounted glusterfs volume (so glusterfs does work on frontend/nodes). If I try to import an image from marketplace I get the following error: [ImageAllocate] Cannot determine Image SIZE. Datastore driver 'shared' not available. Gr. Stefan -- | BIT BV http://www.bit.nl/Kamer van Koophandel 09090351 | GPG: 0xD14839C6 +31 318 648 688 / i...@bit.nl ___ Users mailing list Users@lists.opennebula.org http://lists.opennebula.org/listinfo.cgi/users-opennebula.org
Re: [one-users] Request for comments: Sunstone dashboard
Quoting Carlos Martín Sánchez (cmar...@opennebula.org): > Dear community, > > As you may know if you tried the beta release, we are doing a facelift to > Sunstone. We are now improving the refresh mechanism to improve the > performance and alleviate the load Sunstone puts on OpenNebula. > > To remove as many refresh calls, we are going to simplify the dashboard. > Right now it shows aggregated historic graphs for the VM network speed, and > the Host CPU and Memory. > > We will change it to show only the instantaneous number of VMs in each > state, and the instantaneous cpu and memory usage. Plus the existing > storage, users, and network stats. Possibly, we will also add the current > quota usage. > > The information shown by the graphs we are going to remove will be still > accessible for each VM/Host in the individual info view. > > > We'd appreciate your comments on what information you would like to see in > the dashboard, for both admins and regular users. I've been discussing this with a collegue of mine. We have some ideas on improving the Dashboard. It might be more than a facelift though, but we think it would make the dasyboard more useful than it's now. Here we go: - 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. - 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.). - Add Virtual Datacenter / Resource groups Would be nice to have stats/quota info on these as well. - Add quota / accounting I would definately add "quota" and "accounting" info. Especially for "users" and VDC admins this will be useful info. Instead of having absolute numbers, like "2.1 TB" used storage, a (VDC) / oneadmin would like to know how close he is to limits. 2.1 TB of used storage might be 1% of capacitiy, but if that would mean datastores are 95% full the (VDC) / oneadmin definately wants to take action immediately. Elaborating on this: It would be nice to get an (overall) "cloud health". All resources below thresholds would give "green" light, all the way up to "red" (resources exceeding critical thresholds). This would definately make it dashboard like and "cloud managers" will love it ;). Our 2 cents. Gr. Stefan -- | BIT BV http://www.bit.nl/Kamer van Koophandel 09090351 | GPG: 0xD14839C6 +31 318 648 688 / i...@bit.nl ___ Users mailing list Users@lists.opennebula.org http://lists.opennebula.org/listinfo.cgi/users-opennebula.org
