virt October 2011

virt@lists.centos.org

24 participants
12 discussions

centos6 and xen4 dom0 memory allocation
by Peter Peltonen 08 Jul '12

08 Jul '12

Hi, This is slightly OT as xen is not officially supported in EL6. But maybe someone else is running it on centos6 as well and could help me out here: I have problems allocating a static amount of memory for my xen dom0. It seems to reduce when new domUs are created, even though I *think* ballooning is disabled. Can someone advise me what I'm doing wrong? I have xen-4.1.1-3 installed from http://xenbits.xen.org/people/mayoung/EL6.xen on my centos6 server. I am running 2.6.32-131.12.1.el6.xendom0.x86_64 kernel available from the same place. 'xl info' reports about available memory as follows: total_memory : 24567 free_memory : 3698 xen_commandline : dom0_mem=2048M and in xend-config.sxp I have: (dom0-min-mem 256) (enable-dom0-ballooning no) (total_available_memory 0) Now one would think that I should have 2048M memory for dom0, right? I started from there, but when starting domUs it has been reduced to 1140M even though I think ballooning is disabled and there is free memory available? And the weird thing is that dom0 has done some swapping: # free -mt total used free shared buffers cached Mem: 1140 1052 88 0 2 12 -/+ buffers/cache: 1038 102 Swap: 8194 85 8109 Total: 9335 1138 8197 Shutting down domUs and making more memory available does not reverse the effect, dom0 does not allocate more memory. What am I missing here or is my xen setup acting strangely here? Best regards, Peter

2 4

how to add file-based disk space to a guest
by thomas veymont 31 Oct '11

31 Oct '11

hi there, I'm following these documentations to add a file-based disk volume to a KVM guest under Centos 6.0 : http://docs.redhat.com/docs/en-US/Red_Hat_Enterprise_Linux/6/html/Virtualiz… as instructed, I created a "pool" then a "volume", file-based, e.g : mkdir /mnt/raid/kvm_pool1 virsh # pool-define-as pool1 dir - - - - "/mnt/raid/kvm_pool1" virsh # pool-autostart pool1 virsh # vol-create-as pool1 volume1 20G --allocation 15G --format qcow2 now I want to associate "volume1" to my guest OS. Following this doc: http://docs.redhat.com/docs/en-US/Red_Hat_Enterprise_Linux/6/html/Virtualiz… - why does this ask me to create a file with "dd" ? it's already been created before with the virsh pool commands, isn't it? Seems to me I'm bypassing the libvirt/virsh layer if I do that. - after that, the doc tells me to do some stuff with guest XML files. Is'nt there some specific commands provided by virsh to associate a managed Pool to a managed Guest ? - in this case, should I use the virsh "attach-disk" command ? thanks. -- Tom

2 4

using local media file to install guest
by Bob Hoffman 31 Oct '11

31 Oct '11

Hi, Been fighting with this for a long time and have narrowed my issues down to one item... How to use a iso image file to install a guest os with centos6? I installed centos6 with the virt host package. According to few obscure sites virt-install cannot use local media iso to install guests. With virt-manager I was able to use the same iso to make a guest, but I do not want the gui on the server. I tried mounting, umounting it..I tried using dd to bring the bootable iso from the cdrom. the issue I think, but not sure, is the use of the console directive with virt-install. It appears --extra-args="console=tty0 console=ttyS0,115200" works fine when using --location, but not --cdrom Unfortunately '--location" will ignore the path provided for the local iso and want to go to a url, an nfs, a virtual hard drive or cdrom....thus not able to install from local media. However, the text installer does come up. when using "--cdrom" and use location to the different ISO's I added the console will not come up as the --extra-args="console=tty0 console=ttyS0,115200" is ignored I assume. I cannot believe that the only way to install a guest on my machine is to use the desktop gui or a remote iso only. There has to be a way to install via the local iso... I have tried about 40 times with different ideas, even playing with the files on the mounted bootable iso. However, the console only works with --location thus unusable.....sigh. any help appreciated. here is just a basic example of my attempts virt-install --name test --ram=1024 --vcpus=1 --prompt --nographics --location=/backups/backup1/centos6.iso --extra-args="console=tty0 console=ttyS0,115200" I do not want VNC since again, that is not part of the normal install and is again asking me to use a remote computer to install on my local one, which is just not right. There must be a way to install on my local computer, from my local computer, from local file, a local guest.... isn't there?

6 17

onboot=yes not working for eth0.x:y interfaces?
by Dennis Jacobfeuerborn 25 Oct '11

25 Oct '11

Hi, I ran into a Problem when using alias interfaces with vlans on a centos 5 box. When I define such an interface and do a "service network restart" the alias interface doesn't get started. Starting it manually works fine. Also after changing DEVICE=eth0.10:0 to DEVICE=eth0:0 in the config file the interface is started after a network restart. Does anyone have an idea what the problem might be or a suitable workaround? Regards, Dennis

2 1

bridging bonded NICs, centos6
by Bob Hoffman 25 Oct '11

25 Oct '11

Trying to think this through, hope someone can lend a hand. I have 3 eths I want to bond together and that seems simple enough. eth0,1,2 <-> bond0 , though not sure how this would be for multiple ip addresses needed for the virtual machines... perhaps eth:0, eth:1 etc....to bond0:1, bond0:2 ? then I want to bridge to my virtual machines.... I am going to be using virtual machines, each with one ip address (websites) some with possibly more than one ip... so... for each ip do I need to bond all three nics, go to a new bonding file, then to the bridge file for that virtual machine? what happens when I use two IP addresses for the same virtual machine? Each machine wants to use just one bridge, but will it use more than one? I am pretty sure each bonding file must have an ipaddress on it...so perhaps it will be multiple eth files pointing to multiple bonding files pointing to single bridge files for each machine? somewhat confused. any help appreciated. am I even close?

2 1

disk i/o benchmarking inside vm
by Dennis Jacobfeuerborn 23 Oct '11

23 Oct '11

Hi, I'm having trouble benchmarking disk i/o in my vm's. The data I get seems bogus. I have two centos 6 guests which use a raw image as volume. Each volume is stored on its own physical disk and both disks are the same model. The host system is fedora 15 with the virt-preview repo enabled. The disks for the guests use virtio and caching is set to none. My problem is that I get very different results when I benchmark I/O in these guests even though I shouldn't. Doing a seekmark I get: guest 1: 120 seeks/s guest 2: 220 seeks/s "hdparm -t" shows: guest 1: 100 MB/s guest 1: 160 MB/s What's worse is that when I restart the guests the results change and suddenly guest 1 is a lot faster and guest 2 is a lot slower however as long as the guests are running repeating the benchmarks give consistent results. When I test the disks on the host directly I get: seekmark: /dev/sdb: 75 seeks/s /dev/sdc: 75 seeks/s hdparm -t: /dev/sdb: 140 MB/s /dev/sdc: 140 MB/s What bugs me is not so much the absolute numbers (for now) but the fact that these guests give so wildly inconsistent results. Even if the jump from 75 seeks/s to 120 seeks/s from host to guest is explainable by the way block i/o is handled in the virtualization layer I would still expect both guests to return similar results and I would also expect to see similar results across restarts of a single guest. I've attached the definition for both guests even though they are virtually identical. Does anyone have an idea what's happening here? Regards, Dennis

5 10

Emergency help needed on host network randomly stop working.
by Magicloud Magiclouds 13 Oct '11

13 Oct '11

Hi, This is a Centos 5.5 host with one xen guest. About 2 weeks ago, the host randomly lost network connection. By this I mean I could not connect to the services on it, or ping it. Also was the status of the guest. From serial console, I connected to the host, trying to see what happened. No clue (any error messages) in messages or dmesg. ifdown/ifup the interface did not help, either. Only rebooting was my only choice. Searching through Google, I got the information that some other guys met similar problem, and resolved by setting stp on with the bridge interface. I set it, too. And the problem still occurs. Any idea what I should check now? Thanks. -- 竹密岂妨流水过山高哪阻野云飞

3 3

KVM and shared disc...
by Gilberto Nunes 09 Oct '11

09 Oct '11

Hi all I have deploy some VM's with MS Windows 2008 R2. I deploy it using virt-manager, on CentOS 6. On the same host where I installed CentOS I have installed LIO-Target to, in order to providing iSCSI storage to VM's, 'cause I need install Windows Cluster failover here. I was work with iSCSI initiator from inside each VM's but this pratice show me very instable situations. Now, I use iscsi-initiator direct from host, on physical linux server and I try to seek the iSCSI storage from that host... I was capable to found the device and attached into VM settings, via virt-manager. But, when I try to start the VM, I get this message: libvirtError: internal error Process exited while reading console log output char device redirected to /dev/pts/2 qemu-kvm+ -device lsi,id=scsi0,bus=pci.0,addr=0x7: Parameter 'driver' expects a driver name Try with argument '?' for a list Someone has take this error someday??? Thanks -- Gilberto Nunes (47) 8861-6672 (47) 9676-7530 msn: gilbertonunesferreira(a)hotmail.com Skype: gilberto.nunes36 --

1 0

Should I switch and if so what is the procedure
by Rich 07 Oct '11

07 Oct '11

Since the Xen and Linux kernel people have finally made peace and Xen is going to be included with the kernel, should I keep using the Xen virtual server with Centos or should I switch to KVM? I am running Centos 5.7 now. I guess the real question is can I still use Xen with Centos 6?

10 13

kvm-qemu: unable to execute QEMU command savevm (monitor missing?)
by Jason Brian Friedrich 05 Oct '11

05 Oct '11

System: CentOS Linux release 6.0 (final) Kernel: 2.6.32-71.23.1.el6.x86_64 KVM: QEMU PC emulator version 0.12.1 (qemu-kvm-0.12.1.2) Libvirt: ibvirtd (libvirt) 0.8.1 Hi everyone, I only recently subscribed to this list and hope you can shed some light on the following error. I created a VM on my Centos 6 KVM machine, used a qcow2 image and wanted to create a snapshot via 'virsh snapshot-create' command: // [root@kvmhost ~]# virsh snapshot-create server01 error: internal error unable to execute QEMU command 'savevm': The command savevm has not been found \\ I googled before the post, found some [0] threads [1], but could not find an answer how to solve the problem. If the kvm-qemu lacks the support of a savevm monitor, how can I add one? Do I need to recompile kvm-qemu with special flags or is simply a RPM package or a module missing? Thanks in advance, - Jason [0] http://lists.centos.org/pipermail/centos-virt/2011-July/002557.html [1] https://www.redhat.com/archives/libvirt-users/2011-August/msg00011.html

2 3

overlay image on centos 6 do not work
by Kai Krueger 04 Oct '11

04 Oct '11

Hello, I have some problems with overlay images on a host running centos 6.0 * I have 2 physical hosts, one running centos 5.5, the other running centos 6.0 * I created a virtual linux host with a raw file image centos6.img I can start this vm on both machines. * then I created an overlay of the file image with qemu-img create -b centos6.img -f qcow2 centos6.ovl when I start the VM from the ovelay on the centos6 machine I get: --- Booting from Hard Disk... Boot failed: not a bootable disk No bootable device --- When booting from the raw-image and adding the overlay as second disk I see the following: ========== cut here ========== fdisk -l Disk /dev/vda: 8598 MB, 8598323200 bytes 255 heads, 63 sectors/track, 1045 cylinders Units = cylinders of 16065 * 512 = 8225280 bytes Sector size (logical/physical): 512 bytes / 512 bytes I/O size (minimum/optimal): 512 bytes / 512 bytes Disk identifier: 0x000bb12f Device Boot Start End Blocks Id System /dev/vda1 * 1 1046 8395776 83 Linux Disk /dev/vdb: 0 MB, 262144 bytes 16 heads, 63 sectors/track, 0 cylinders Units = cylinders of 1008 * 512 = 516096 bytes Sector size (logical/physical): 512 bytes / 512 bytes I/O size (minimum/optimal): 512 bytes / 512 bytes Disk identifier: 0x00000000 Disk /dev/vdb doesn't contain a valid partition table ========== cut here ========== As one can see, the overlay image does not have any partition table. Starting the vm with exactly these images (per nfs-mount) on the centos-5.5 host works properly. Is my problem an known bug or is it an old feature that has been removed? Best regards, Kai Krueger ---------------------------------------------------------------- This message was sent using IMP, the Internet Messaging Program.

1 0

Strange problem with guest having more than 3 virtual interfaces.
by Veli-Pekka Kestilä 02 Oct '11

02 Oct '11

Hi, I have 6 DomU's with 4-virtual interfaces and 1 DomU with only one. When starting the DomU's in order where the domain with only one interface is started last it refuses to start with following error. If I start it first there is no problem starting them all. ---Error--- [2011-10-02 18:37:33 xend.XendDomainInfo 6728] DEBUG (XendDomainInfo:1335) XendDomainInfo.handleShutdownWatch [2011-10-02 18:37:33 xend 6728] DEBUG (DevController:179) Waiting for devices vif. [2011-10-02 18:37:33 xend 6728] DEBUG (DevController:184) Waiting for 0. [2011-10-02 18:37:33 xend 6728] DEBUG (DevController:571) hotplugStatusCallback /local/domain/0/backend/vif/8/0/hotplug-status. [2011-10-02 18:39:13 xend 6728] DEBUG (DevController:94) Lock released [2011-10-02 18:39:13 xend 6728] DEBUG (DevController:96) Lock not acquired, cannot release [2011-10-02 18:39:13 xend 6728] ERROR (SrvBase:88) Request wait_for_devices failed. Traceback (most recent call last): File "/usr/lib64/python2.4/site-packages/xen/web/SrvBase.py", line 85, in perform return op_method(op, req) File "/usr/lib64/python2.4/site-packages/xen/xend/server/SrvDomain.py", line 72, in op_wait_for_devices return self.dom.waitForDevices() File "/usr/lib64/python2.4/site-packages/xen/xend/XendDomainInfo.py", line 2583, in waitForDevices self.waitForDevices_(c) File "/usr/lib64/python2.4/site-packages/xen/xend/XendDomainInfo.py", line 1563, in waitForDevices_ return self.getDeviceController(deviceClass).waitForDevices() File "/usr/lib64/python2.4/site-packages/xen/xend/server/DevController.py", line 181, in waitForDevices return map(self.waitForDevice, self.deviceIDs()) File "/usr/lib64/python2.4/site-packages/xen/xend/server/DevController.py", line 192, in waitForDevice raise VmError("Device %s (%s) could not be connected. " VmError: Device 0 (vif) could not be connected. Hotplug scripts not working. [2011-10-02 18:39:13 xend.XendDomainInfo 6728] DEBUG (XendDomainInfo:2387) XendDomainInfo.destroy: domid=8 [2011-10-02 18:39:13 xend 6728] DEBUG (DevController:94) Lock released [2011-10-02 18:39:13 xend 6728] DEBUG (DevController:94) Lock released [2011-10-02 18:39:13 xend.XendDomainInfo 6728] INFO (XendDomainInfo:2528) Dev 51712 still active, looping... [2011-10-02 18:39:14 xend.XendDomainInfo 6728] INFO (XendDomainInfo:2528) Dev 51712 still active, looping... [2011-10-02 18:39:14 xend.XendDomainInfo 6728] INFO (XendDomainInfo:2528) Dev 51712 still active, looping... ---Error--- Greetings, Veli-Pekka Kestilä

1 0

2024

2023

2022

2021

2020

2019

2018

2017

2016

2015

2014

2013

2012

2011

2010

2009

2008

2007

virt October 2011