Re: [CentOS-virt] OS-level virtualization using LXC and systemd-nspawn containers
Greetings, - Original Message - > Can you share your experience with LXC and/or systemd-nspawn > for RHEL 8 / CentOS 8 operating system on the hardware node? > I can't use host network for [system] containers. > Each container must have its own private network. In that case, perhaps you'd like docker/podman's private networking? > Backuping persistent containers and restoring from backup - issue. > I don't want have deal with a mash of different images and layers. I haven't thought of backups. I assume there are a number of backup solutions for docker/podman containers but I'm completely ignorant. > Each my systemd-nspawn container located in separate filesystem: > > # zfs list > NAME USED AVAIL REFER MOUNTPOINT > tank 531G 1.13T 96K /tank > tank/containers 528G 1.13T 168K /tank/containers > tank/containers/119.1G 1.13T 8.00G /tank/containers/1 Ok, so you are turning off SELinux and using ZFS too? And you still want to stay with EL? Why? > Upstream also doesn't support ZFS, but this is extraordinary file system > with excellent feature set. Ubuntu and LXD do support ZFS and Canonical's lawyers seem happy to allow ZFS to be bundled with Ubuntu by default. You should get along nicely. TYL, -- Scott Dowdle 704 Church Street Belgrade, MT 59714 (406)388-0827 [home] (406)994-3931 [work] ___ CentOS-virt mailing list CentOS-virt@centos.org https://lists.centos.org/mailman/listinfo/centos-virt
Re: [CentOS-virt] OS-level virtualization using LXC and systemd-nspawn containers
Greetings, - Original Message - > > LXD is a management layer on top of it which provides for easy > > clustering and even managing VMs. I think it is the closest thing > > to vzctl/prlctl from OpenVZ. > > "Yes, you could use LXC without LXD. But you probably would not want to. > On its own, LXC will give you only a basic subset of features. > For a production environment, you’ll want to use LXD". Have you tried LXD? Again, I'd only recommend it on Ubuntu LTS and I believe your target is CentOS so that is probably why you are excluding it, eh? > podman is replacement for Docker, > it is not replacement for OpenVZ 6 containers. Docker definitely targets "Application Containers"... with one service per container. podman says they can also do "System Containers" by running systemd as the entry point. Of course the vast majority of pre-made container images you'll find in container image repositories aren't built for that, but you can use distro provided images and build a system container image out of them. I have a simple recipe for Fedora, CentOS, and Ubuntu. I don't know how many people are using podman in this capacity yet, and I don't know if it is mature or not for production... but the limited testing I've done with it, has worked out fairly well... using Fedora or CentOS Stream 8 as the host OS... and yes, even running the container as a regular user after doing: setsebool -P container_manage_cgroup on Yes, podman does still use it's own private network addressing, but I guess that can be overcome by telling it to use the host network. I haven't tried that. Not exactly like OpenVZ's container networking for sure. > I have containers with 1.6 TiB of valuable data - podman > not designed to work in this mode and in such conditions. Persistent data really isn't an issue. You just have to understand how it works. Plenty of people run long-term / persistent-data Docker and podman containers... although granted, most folks say if you are using persistent data containers, you are doing it wrong. I guess I prefer to do it wrong. :) > So I have only two alternatives for OS-level virtualization: > LXC or systemd-nspawn. If CentOS is your target host, I'd guess that neither of those really are a good solutions... simply because they aren't supported and upstream doesn't care about anything other than podman for containers. LXC varies from one distro to the next... with different kernels, and different versions of libraries and management scripts. Again, LXD on an Ubuntu LTS host is probably the most stable... with Proxmox VE as a close second. Both of those upstreams care about system containers and put in a lot of effort to make it work. Good luck. TYL, -- Scott Dowdle 704 Church Street Belgrade, MT 59714 (406)388-0827 [home] (406)994-3931 [work] ___ CentOS-virt mailing list CentOS-virt@centos.org https://lists.centos.org/mailman/listinfo/centos-virt
Re: [CentOS-virt] OS-level virtualization using LXC and systemd-nspawn containers
Greetings, - Original Message - > OpenVZ 7 has no updates, and therefore is not suitable for > production. The free updates lag behind the paid Virtuozzo 7 version and plenty of people are using it in production. I'm not one of those. > LXC/LXD is the same technology, as I understand from > linuxcontainers.org linuxcontainers.org is owned by Canonical and yes it documents LXC... but LXD is a management layer on top of it which provides for easy clustering and even managing VMs. I think it is the closest thing to vzctl/prlctl from OpenVZ. > podman can't be a replacement for OpenVZ 6 / systemd-nspawn because > it destroys the root filesystem on the container stop, and all > changes made in container configs and other container files will be lost. > This is a nightmare for the website hosting server with containers. No, it does NOT destroy the delta disk (that's what I call where changes are stored) upon container stop and I'm not sure why you think it does. You can even export a systemd unit file to manage the container as a systemd service or user service. volumes are a nice way to handle persistence of data if you want to nuke the existing container and make a new one from scratch without losing your data. While it is true you have to approach the container a little differently, podman systemd containers are fairly reasonable "system containers". TYL, -- Scott Dowdle 704 Church Street Belgrade, MT 59714 (406)388-0827 [home] (406)994-3931 [work] ___ CentOS-virt mailing list CentOS-virt@centos.org https://lists.centos.org/mailman/listinfo/centos-virt
Re: [CentOS-virt] OS-level virtualization using LXC and systemd-nspawn containers
Greetings, - Original Message - > I found only two possible free/open source alternatives for OpenVZ 6: > > - LXC > - systemd-nspawn Some you seem to have overlooked?!? 1) OpenVZ 7 2) LXD from Canonical that is part of Ubuntu 3) podman containers with systemd installed (set /sbin/init as the entry point) I use LXC on Proxmox VE (which I guess should be #4 above) some although I primarily use it for VMs. Oh, LXD is supposedly packaged for other distros but given that they aren't much into SELinux and they are into snaps, I'd not really recommend it outside of Ubuntu. TYL, -- Scott Dowdle 704 Church Street Belgrade, MT 59714 (406)388-0827 [home] (406)994-3931 [work] ___ CentOS-virt mailing list CentOS-virt@centos.org https://lists.centos.org/mailman/listinfo/centos-virt
Re: [CentOS-virt] Apparent discontinuity between advertised centos7 release 1803_01 and content of centos-release file
Greetings, - Original Message - > Hello, > > I searched centos7 in the AWS marketplace for the > at-time-of-writing-latest centos7 image: > https://aws.amazon.com/marketplace/pp/B00O7WM7QW?qid=1524138193326=0-1_=srh_res_product_title > > I built a standard free tier t2.micro from this putative 1803_01 AMI. > I see from the docs, this is thus a March 2018 compilation. > When I get CLI, I get this: > > [centos@ip-172-31-27-32 etc]$ cat centos-release > CentOS Linux release 7.4.1708 (Core) > > which suggests that the AMI I just launched was built on a version of > centos compiled from upstream sources in August 2017. > Does the release of centos7 on the marketplace sport a release > version which does not pertain to the machine's release? Unlike with > releases published outside AWS, the Centos cloud page does not > specify version info as listed on AWS. > > Thanks v much IA, RHEL releases about every 6 months... and then CentOS lags behind a little bit as a rebuilder. The most current release of CentOS is based on RHEL 7.4... and is dated August 2017 (1708). RHEL 7.5 was released a week or so ago and CentOS is frantically working on coming out with CentOS release based on it. When that comes out, the latest release will be 1804 (or 1805, etc). That doesn't mean there aren't in-between release updates because there are... but they don't re-number the whole release because of regular updates. CentOS has a lot of products that they produce and some of them may be rebuilt and distributed more frequently (like CentOS Atomic Host or their Vagrant image, etc)... but not the oldest, main product. Did that answer your question? TYL, -- Scott Dowdle 704 Church Street Belgrade, MT 59714 (406)388-0827 [home] (406)994-3931 [work] ___ CentOS-virt mailing list CentOS-virt@centos.org https://lists.centos.org/mailman/listinfo/centos-virt
Re: [CentOS-virt] LXC on CentOS 7 HowTo: PAM Configuration
Greetings,
- Original Message -
> I am trying to implement something like an "LXC on CentOS 7 HowTo"
> for internal use. (Might as well get public afterwards.) I am following
> the HowTo for CentOS 6
> (https://wiki.centos.org/HowTos/LXC-on-CentOS6). So, here's what I
> did so far (Steps 1-6 can easily be omitted, but I am trying to be
> complete.)
Do you want to use the libvirt tools or the lxc-{whatever} tools?
I haven't worked with LXC on EL6 nor EL7 much at all... but I have been playing
with it some on Fedora 23.
Anyway, to create a CentOS container, the lxc tools can do a lot of the work
for you... and I don't know that all of the steps are needed from that wiki...
at least if you use the lxc tools rather than libvirt... although you'll still
use libvirt for it's networking stuff.
To create a CentOS 7 container:
lxc-create -t download -n {desired-name}
That should give you a list of available Templates... and you would type in:
Distribution: centos
Release: 7
Architecture: amd64
It should download the template and put it under /var/cache/lxc/ and create the
container under /var/lib/lxc/.
The Template should just work and not require any fiddling with... I'm hoping.
LXC is still rather lacking in isolation features as it does not give the
container a subset of /proc... so within the container you can see all of the
RAM and disk... and your root user can do bad things if you don't trust them.
That is with a "privileged" container. Supposedly there is a way to run a
container as a user and then grant capabilities as needed to reduce the
security footprint but I don't know much about that.
Docker is a subset of that design for Applications (rather than the full distro
with an init system of its own) that provides a really nice image library and
image builder... but unless you are trying to do fleet computing (aka
microservices) then Docker really isn't the container I've been looking for.
If you want privileged containers you don't have to worry about, you'll most
likely want tp create an OpenVZ host (warning, third-party repo / kernel /
tools needed). The current stable version of OpenVZ is "OpenVZ Legacy" which
is EL6-based. They have been working hard on "Virtuozzo 7" which is merger of
OpenVZ and the upstream Virtuozzo product-line still offering a FLOSS
version... that is based on EL7 and also provides KVM VM management along-side
of containers. They are trying to integrate Virtuozzo support into libvirt and
the libvirt-based tools like virsh and virt-manager... and get as much of that
work upstreamed as possible... and switch from the kernel-patch based
checkpoint code they have in OpenVZ Legacy to the mostly upstreamed CRIU C/R.
Hopefully in the next 3-6 months Virtuozzo 7 will go GA. They basically have
created a complete distro for it which is based on CentOS.
I'd be interested to hear of the lxc tools work for you or not. The little bit
I tried them on EL7 I seemed to get journald CPU max-outs on the host node.
TYL,
--
Scott Dowdle
704 Church Street
Belgrade, MT 59714
(406)388-0827 [home]
(406)994-3931 [work]
___
CentOS-virt mailing list
CentOS-virt@centos.org
https://lists.centos.org/mailman/listinfo/centos-virt
Re: [CentOS-virt] Hostname inside lxc container
Greetings, - Original Message - > I have installed a CentOS6 lxc guest under a Debian 8.x LXC host. > All it is working ok but I can't change the hostname for the centos6 lxc > container (it is using the same hostname from Debian host). I have > modifyed HOSTNAME under /etc/sysconfig/network and /etc/hosts file, > but it doesn't works. > > Do I need to change anything else?? If I remember correctly, it is generally set in /etc/sysconfig/network on a regular EL6 host. If that isn't working when run within an LXC container on a Debian host... ask whoever packaged up the LXC container template you used. There is a lot of variance in LXC from one distro to another. Some recipes say to install from a .tar.gz file (which I prefer to call an OS Template) others say to do a chroot install using a distro's native package manager. My point is, that LXC tools and methods really vary... and I don't think the CentOS project can help much in this situation. For containers, I've been using OpenVZ (now called OpenVZ Legacy while they work on the newer Virtuozzo 7) for 10 years now... and they have a standard set of OS Templates for a range of distros and everything (generally) works as it is supposed to... and even with that... the CentOS Project isn't interested in helping OpenVZ users running CentOS containers... because it uses a non-CentOS provided kernel and management tools... although the recommended OpenVZ Legacy hostnode distro is CentOS 6.x. Virtuozzo 7 is its own distro rebased from CentOS 7. One container technology they are interested in supporting is Docker (app containers) especially when using the official CentOS Docker images built/provided by the CentOS Project... running on a CentOS host. TYL, -- Scott Dowdle 704 Church Street Belgrade, MT 59714 (406)388-0827 [home] (406)994-3931 [work] ___ CentOS-virt mailing list CentOS-virt@centos.org https://lists.centos.org/mailman/listinfo/centos-virt
Re: [CentOS-virt] Can KVM and VirtualBox co-exist on same host?
Greetings, - Original Message - A supplemental question: Is there any way to convert a VB guest image into a KVM guest image? I believe qemu-image can convert between a few different virtualization disk image formats. Of course that just changes the disk image itself... and not the drivers inside... so if you do convert it (I'd recommend working on a copy)... then you'll probably have to pull the VirtualBox guest addons out... and install the KVM guest stuff... but it shouldn't be that difficult. TYL, -- Scott Dowdle 704 Church Street Belgrade, MT 59714 (406)388-0827 [home] (406)994-3931 [work] ___ CentOS-virt mailing list CentOS-virt@centos.org http://lists.centos.org/mailman/listinfo/centos-virt
Re: [CentOS-virt] Fwd: About Centos 7 + Virt-manager
Greetings, - Original Message - while much of what you say is true you somehow could lead unaware users to the conclusion that docker and lxc are two very different container technologies. In fact, docker uses lxc for containers. So it's more a management abstraction layer with an API. Nevertheless for true and secure containerization you'll need openvz atm, sadly it's not in the kernel yet. Docker dropped LXC with version 0.6 or was it 1.0? They have their own library that they use now. Since I can't remember what version they switched, nor what version Red Hat has in RHEL7 (I'll leave it as an exercise for the reader)... perhaps the Docker currently shipping with RHEL7 still uses LXC?!? The bulk of OpenVZ will never make it into the kernel although the Parallels kernel devs do try to integrate existing kernel features into OpenVZ (so the patch becomes smaller over time) as well as getting bits and pieces into the kernel or into userland (criu for example). I wonder how much change OpenVZ will undergo in the port to the RHEL7 kernel... where considerable container building blocks are already part of the kernel. TYL, -- Scott Dowdle 704 Church Street Belgrade, MT 59714 (406)388-0827 [home] (406)994-3931 [work] ___ CentOS-virt mailing list CentOS-virt@centos.org http://lists.centos.org/mailman/listinfo/centos-virt
Re: [CentOS-virt] Fwd: About Centos 7 + Virt-manager
Greetings, - Original Message - Thanks a lot, for your replies, my boss is a big fan of lxc, but I have read many forums, and what I perceive is rhel7 - docker, centos7 --- openvz With great difficulty, I managed a container with virt-manager, I even noticed a bug when trying to create a bridge. Conclusion, as we want to use a container operating system is better to use openvz, now is there interfaces that allow a user no expert reserve resources such as memory, cpu, etc without going to browse cgroups? Just to clarify, the OpenVZ kernel runs fine on RHEL(56) too. In fact I have a couple of RHEL hosts running OpenVZ. So far as resource management goes, in the EL6-based OpenVZ kernel, vSwap-based configuration/management is preferred whereas in the EL5-based kernel (which OpenVZ is EOL'ing in Oct. I think), user_beancounters are what you have. vzctl's resource management facilities do what the vast majority of container users need. vzctl-core is available for non-OpenVZ kernels but it is missing quite a few features compared to when run on an OpenVZ-based kernel. See: https://wiki.openvz.org/Vzctl_for_upstream_kernel I don't think it is well tested on upstream kernels. TYL, -- Scott Dowdle 704 Church Street Belgrade, MT 59714 (406)388-0827 [home] (406)994-3931 [work] ___ CentOS-virt mailing list CentOS-virt@centos.org http://lists.centos.org/mailman/listinfo/centos-virt
Re: [CentOS-virt] Fwd: About Centos 7 + Virt-manager
Greetings, - Original Message - Am 16.07.2014 15:16, schrieb Scott Dowdle: Docker dropped LXC with version 0.6 or was it 1.0? They have their own library that they use now. This is not correct, or the docker docs are out of date: Docker combines these components into a wrapper we call a container format. The default container format is called libcontainer.Docker also supports traditional Linux containers using LXC. Source:[1] [1]https://docs.docker.com/introduction/understanding-docker/#the-underlying-technology Yes, if you read that it says the default container format is called libcontainer. It also supports LXC. Originally it was LXC only. Then they switched to libcontainer... but they say they support several formats (including OpenVZ although I haven't seen any elaboration as to how) and plan to add more in the future. Docs that were written before the move to libcontainer (and in all honesty, I'm not really sure what libcontainer is vs. LXC) won't mention it. Everything after the move, should mention it. For more information about libcontainer... and it was the 0.9 release of Docker that announced it (not 0.6 nor 1.0 as I had previously guessed)... can be found here: http://blog.docker.com/2014/03/docker-0-9-introducing-execution-drivers-and-libcontainer/ TYL, -- Scott Dowdle 704 Church Street Belgrade, MT 59714 (406)388-0827 [home] (406)994-3931 [work] ___ CentOS-virt mailing list CentOS-virt@centos.org http://lists.centos.org/mailman/listinfo/centos-virt
Re: [CentOS-virt] Fwd: About Centos 7 + Virt-manager
Greetings again, - Original Message - Am 16.07.2014 15:16, schrieb Scott Dowdle: Docker dropped LXC with version 0.6 or was it 1.0? They have their own library that they use now. This is not correct, or the docker docs are out of date: Docker combines these components into a wrapper we call a container format. The default container format is called libcontainer.Docker also supports traditional Linux containers using LXC. Source:[1] [1]https://docs.docker.com/introduction/understanding-docker/#the-underlying-technology For anyone not willing to take the time to visit the Docker 0.9 release blog post, here's a snippet from it: Thanks to libcontainer, Docker out of the box can now manipulate namespaces, control groups, capabilities, apparmor profiles, network interfaces and firewalling rules – all in a consistent and predictable way, and without depending on LXC or any other userland package. This drastically reduces the number of moving parts, and insulates Docker from the side-effects introduced across versions and distributions of LXC. Is that more clear? TYL, -- Scott Dowdle 704 Church Street Belgrade, MT 59714 (406)388-0827 [home] (406)994-3931 [work] ___ CentOS-virt mailing list CentOS-virt@centos.org http://lists.centos.org/mailman/listinfo/centos-virt
Re: [CentOS-virt] Fwd: About Centos 7 + Virt-manager
Greetings, - Original Message - Thanks Nux, but now I'm fighting with the network settings... There is some tutorial for this?, How do you create centos template, can you give me some steps for this... Thanks in Advance Pablo On Tue, Jul 15, 2014 at 2:32 PM, Nux! n...@li.nux.ro wrote: Hi, I've got an old image here: http://li.nux.ro/download/LXC/ Use at your own risk etc :-) -- - Original Message - From: Pablo Silva psil...@gmail.com To: centos-virt@centos.org Sent: Tuesday, 15 July, 2014 3:53:45 PM Subject: [CentOS-virt] Fwd: About Centos 7 + Virt-manager Dear Colleagues: I'm trying to use virt-manager to create a container with lxc, as the picture I attached, I indicated that I must point to a directory where the template is to manage it with virt-manager. I tried to use a template of http://wiki.openvz.org/Download/template/precreated , but not working, being on the console option displays a black screen with nothing. I wonder, where can I get a template to work with virt-manager, or what are the steps to follow so you can create a template with Centos Minimal, do not know if the template must be the same version of Centos 7 or can be a template with Centos 6.5 In advance, thank you very much. -Pablo http://picpaste.com/Captura_de_pantalla_2014-07-15_a_las_10.47.26-DHi28gtt.png http://picpaste.com/Captura_de_pantalla_2014-07-15_a_las_10.49.02-y4TkGxLY.png While virt-manager has had an lxc option for some time, and yes, I've actually seen a few tutorials by adventurous Fedorians on Fedora Planet over the last couple of years... to the best of my knowledge, almost no one is using virt-manager with containers. Red Hat only sees containers for applications and they are favoring strongly Docker. Among the many guides released with RHEL7 was one entitled, Resource Management and Linux Containers (https://access.redhat.com/documentation/en-US/Red_Hat_Enterprise_Linux/7/html-single/Resource_Management_and_Linux_Containers_Guide/index.html). It does have a section on LXC containers but they really don't go into it too much. Canonical is the main lead on LXC containers... so maybe they have some documentation? If you want to containerize applications, then Docker might be the way to go. If you want a full-blown container (complete distro with independent accounts, networking, services, etc) then OpenVZ is the way to go. OpenVZ is a third-party patch to the kernel that will never get into the mainline kernel so you have to use an OpenVZ-provided kernel (their stable branches are based on RHEL-kernels) and utils. They have EL5 and EL6-based kernels... and are working on an EL7-based one but no date on when that will be released. I'm a big OpenVZ user (since 2005) so if you have questions, feel free to email me directly if desired... or find me in #openvz on freenode during MST business hours. TYL, -- Scott Dowdle 704 Church Street Belgrade, MT 59714 (406)388-0827 [home] (406)994-3931 [work] ___ CentOS-virt mailing list CentOS-virt@centos.org http://lists.centos.org/mailman/listinfo/centos-virt
Re: [CentOS-virt] Fwd: About Centos 7 + Virt-manager
Greetings, - Original Message - Canonical is the main lead on LXC containers... so maybe they have some documentation? I haven't processed it yet... but they appear to have some documentation. See the LXC section of their Server manual starting on page 323: https://help.ubuntu.com/14.04/serverguide/serverguide.pdf That doesn't help much on Fedora nor CentOS... because LXC varies greatly from kernel to kernel and distro to distro. TYL, -- Scott Dowdle 704 Church Street Belgrade, MT 59714 (406)388-0827 [home] (406)994-3931 [work] ___ CentOS-virt mailing list CentOS-virt@centos.org http://lists.centos.org/mailman/listinfo/centos-virt
[CentOS-virt] How to create an OpenVZ OS Template for CentOS 7 Public QA
Greetings,
First start of by working on a physical system, virtual machine, or container
that matches the OS Template you are wanting to build. I used my CentOS 7
Public QA OS OpenVZ container to build it.
You must of course have a working yum. Once we are beyond Public QA and there
is stuff in /etc/yum.repos.d/ this won't be a problem. One thing to note is
that the --enablerepo= must refer to a repo your build host has and viewable
via yum repolist. That repo should point to the desired CentOS 7 build tree
directory.
A note about the package list. Yes, listed out every individual packcage is
tedious. Perhaps some package groups could be used but they typically drag in
a lot of unwanted additional packages. Suggestions welcome.
Here is a simple script and please don't nag at me because I'm a scripting
novice. I hope email client word wrapping and screen sizes don't butcher it too
bad:
- - - - -
# To get a package list without version numbers from a target system
# rpm -qa --qf %{n} packages.txt
# Put contents of packages.txt after -y install \ line below
mkdir /ostemplate
yum \
--installroot /ostemplate \
--nogpg \
--releasever=7 \
--enablerepo=centos7pubqa \
-y install \
centos-release filesystem ncurses-base mailcap tzdata glibc-common xz-libs \
ncurses-libs pcre libselinux info libdb popt sed libcom_err libuuid expat \
libacl libgpg-error dbus-libs gawk lua libxml2 glib2 shared-mime-info apr cpio \
gmp p11-kit tcp_wrappers-libs perl-parent perl-podlators perl-Text-ParseWords \
perl-Pod-Escapes perl-libs perl-threads perl-constant perl-Filter \
perl-Time-Local perl-threads-shared perl-File-Path perl-Scalar-List-Utils \
perl-Getopt-Long libcap-ng nss-softokn libassuan libunistring diffutils
gpm-libs \
libnfnetlink keyutils-libs gettext-libs p11-kit-trust nettle \
gobject-introspection vim-minimal pinentry make libselinux-utils ncurses \
libverto libsemanage krb5-libs openldap cracklib libmount systemd-libs libuser \
pam libblkid util-linux python-libs dhcp-libs libcurl python-urlgrabber
rpm-libs \
dhcp-common libselinux-python python-iniparse python-chardet
yum-metadata-parser \
python-backports-ssl_match_hostname newt-python pyxattr binutils logrotate \
procps-ng mariadb-libs fipscheck-lib openssh libmnl iptables json-c \
device-mapper cryptsetup-libs dbus iputils cronie-anacron crontabs libestr \
gnupg2 rpm-python pygpgme libnl3 yum-utils man-db dhclient audit openssh-server
\
libgudev1 net-tools elinks python-pyudev policycoreutils python-configobj \
pygobject3-base sudo wget file tar which psmisc libpcap libsysfs libdaemon lzo \
libgcc setup basesystem kbd-misc bind-license nss-softokn-freebl glibc
libstdc++ \
bash libsepol zlib audit-libs nspr chkconfig bzip2-libs nss-util grep libattr \
libcap elfutils-libelf libgcrypt readline libidn libffi pkgconfig sqlite \
groff-base file-libs libtasn1 slang gdbm perl-HTTP-Tiny perl-Pod-Perldoc \
perl-Encode perl-Pod-Usage perl-macros perl-Storable perl-Carp perl-Exporter \
perl-Socket perl-File-Temp perl-PathTools perl-Pod-Simple perl apr-util
libcroco \
cyrus-sasl-lib libgomp kmod-libs libedit hostname js newt ca-certificates less \
dbus-glib acl libdb-utils findutils xz sysvinit-tools ustr nss-tools \
openssl-libs gzip cracklib-dicts nss libpwquality coreutils shadow-utils \
libutempter nss-sysinit python libssh2 python-pycurl curl rpm python-decorator \
python-slip dbus-python python-kitchen python-backports python-setuptools \
pyliblzma centos-logos kmod openssl nss_compat_ossl bind-libs-lite fipscheck \
httpd-tools libnetfilter_conntrack iproute qrencode-libs device-mapper-libs \
systemd systemd-sysv initscripts cronie libpipeline pth rpm-build-libs gpgme
yum \
libnl3-cli rsyslog mlocate kbd postfix httpd ebtables openssh-clients
authconfig \
python-slip-dbus mc gettext screen passwd gnutls elfutils-libs libss nano
snappy \
libndp ethtool hardlink rootfiles
ln -sf /proc/mounts /ostemplate/etc/mtab
# I want Mountain time to be the default
ln -sf /usr/share/zoneinfo/America/Denver /ostemplate/etc/localtime
# Now compress that sucker
cd /ostemplate ; tar -cvJf /root/centos-7-x86_64-viayum.tar.xz . ; cd
ls -lh /root/centos-7-x86_64-viayum.tar.xz
echo Done building OS Template. Now test it.
- - - - -
TYL,
--
Scott Dowdle
704 Church Street
Belgrade, MT 59714
(406)388-0827 [home]
(406)994-3931 [work]
___
CentOS-virt mailing list
CentOS-virt@centos.org
http://lists.centos.org/mailman/listinfo/centos-virt
[CentOS-virt] OpenVZ variant
Greetings, I was reading the LWN article from today (free to non-subscribers next Thursday). Here's a subscriber link for those who might want to see it now: CentOS and Red Hat - http://lwn.net/SubscriberLink/592723/485ea802859f6c36/ I saw that Xen was mentioned as an area where CentOS went beyond RHEL with CentOS 6... and being that I'm deeply in the OpenVZ community, I thought it might be natural to have an OpenVZ CentOS Variant. I just noticed that the CentOS Virt-SIG page already mentions OpenVZ. Is this only for the upcoming CentOS 7 or would it be possible to produce a spin/remix that is CentOS 6-based that includes the OpenVZ kernel and OpenVZ utils? Looking at the stats provided by the OpenVZ Project (http://stats.openvz.org/) it is obvious that CentOS is the most popular platform for both OpenVZ hosts and OpenVZ containers: Top host distros --- CentOS 56,725 Scientific2,471 RHEL869 Debian 576 Fedora 111 Ubuntu 82 Gentoo 54 openSUS 18 ALT Linux10 Sabayon 6 and Top 10 CT distros --- centos 245,468 debian 106,350 ubuntu 83,197 OR8,354 gentoo7,017 pagoda4,024 scientific3,604 fedora3,173 seedunlimited 1,965 This morning I sent out some feelers to the OpenVZ community (via the OpenVZ Users mailing list, blog.openvz.org, and the #openvz IRC channel) to see if any OpenVZ users were already working with the CentOS project (I'm not). So does anyone that is part of this SIG care to tell me how much OpenVZ interest there currently is and how I might become a part of the effort? I know the virt-sig is probably quite broad beyond OpenVZ. TYL, -- Scott Dowdle 704 Church Street Belgrade, MT 59714 (406)388-0827 [home] (406)994-3931 [work] ___ CentOS-virt mailing list CentOS-virt@centos.org http://lists.centos.org/mailman/listinfo/centos-virt
Re: [CentOS-virt] kvm
Mattias, - Original Message - i have installed kvm on centos 6 but how to run it i not use libvirt There is a fine virtualization guide written by Red Hat so check that if you haven't. Basically you DO want to use libvirt in so much as you use a client that uses it. What client? Like virt-manager and/or virsh. Those should be the clues that you need. TYL, -- Scott Dowdle 704 Church Street Belgrade, MT 59714 (406)388-0827 [home] (406)994-3931 [work] ___ CentOS-virt mailing list CentOS-virt@centos.org http://lists.centos.org/mailman/listinfo/centos-virt
Re: [CentOS-virt] create a guest
Greetings, - Original Message - but i allredy have the freebsd disc image file on the server DUH... I thought you were saying you had a disc or disc image of the FreeBSD install media. You already have a FreeBSD VM disk image? In that case, what is it from? Is it a KVM VM or did it come from some other virt platform like VMware, VirtualBox, Parallels, Xen or what? You may run into an issue with drivers if it came from another virt platform that uses product specific tools (like VMware tools for example)... where to make it work correctly you have to remove their tools. This is probably less of an issue with a FreeBSD VM though. There is also v2v which supposedly can convert a disk image of a VM from one product format to another. I haven't used it. There should be good documentation for v2v if you do a search. TYL, -- Scott Dowdle 704 Church Street Belgrade, MT 59714 (406)388-0827 [home] (406)994-3931 [work] ___ CentOS-virt mailing list CentOS-virt@centos.org http://lists.centos.org/mailman/listinfo/centos-virt
Re: [CentOS-virt] create a guest
Greetings, - Original Message - stacklet.com kvm image In that case, what I would do would be to create a new VM with virt-manager but use the disk image file provided. That will basically create /etc/libvirt/qemu/whatever.xml where whatever is the name you gave the VM in virt-manager. Then you can use virt-manager to start, stop, console connect etc... or you can use virsh from the command line. BTW, if the VM is to have a public IP address then you want to setup a bridge device if you don't already have one, and associate the VM with that when you create it. If it is going to use a private IP address, then you can just use the default NAT. KVM is a little complicated to get going with but the effort is definitely worth it. And again, there is good documentation if you do a few searches. TYL, -- Scott Dowdle 704 Church Street Belgrade, MT 59714 (406)388-0827 [home] (406)994-3931 [work] ___ CentOS-virt mailing list CentOS-virt@centos.org http://lists.centos.org/mailman/listinfo/centos-virt
Re: [CentOS-virt] (no subject)
Shawn, - Original Message - On an otherwise completely idle system I've noticed the load to be 1.0 to 1.5 range. Running top shows the culprit to be: qemu-kvm. Is this normal behavior? I would have expected the load to be pretty light. You have to remember, or at least as I understand it, that a load of 1 is full for a single CPU/core. Since you have 24, a full load would be 24. Linux gets even weirder with more cores. I have one system that has 64 cores... and it has a lot of threads/process running just to support those. Another thing that uses quite a bit of CPU is ksm. If you don't have a number of similar VMs then I don't think it is very helpful... and it seems to eat up quite a bit of CPU resources trying to be helpful. Ok, now the uber-CentOS geeks can tell me how stupid I am. Mmmm... go. TYL, -- Scott Dowdle 704 Church Street Belgrade, MT 59714 (406)388-0827 [home] (406)994-3931 [work] ___ CentOS-virt mailing list CentOS-virt@centos.org http://lists.centos.org/mailman/listinfo/centos-virt
Re: [CentOS-virt] turning off udev for eth0
Greetings, - Original Message - I have set up a kvm host and configured a standard clone prototype for generating new guests. One persistent (pun intended) annoyance when cloning is the behaviour of udev with respect to the virtual network interface. The prototype is configured with just eth0 having a dedicated IP addr. When the prototype is cloned udev creates rules for both eth0 and eth1 in the clone. Because eth1 does not exist in the cloned guest one has to manually edit /etc/udev/rules.d/70-persistent-net.rules to get rid of the bogus entries and then restart the clone instance to have the changes take effect. All this does is return the new guest to the prototype eth0 configuration. Is there no way to alter udev's behaviour? Is udev even needed on a server system using virtual hardware? Altering the rules file not a big deal in itself but it adds needless busywork when setting up a new guest. That's how it is on physical machines (I image lab computers and have to clean up the udev rules file among other things) and would expect the same behavior from virtual machines. The limitations of virt-clone are known and are being addressed in virt-sysprep... which hasn't made it to RHEL yet I don't think... but you can find out about it here: http://rwmj.wordpress.com/2011/10/08/new-tool-virt-sysprep/ TYL, -- Scott Dowdle 704 Church Street Belgrade, MT 59714 (406)388-0827 [home] (406)994-3931 [work] ___ CentOS-virt mailing list CentOS-virt@centos.org http://lists.centos.org/mailman/listinfo/centos-virt
Re: [CentOS-virt] Now on to creation of disk images
Greetings, - Original Message - Mr. Heron was so kind to make a suggestion that I should use disk images to install VMs. Upon further thought, I kinda like the idea. So I re-read the manual and google a little, and discover I still don't know what should be in these disk images. Should I copy the contents of the CDs to a file or what? I haven't been following this conversation but I think I can answer that last question. Using a disk image file rather than a physical disk or partition(s) on a physical means that you use the file as if it were a disk. You don't need to put anything on it... and you boot install media and then select the disk image file as the disk you want to use to install your OS too. Either that, or you are talking about using .iso files on disk as install media rather than physical optical media. TYL, -- Scott Dowdle 704 Church Street Belgrade, MT 59714 (406)388-0827 [home] (406)994-3931 [work] ___ CentOS-virt mailing list CentOS-virt@centos.org http://lists.centos.org/mailman/listinfo/centos-virt
Re: [CentOS-virt] Reading the new 6.0 manual - now questions
Greetings, - Original Message - RH makes this an add-on to their license. Does anyone know if the upcoming Centos 6 will provide the virtualization packages (right away or in the future)? Just to clarify... Red Hat's virtualization entitlement is for management/support from RHN. The way they sell RHEL... you can have 1 VM, 4 VMs or unlimited VMs. When I say VMs there I mean supported RHN subscribed RHEL installs where you register them with RHN and they get updates like any RHEL box would. So you are affectively getting 2, 5 or unlimited RHEL update entitlements. This is done by installing an additional package or two in the RHEL VM, and registering it with RHN so it knows it is a VM and RHN knows which physical host it is associated with. If you want to run any number of virtual non-RHEL OSes, go for it. They are not accounted for. The only thing accounted for are RHN subscriptions by physical or virtual machines. It isn't like virt-manager phones home... it does not. None of that entitlement stuff applies to the free RHEL clones so it isn't an issue. Secondly, I'm not sure I understand the CPU allocation stuff. If I have 6 cores, it appears I can only create VMs that use 6 cores total. It is my understanding that you can allocate basically all of the vcpus you want... the only rule though is that you can't assign more vcpus to a single VM than you have physical cpus as the OS sees them. So if you have two quad-core CPUs and they can do multiple threads per core... just look at /proc/cpuinfo to see how many cpus are listed there. It is probably the total number of threads times the total number of cores per CPU times the number of CPUs. You can't go over that number of vcpus in a single VM. So if /proc/cpuinfo on the physical host shows 16 CPUs you could make x number of VMs with 16 or less vcpus each. It doesn't matter what the total number is across VMs. From a performance stand point some might want to allocate only as many vcpus as they have physical cores or cpus and then pin them so they get a 1-to-1 allocation... but for most folks, as long as their hardware isn't bogged down too much, it is a freeforall.:) That's my understanding anyway. TYL, -- Scott Dowdle 704 Church Street Belgrade, MT 59714 (406)388-0827 [home] (406)994-3931 [work] ___ CentOS-virt mailing list CentOS-virt@centos.org http://lists.centos.org/mailman/listinfo/centos-virt
Re: [CentOS-virt] SPICE Benchmark
Greetings, - Original Message - Google's NX implementation is called 'neatx': http://code.google.com/p/neatx/ Thanks. I was looking for that. NX the protocol is open already.. :) It is for all versions before 4.0. 4.0 will be completely closed. TYL, -- Scott Dowdle 704 Church Street Belgrade, MT 59714 (406)388-0827 [home] (406)994-3931 [work] ___ CentOS-virt mailing list CentOS-virt@centos.org http://lists.centos.org/mailman/listinfo/centos-virt
Re: [CentOS-virt] SPICE Benchmark
Greetings, - Original Message - So as I understand it correctly, this whole SPICE thing is just something like VNC on steroids? Why can't we have this SPICE thing work on physical hosts as well? SPICE was specifically designed to be a display protocol for a KVM virtual machine. Most remote display protocols have two pieces... a client and a server. SPICE has three... a client, a server, and a qxl device driver provided inside of a KVM virtual machine. It is probably possible to take the SPICE protocol and adapt it to work without the qxl device driver... but no one has done that yet. The best experience I've seen for a remote Linux box was provided by No Machine's NX protocol. FreeNX comes from NX but it appears FreeNX has stalled. Google created some project, I forget the name, forked from FreeNX I believe. No Machine is working on version 4.0 of NX and supposedly will be releasing a beta in the not too distant future. Some time ago they posted a lot of information about NX 4.0 on their website and it seems to rival SPICE to a certain degree... but that remains to be seen. What I'd like to see happen would be for SPICE to be adapted to a general purpose remote display protocol or perhap Red Hat could buy No Machine and open source that protocol too. :) TYL, -- Scott Dowdle 704 Church Street Belgrade, MT 59714 (406)388-0827 [home] (406)994-3931 [work] ___ CentOS-virt mailing list CentOS-virt@centos.org http://lists.centos.org/mailman/listinfo/centos-virt
Re: [CentOS-virt] SPICE Benchmark
Greetings, - Original Message - I don’t suppose you have any websites you would recommend that shows how to install and use spice? I second that question. While I've found instructions here and there, and have even given Fedora 14 a try as well... the processes is very manual and I have yet to find a very detailed set of instructions for getting SPICE going. I understand that the benchmark paper used RHEV for Desktops... which I assume does all of the work for you... but what about us folks who are using RHEL 5.5, RHEL 6, and/or Fedora 14. All of them come with KVM and SPICE packages but we need some detailed instructions on setting it up and making it work. Thanks in advance for any consideration, -- Scott Dowdle 704 Church Street Belgrade, MT 59714 (406)388-0827 [home] (406)994-3931 [work] ___ CentOS-virt mailing list CentOS-virt@centos.org http://lists.centos.org/mailman/listinfo/centos-virt
Re: [CentOS-virt] SPICE Benchmark
Greetings, - Original Message - If you want just to see SPICE in action it is not hard. You need qemu with SPICE support on server and SPICE client on client. You need to start qemu on server with additional options: -spice port=port,disable-ticketing - use this one if you do not need password protection OR -spice port=port,password=secret - if you need to protect connection After it you can connect from client using spicec -h host -p port That isn't quite all there is to it. What about installing the xorg-x11-drv-qxl package in the guest VM and configuring it in xorg.conf? How exactly is that done? Also I believe spice-server needs to be running on the VM host. I'm kinda going by Fedora 14's SPICE feature page (http://fedoraproject.org/wiki/Features/Spice) as well as a CentOS related one (http://www.geekgoth.de/tag/centos/). Are all the packages one needs to get SPICE going on the VM host included in CentOS 5.5? What packages are those? Are there any steps required for inside of the VM? Simply having the right qemu-kvm and starting up the VM with the -spice flag isn't all there is to it. What many of us need are step by step instructions. TYL, -- Scott Dowdle 704 Church Street Belgrade, MT 59714 (406)388-0827 [home] (406)994-3931 [work] ___ CentOS-virt mailing list CentOS-virt@centos.org http://lists.centos.org/mailman/listinfo/centos-virt
Re: [CentOS-virt] Can KVM be run headless?
Greetings,
- Original Message -
On 11/10/2010 06:46 PM, compdoc wrote:
If I am running a windows client, I would run something like
Ultra VNC server. And on my host, communicate with something
like krdc. Am I correct?
There are a few ways to do it.
If I'm on a Linux box (which I am 99.9% of the time), I ssh -X into the host
running the KVM virtual machines, and run virt-viewer {vm-name} and a
graphical display of the machine pops up. If I just want text access, I ssh
into the VM directly. If I want to run just one graphical app, I might ssh -X
into the VM.
If you are on a Windows client then you can use an ssh client (PuTTY for
example) or if you want the GUI stuff, you can install an Xserver app on
Windows and tunnel the X traffic over ssh. A nice free Xserver for Windows is
Xming.
If you would prefer to run a remote display protocol server inside of the VM
and connect with that you can install vnc-server or No Machine's NX server. NX
is a ton faster but the free version (not open source but free of charge)
limits you to two user connections which is plenty for most people on a server.
Then you install the vnc-client or NX client on your Windows box and use that
to connect to the remote display server in the VM.
How do you start and stop your headless machines?
You can use virt-manager if you have GUI access to the host or you can use the
command line tool virsh. For example:
virsh list --all (Shows all VMs on the host and shows their status)
virsh start {vm-name} (Starts a VM)
virsh shutdown {vm-name} (Shuts down your VM)
You really should check out Red Hat's Virtualizaiton Guide. They have a book
quality guide that explains KVM, virt-manager and virsh.
http://docs.redhat.com/docs/en-US/Red_Hat_Enterprise_Linux/5/html/Virtualization/index.html
RHEL6 just came out today and they have greatly enhanced virt-manager and
KVM... and you can expect a CentOS 6 release in 1 - 2 months... or so goes the
pattern. If you want to see what KVM is like in RHEL6 you can check out their
updated Virtualization Guide:
http://docs.redhat.com/docs/en-US/Red_Hat_Enterprise_Linux/6/html/Virtualization/index.html
TYL,
--
Scott Dowdle
704 Church Street
Belgrade, MT 59714
(406)388-0827 [home]
(406)994-3931 [work]
___
CentOS-virt mailing list
CentOS-virt@centos.org
http://lists.centos.org/mailman/listinfo/centos-virt
Re: [CentOS-virt] Djbdns Working in VPS ?
Ludwig, - Lodewijk christoffel [EMAIL PROTECTED] wrote: It is said at djbdnsrocks.org that it shouln't work on VPS or jails...i'm try it out on my VPS and nothing seems work, at least to see this thing working by using ps -aux. I see on this (http://www.djbdnsrocks.org/single/getting_started.htm) page they say, Virtual private servers (jails) will usually NOT work. That implies that a VPS is a jail. OpenVZ is much, much more than a jail and I see no reason it shouldn't work. TYL, -- Scott Dowdle 704 Church Street Belgrade, MT 59714 (406)388-0827 [home] (406)994-3931 [work] ___ CentOS-virt mailing list CentOS-virt@centos.org http://lists.centos.org/mailman/listinfo/centos-virt
[CentOS-virt] Check out Proxmox VE... can CentOS improve on this?
Greetings, I'm a big OpenVZ fanboy. I've sent a few emails on this list that proves that... and I'm sure I've annoyed some people... but be that as it may... I would like to draw everyone on this list's attention to Proxmox VE. What is Proxmox VE? Here's a review from the end of June: ProxMox: The high-performance virtualization server for the rest of us http://blogs.zdnet.com/BTL/?p=9181 It is a bare-metal Linux distribution based on Debian that has been stripped down to a minimum that includes a kernel that provides support for both KVM and OpenVZ so it does fully virtualized machines (requiring hardware support in the CPU) and containers. It has a nice web-based management system that is fairly feature complete and allows for the easy creation and management of KVM virtual machines and OpenVZ containers. It also has clustering features. That's a long enough description. Ok, how does this differ from everything else out there? 1) It is bare-metal... just pop in the CD (which is a 250MB download)... boot up the machine, answer about two questions... a few minutes worth of install time... and a reboot later... you have what looks very similar to a VMware ESX host... with a console (text) login screen that says... I man a Proxmox VE machine... connect to me at the following URL. Start browsing. 2) It supports both fully virtualized machines AND containers 3) It has a really nice, maturing (two releases so far) web-interface for managing everything 4) It is cluster aware - add additional Proxmox VE machines, use the really simple command line program to make each machine aware of all of the others... and bang the web-interface on any machine sees all of the virtual machines on all nodes 5) The web-interface includes VNC-java-applet based access so you can graphically attach to any virtual machine including the console of an OpenVZ container. For KVM machines, you get to see the BIOS / boot from ISO image before virtual machine is even on the network 6) Proxmox VE shows the potential that exists with a FOSS OS, FOSS Virtualization products... in a completely FOSS product... that is freaking easy to install, setup and use... that works well Problems - Some of the features aren't done yet. Proxmox VE is a little ahead of its time. Live migration doesn't quite work for OpenVZ yet (although it works fine in stock OpenVZ with their stable kernel branches). Proxmox VE is still in beta. Basically, if and when the Proxmox VE concept is fully realized and matured it'll be a killer app that makes business folks go... Linux DUH... for those that care about virtualization... like yoous guyyys on this hea list. Ok... so any chance CentOS or some of the members of the CentOS development community would like to borrow the Proxmox VE idea (perhaps even their web-interface code) and make something like Proxmox VE... but based on CentOS... that supports Xen and OpenVZ? It would have the benefit of being able to run on both systems with and without hardware support for Virtualization - if you have VT you can run fully virtualized Xen AND para-virtualized machines... if not... para-virtualized VMs only... and in both cases OpenVZ containers. The form-factor would have to retain all of the properties I mentioned above... I think... for it to be a huge success. Where to start? Get two machines to test on... desktops are fine as long as they have VT in the CPU. Download the 250MB Proxmox VE iso. Burn disc. Boot disk. Answer two questions. Wait 5 minutes. Reboot. Play with it. See what you think... and use your imagination. If impressed, plan the take over of the world with a similar setup based on CentOS. Then when RHEL6 comes out and KVM is here... switch to KVM. BTW, the OpenVZ Project had a kernel package built on the RHEL kernel that included both Xen and OpenVZ but I can't seem to find it now. Notice I'm not providing any links to Proxmox VE. You have to care enough to google for it. :) TYL, -- Scott Dowdle 704 Church Street Belgrade, MT 59714 (406)388-0827 [home] (406)994-3931 [work] ___ CentOS-virt mailing list CentOS-virt@centos.org http://lists.centos.org/mailman/listinfo/centos-virt
Re: [CentOS-virt] Djbdns Working in VPS ?
Ludwig, There isn't anything about djbdns that should cause it not to run inside of an OpenVZ container. Of course there aren't any easy to install packages for it I don't think. - Lodewijk christoffel [EMAIL PROTECTED] wrote: Hi, I try dig google for this question and found little notes in there, I already try it on my VPS machine and end up with nothing working... I'm using CentOS 4.6 in VPS (openvz), at first i'm trying BIND and it's just nice, and now i want to try djbdns... Is it djbdns working in VPS ? if not, is there anything that i can work on it ? i can only afford VPS for now. Thank you... Regards, Ludwig TYL, -- Scott Dowdle 704 Church Street Belgrade, MT 59714 (406)388-0827 [home] (406)994-3931 [work] ___ CentOS-virt mailing list CentOS-virt@centos.org http://lists.centos.org/mailman/listinfo/centos-virt
Re: [CentOS-virt] Virtuozzo GFS
James, - James Thompson [EMAIL PROTECTED] wrote: I have just finished deploying two Dell PowerEdge 1950s with CentOS 5.1 and Virtuozzo 4. GFS is up and running and Virtuozzo is configured for shared-storage clustering. Everything works adequately but I am wondering if anyone else has experienced load issues like I am seeing. I have three VEs/VMs running, two on one node and one on the other node. One of the VEs on each node are doing very little (one is just idling with apache and mysql and the other is running rsync every six hours). The other is running Zimbra. Every so often load will spike on the node running the Zimbra VE to as high as 2 or 3 then settle down a short while later to around 0.8 or 0.9. During the spikes the node not running Zimbra will other see an increase from its idle load of 0.4 or so up to as high as 1.7 as I have seen. I notice when running top that dlm_send and dlm_recv will jump to the top fairly frequently when these load spikes occur. What I am wondering is whether anyone else has experienced these kind of load scenarios with GFS and what they have done to deal with them? We are hoping to deploy a bit more densely on this setup so I'd like to make any performance adjustments I can at this stage. Thanks, James Thompson A load of 2-3 isn't much at all... so I don't think I'd call that much of a spike. I have run OpenVZ at work and on a hobby server. In both cases I have about 7 containers... one of them being Zimbra. The other 6 containers are fairly busy so the two machines see a decent amount of load. I am NOT using GFS though. What is dlm_send and dlm_recv part of? GFS? TYL, -- Scott Dowdle 704 Church Street Belgrade, MT 59714 (406)388-0827 [home] (406)994-3931 [work] ___ CentOS-virt mailing list CentOS-virt@centos.org http://lists.centos.org/mailman/listinfo/centos-virt
Re: [CentOS-virt] i386 VM on x86_64 host in Xen
Christopher, - Christopher G. Stach II [EMAIL PROTECTED] wrote: It would really suck to have 3795 virtual machines die all at the same time from a single kernel panic. Yes, absolutely it would. I use the OpenVZ kernels that are based on the RHEL4 and RHEL5 kernels and I haven't had any problems with them... just like I haven't had any problems with the stock RHEL4 and RHEL5 kernels... nor CentOS kernels. I usually end up rebooting host node machines because of kernel upgrades... so my machines don't get a chance to have longish uptimes... but on one remote colocation machine I have for hobby stuff... it currently has an uptime of 106 days. It has 7 VPSes on it and they are fairly fat as they all run a full set of services. I know I've been running that machine for close to 2 years now... and if I remember correctly it started out with CentOS 4.0. I've upgraded to each release (on the host node and the VPSes) and am currently at CentOS 4.5. I look foward forward to 4.6. Here's what they look like (ip addresses and hostnames obscured): [EMAIL PROTECTED] ~]# vzlist VEID NPROC STATUS IP_ADDR HOSTNAME 101 53runningxx.xx.xx.xxvps101.localdomain 102 44runningxx.xx.xx.xxvps102.localdomain 103 44runningxx.xx.xx.xxvps103.localdomain 104 32runningxx.xx.xx.xxvps104.localdomain 105322 runningxx.xx.xx.xxvps105.localdomain 106 32runningxx.xx.xx.xxvps106.localdomain 107 29runningxx.xx.xx.xxvps107.localdomain Looking at the number of processes, can you tell which VPS is running Zimbra? :) 6 of the 7 VPSes are CentOS and the remaining 1 is Debian. Speaking of uptimes, I have a legacy machine at work running Linux-VServer on a 2.4.x series kernel. It had the longest uptime of any machine I've had... and was well over 400 days... when a power outage that outlasted its UPS took it down. That particular machine runs three VPSes that are mail relay/frontends and they get pounded... so that uptime is notable. So, my experience has been that physical failures and power failures (although pretty rare) are more common that kernel panics that take down all of my virtual machines. TYL, -- Scott Dowdle 704 Church Street Belgrade, MT 59714 (406)388-0827 [home] (406)994-3931 [work] ___ CentOS-virt mailing list CentOS-virt@centos.org http://lists.centos.org/mailman/listinfo/centos-virt
