Hi Ulli,
we've dropped rawio (in due to security reasons) and didn't face any
problems. The lxc host seems to do all of necessary /proc operations. We
also mounted /proc filesystem ro in containers.
Regards,
Andreas
Am 24.10.2013 09:19, schrieb Ulli Horlacher:
> So far, I drop these capabiliti
You have to run the tests on both containers at the same time, then you
will see the difference.
If no container need IO, your first container will get the same speed as
your second.
Regards,
Andreas
Am 17.10.2013 10:54, schrieb autumn_sky_is:
Hi, i'm using lxc in my project, i want to control
Hi.
I think the best is the traditionally way between two machines: Connect
via ssh and do what you want to.
Its not a good idea in my eyes to break through the security isolation
of containers because that could affect all of them.
(something bad like getting root of lxc host and kill process)
if we set
lxc.network.hwaddr.
Sometimes an ip won't get configured for the first time but that repairs
a arping or pinging the gateway inside container.
Regards,
Andreas
Am 11.10.2013 14:58, schrieb Tamas Papp:
> On 10/11/2013 02:42 PM, Andreas Laut wrote:
>> Hi,
>>
>> actually can
Sorry, found the mistake in my lxc config by myself, need to do further
tests.
Am 11.10.2013 14:42, schrieb Andreas Laut:
Hi,
actually can't get lxc nightly compiled with debian right now,
configure has problems with pkg-config python3-dev (package pkg-config
and python3-dev is inst
this helps.
Our lxc-host bridge is configured like
"auto br0
iface br0 inet static
bridge_ports eth4
bridge_stp off
address 10.5.255.80
netmask 255.255.0.0
gateway 10.5.255.252"
Andreas
Am 11.10.2013 10:45, schrieb Tamas Papp:
On 10/11/2013 10
g. a bridge on the lxc host and veth's on the
> containers?
>
> A guess might be that you have a MAC address clash; did you override the
> lxc.network.hwaddr?
>
> Guido
>
>> -Original Message-
>> From: Andreas Laut [mailto:andreas.l...@spark5.d
Dear list,
we are using lxc 0.8 on Debian Wheezy (official debian package).
Now we wanted to start a lxc with more than one IP address and we have
gotten strange behaviors.
As starting the lxc some IPs are reachable, some not. If we shut down
the lxc and boot again, some other IPs are reachable
Hi,
thats usual behavior with ubuntu/debian and no need to worry about.
The rootfs mount entry is fake, not real. (you can test this using the
command "umount rootfs" or type "mount")
Regards,
Andreas
Am 05.10.2013 06:50, schrieb Kalyana sundaram:
Hi
I run a group of lxc containers over a u
You can also symlink to the path you want.
rmdir /var/lib/lxc
ln -s /mywantedpath /var/lib/lxc
Regards,
Andreas
Am 04.10.2013 07:40, schrieb Tamas Papp:
> On 10/04/2013 06:03 AM, Kalyana sundaram wrote:
>> Hi
>> lxc by default creates rootfs and fstab on /var/lib/lxc
>> Is it possible to use some
Hi list.
I've tried to mount an external nfs mount into a container. But I got
the following error message after lxc-start:
lxc-start: Invalid argument - failed to mount 'nfsserver:/srv/services'
on '/usr/lib/x86_64-linux-gnu/lxc//home'
And my lxc.mount.entry looks like:
nfsserver:/srv/services
Ubuntu 13.04 comes with lxc 0.9?
I used same version (from debian testing) on wheezy and I have no
problem with read only proc.
And my mount options are the same as yours, strange.
lxc.mount.entry = proc
/srv/vserver/vs-db01-dev/rootfs/proc proc nodev,noexec,nosuid,ro 0
Dear list.
I've got the following error message with Debian squeeze (Kernel
2.6.32.5) and lxc 0.8:
"lxc-start: No such file or directory - failed to rename cgroup
/sys/fs/cgroup//lxc/18381->/sys/fs/cgroup//lxc/vs-db"
lxc 0.7xx creates a folder under /sys/fs/cgroup/[init-process-id] and
ren
Hi list,
usually lxc container mounting proc and sysfs read-write. With this
configuration one container can easily kill the host system and all the
running containers on it. (as both are global)
So we think about mounting proc and sysfs read-only.
Our test server/container runs smoothly and do
Quoting Andreas Laut (andreas.l...@spark5.de):
>> Hi list,
>>
>> usually lxc container mounting proc and sysfs read-write. With this
>> configuration one container can easily kill the host system and all the
>> running containers on it. (as both are global)
>>
>&
15 matches
Mail list logo