Hi list,
I've a question on freezer.c.
In freeze_unfreeze() is a while(1) loop which write()s "FROZEN" or
"THAWED" until read() returns the same.
Why do we need this loop?
It looks like a hack to me.
Thanks,
//richard
--
On Mon, Oct 11, 2010 at 9:42 AM, Daniel Lezcano wrote:
> On 10/11/2010 08:48 AM, richard -rw- weinberger wrote:
>>
>> Hi list,
>>
>> I've a question on freezer.c.
>> In freeze_unfreeze() is a while(1) loop which write()s "FROZEN" or
>> &qu
Hi,
Can KSM (http://lwn.net/Articles/306704/) help LXC?
I guess there are a lot of identical pages when logs of containers are running.
--
Thanks,
//richard
--
The modern datacenter depends on network connectivity to ac
On Tue, Feb 8, 2011 at 12:10 PM, Guido Trotter wrote:
> On Mon, Feb 7, 2011 at 9:04 PM, Daniel Lezcano wrote:
>> On 02/07/2011 08:19 PM, richard -rw- weinberger wrote:
>>> Hi,
>>>
>>> Can KSM (http://lwn.net/Articles/306704/) help LXC?
>>> I guess th
On Sat, Feb 19, 2011 at 7:54 PM, Hugh Dickins wrote:
> On Sat, 19 Feb 2011, supercilious.d...@gmail.com wrote:
>>
>> Is there a way to enable KSM globally for all eligible pages in the system
>> such that applications need not call madvise() themselves? If not, is there
>> a way to do so on behalf
hi!
i'm trying to use lxc 0.7.4 on centos 5.5 (kernel 2.6.37.3).
lxc-start fails when i configure a network.
e.g:
lxc-start: failed to create vethivvZMG-vethGjB0lG : Operation not supported
lxc-start: failed to create netdev
lxc-start: failed to create the network
lxc-start: failed to spawn 'c0'
On Fri, Mar 11, 2011 at 7:55 AM, Daniel Lezcano wrote:
> On 03/10/2011 10:45 PM, richard -rw- weinberger wrote:
>>
>> hi!
>>
>> i'm trying to use lxc 0.7.4 on centos 5.5 (kernel 2.6.37.3).
>>
>> lxc-start fails when i configure a network.
>&
hi,
i'm using this minimal config:
lxc.utsname = c0
lxc.mount.entry = none /lxc/c0/proc proc defaults 0 0
lxc.rootfs = /lxc/c0/
but lxc does not mount proc into /lxc/c0/proc.
using lxc.mount = /path/to/fstab also does not work.
it's also not only proc, it does not mount anything.
mounting within
On Fri, Mar 11, 2011 at 10:06 PM, Michael Tokarev wrote:
> 11.03.2011 22:47, richard -rw- weinberger пишет:
>> hi,
>>
>> i'm using this minimal config:
>> lxc.utsname = c0
>> lxc.mount.entry = none /lxc/c0/proc proc defaults 0 0
>> lxc.rootfs = /lxc/c0/
On Fri, Mar 11, 2011 at 10:13 PM, Daniel Lezcano wrote:
> On 03/11/2011 08:47 PM, richard -rw- weinberger wrote:
>>
>> hi,
>>
>> i'm using this minimal config:
>> lxc.utsname = c0
>> lxc.mount.entry = none /lxc/c0/proc proc defaults 0 0
>> lxc.r
On Tue, Mar 22, 2011 at 10:20 AM, Nathan McSween wrote:
> Can I get a quick rundown of what is implemented w.r.t UID/GID
> containerization, is it safe yet to give containerized root to an
> everyday user without huge security issues?
Drop all dangerous capabilities and mount /proc read-only.
H
Hi!
Is there any use case where a user wants to omit "-f"?
Yesterday I've wasted a few hours with finding out why LXC set up a
broken container.
Then I found out that I've created it using "lxc-create -n foo
foo.cfg" instead of "lxc-create -n foo -f foo.cfg".
--
Thanks,
//richard
--
On Tue, May 29, 2012 at 12:06 AM, Stéphane Graber wrote:
> At least in Ubuntu -f isn't mandatory because calling lxc-create without
> it will simply default to a system wide configuration in /etc/lxc/lxc.conf.
Is this really useful?
My openSUSE system does not have /etc/lxc/lxc.conf,
so lxc-creat
On Tue, May 29, 2012 at 12:36 AM, Stéphane Graber wrote:
> On 05/28/2012 06:29 PM, richard -rw- weinberger wrote:
>> On Tue, May 29, 2012 at 12:06 AM, Stéphane Graber
>> wrote:
>>> At least in Ubuntu -f isn't mandatory because calling lxc-create without
>>&
On Tue, May 29, 2012 at 6:56 AM, Michael Tokarev wrote:
>> Yesterday I've wasted a few hours with finding out why LXC set up a
>> broken container.
>> Then I found out that I've created it using "lxc-create -n foo
>> foo.cfg" instead of "lxc-create -n foo -f foo.cfg".
>
> Shouldn't it complain abo
On Tue, May 29, 2012 at 10:36 AM, richard -rw- weinberger
wrote:
> Found the problem, patch is in the way!
>
*grr*, forgot that this list is moderated.
--
Using this patch is it possible to catch usage errors like:
lxc-create -n foo foo.cfg
Signed-off-by: Richard Weinberger
---
src/l
On Sun, Jun 17, 2012 at 11:46 AM, cheetah wrote:
> Hi guys,
>
> I am a newbie to lxc and preparing to deploy it in my production environment
> to give each user a container. I have the following two concerns now.
>
> 1. Can user load kernel modules in the guest container without influencing
> the
On Sun, Jun 17, 2012 at 12:13 PM, cheetah wrote:
> Richard, thanks a lot for the clear answer. ;p. Would you recommend openvz
> if it is a hostile environment? What is the answer to the above two
> questions if it is openvz?
I've never used openvz.
So I cannot give you a proper answer. :-)
--
T
Hi!
I'm on Linux 3.8 and lxc 0.9.0.
How are the user namespaces in lxc supposed to work?
I've created an opensuse instance using "lxc-create -n foo -t
/usr/share/lxc/templates/lxc-opensuse" and the added these two lines
to the config:
lxc.id_map = u 0 10 1
lxc.id_map = g 0 10 1
B
On Tue, Apr 9, 2013 at 5:28 AM, Serge Hallyn wrote:
> Quoting richard -rw- weinberger (richard.weinber...@gmail.com):
>> Am I missing something obvious?
>
> lxc-create does not yet convert the rootfs to the mapped uids, so you
> need to do that manually using uidmapshift. Che
On Tue, Apr 9, 2013 at 9:58 AM, richard -rw- weinberger
wrote:
> On Tue, Apr 9, 2013 at 5:28 AM, Serge Hallyn wrote:
>> Quoting richard -rw- weinberger (richard.weinber...@gmail.com):
>>> Am I missing something obvious?
>>
>> lxc-create does not yet convert the roo
On Tue, Apr 9, 2013 at 3:19 PM, Serge Hallyn wrote:
> Quoting richard -rw- weinberger (richard.weinber...@gmail.com):
>> On Tue, Apr 9, 2013 at 9:58 AM, richard -rw- weinberger
>> wrote:
>> > On Tue, Apr 9, 2013 at 5:28 AM, Serge Hallyn
>> > wrote:
>
On Tue, Apr 9, 2013 at 11:07 PM, Serge Hallyn wrote:
> Quoting richard -rw- weinberger (richard.weinber...@gmail.com):
>> On Tue, Apr 9, 2013 at 3:19 PM, Serge Hallyn wrote:
>> > Quoting richard -rw- weinberger (richard.weinber...@gmail.com):
>> >> On Tue, Apr 9
On Tue, Apr 9, 2013 at 11:33 PM, richard -rw- weinberger
wrote:
> On Tue, Apr 9, 2013 at 11:07 PM, Serge Hallyn wrote:
>> Quoting richard -rw- weinberger (richard.weinber...@gmail.com):
>>> On Tue, Apr 9, 2013 at 3:19 PM, Serge Hallyn
>>> wrote:
>>&g
On Wed, Apr 10, 2013 at 8:55 PM, Serge Hallyn wrote:
> Quoting richard -rw- weinberger (richard.weinber...@gmail.com):
>> This one?
>> https://launchpad.net/~ubuntu-lxc/+archive/kernel/+packages
>>
>> I'm not an Ubuntu guy and not very familiar with this ppa thin
On Thu, Apr 11, 2013 at 5:03 PM, Eric W. Biederman
wrote:
> richard -rw- weinberger writes:
>
>> On Thu, Apr 11, 2013 at 7:03 AM, Eric W. Biederman
>> wrote:
>>> richard -rw- weinberger writes:
>>>> {st_mode=S_IFCHR|0644, st_rdev=makedev(5, 1), ...})
On Thu, Apr 11, 2013 at 7:02 PM, Eric W. Biederman
wrote:
> Serge Hallyn writes:
>
>> Quoting Eric W. Biederman (ebied...@xmission.com):
>> After creating the /dev/ttyN we chown them to the root uid inside the
>> container. I've not had failures with this.
>
> Yes that should work fine.
>
> Ther
On Fri, Apr 12, 2013 at 3:37 PM, Serge Hallyn wrote:
> Quoting richard -rw- weinberger (richard.weinber...@gmail.com):
> Note that adding '-l info -o debug.out' will generate a lot of helpful
> debug info in the file debug.out.
>
> Running lxc-start under strace simply do
On Sun, Apr 14, 2013 at 12:14 AM, richard -rw- weinberger
wrote:
> Serge, has magic is hidden in your lxc package? :-)
-EBADENGLISH
Should read ", what magic is hidden in your lxc package"
Serge, can you please try an unmodified lxc-0.9.0 package on your setup?
--
Than
On Sun, Apr 14, 2013 at 5:44 PM, richard -rw- weinberger
wrote:
> On Sun, Apr 14, 2013 at 12:14 AM, richard -rw- weinberger
> wrote:
>> Serge, has magic is hidden in your lxc package? :-)
>
> -EBADENGLISH
>
> Should read ", what magic is hidden in your lxc package&q
The function below is broken, it does never return if a path looks
like "//foo/bar" and corrupts the thread stack.
HINT: dirname("//") does _not_ return "/".
This is most likely the reason why I saw all the strange things.
Debugging this was not fun because valgrind and gdb were useless because th
On Thu, Apr 18, 2013 at 10:48 AM, Stéphane Graber wrote:
> On 04/18/2013 10:34 AM, Stéphane Graber wrote:
>> On 04/18/2013 10:32 AM, Stéphane Graber wrote:
>>> This reverts commit 8de4140644f01180f2fdab55b0ab0f13d1c761c6.
>>>
>>> This commit was preventing container startup on my machine, making t
On Thu, Apr 18, 2013 at 10:32 AM, Stéphane Graber wrote:
> This reverts commit 5a5c35c3a01afec515e688c8366e6f893985518d.
>
> This commit was preventing startup of containers using lxc hooks and
> shutdown of all other containers, requiring the use of a good old
> kill -9 to get rid of lxc-start af
On Thu, Apr 18, 2013 at 8:07 PM, S.Çağlar Onur wrote:
> Hi Richard,
>
> That happened to me yesterday. I think you can easily reproduce the issue by
> just starting a container, logging into it and calling poweroff. You will
> see that it will stuck at the end of it. Alternatively you can use
> py
My patch installed a SIGCHLD handler but I oversaw that the executed hooks
will also produce a SIGCHLD. And if signal_handler(), installed by lxc_poll()
does not see the final SIGCHLD lxc-init my not exit at all.
The whole issue is much more complex than I thought.
Anyway, I'll sent a p
On Thu, Apr 18, 2013 at 8:08 PM, Stéphane Graber wrote:
> He only acked the second patch which made it extremely late to the
> mailing-list (Serge got it early as he was directly CCed on it).
>
>> 3. If you encounter problems with one of my patches the lest thing you
>> can do is CC'ing me.
>
> We
36 matches
Mail list logo
