Bryson Lee | 1 Mar 17:03 2012

OpenVZ on Power?

Hi,

 

I’m looking for an alternative to BLCR to provide checkpoint/restart functionality for a Linux application on IBM Power hardware. Having seen some mentions that OpenVZ supports Power, I wanted to try it out.

 

I’ve tried to rebuild vzkernel-2.6.32-042stab049.6.src.rpm in Mock on an IBM JS-12 blade (Power6) running Fedora 12, and have run into a number of problems.  I’ll note that we have successfully rebuilt the Fedora12 2.6.32-based kernel from SRPM in the same Mock configuration.

 

The initial issue was that the patch-042stab049 introduced a single line of invalid syntax into arch/powerpc/include/asm/elf.h:

 

export struct page *vdso32_pages[1];

 

Correcting “export” to “extern” resulted in a type-redefinition compile error, since vdso32_pages is defined as IIR “static unsigned int” in the PPC vdso.c.

 

Removing the extern declaration from elf.h entirely, since apparently the symbol usage in the cpt/cpt_mm.h is ifdef’d by CONFIG_X86 revealed another stumbling block with undefined functions [un]charge_beancounter_fast() due to CONFIG_BEANCOUNTERS not getting defined.  I added appropriate no-op definitions to the group already present in the #ifndef CONFIG_BEANCOUNTERS section of kernel/include/bc/beancounters.h, but there appears to be a larger problem in that the contents of config-vz aren’t getting reflected in the final kernel config used during the RPM build.

 

My basic question is whether or not there’s any hope of successfully generating a ppc64 OpenVZ kernel.  I tried the stable RHEL5 kernel SRPM as well, but encountered a different build failure.

 

I note that the last e.g. vzctl version that has an RPM download for ppc64 is 3.0.26 from 2/27/2011, and that the next minor release 3.0.26.1 from about a week later has no mention of Power at all.   I reviewed the –announce, -user, and –devel list archives from that timeframe, and didn’t see any explicit mention of support for Power being dropped.

 

Is ppc[64] still a supported architecture for OpenVZ?  If so, is checkpoint/restart available?  How should I go about building a kernel (and, eventually the utilities) for my Fedora12 systems?

 

Thanks in advance,

 

-Bryson Lee

_______________________________________________
Users mailing list
Users@...
https://openvz.org/mailman/listinfo/users
Kirill Korotaev | 1 Mar 18:06 2012

Re: OpenVZ on Power?

1. It should be pretty easy to make OpenVZ compilable/running on PowerPC. Typically it takes a day or so in
the worst case since there is almost no code depending on platform, except for maybe syscalls and it's numbers.

2. However, checkpoint restart is not supported on PowerPC platform (originally we supported x32/x86-64
and IA64 platforms, then dropped IA64). Again, it should be pretty straightforward as 95% of state is not
platform dependent. However, nobody really asked before...

Thanks,
Kirill

On Mar 1, 2012, at 20:03 , Bryson Lee wrote:

> Hi,
>  
> I’m looking for an alternative to BLCR to provide checkpoint/restart functionality for a Linux
application on IBM Power hardware. Having seen some mentions that OpenVZ supports Power, I wanted to try
it out.
>  
> I’ve tried to rebuild vzkernel-2.6.32-042stab049.6.src.rpm in Mock on an IBM JS-12 blade (Power6)
running Fedora 12, and have run into a number of problems.  I’ll note that we have successfully rebuilt
the Fedora12 2.6.32-based kernel from SRPM in the same Mock configuration.
>  
> The initial issue was that the patch-042stab049 introduced a single line of invalid syntax into arch/powerpc/include/asm/elf.h:
>  
> export struct page *vdso32_pages[1];
>  
> Correcting “export” to “extern” resulted in a type-redefinition compile error, since
vdso32_pages is defined as IIR “static unsigned int” in the PPC vdso.c.
>  
> Removing the extern declaration from elf.h entirely, since apparently the symbol usage in the
cpt/cpt_mm.h is ifdef’d by CONFIG_X86 revealed another stumbling block with undefined functions
[un]charge_beancounter_fast() due to CONFIG_BEANCOUNTERS not getting defined.  I added appropriate
no-op definitions to the group already present in the #ifndef CONFIG_BEANCOUNTERS section of
kernel/include/bc/beancounters.h, but there appears to be a larger problem in that the contents of
config-vz aren’t getting reflected in the final kernel config used during the RPM build.
>  
> My basic question is whether or not there’s any hope of successfully generating a ppc64 OpenVZ kernel.  I
tried the stable RHEL5 kernel SRPM as well, but encountered a different build failure.
>  
> I note that the last e.g. vzctl version that has an RPM download for ppc64 is 3.0.26 from 2/27/2011, and that
the next minor release 3.0.26.1 from about a week later has no mention of Power at all.   I reviewed the
–announce, -user, and –devel list archives from that timeframe, and didn’t see any explicit
mention of support for Power being dropped.
>  
> Is ppc[64] still a supported architecture for OpenVZ?  If so, is checkpoint/restart available?  How
should I go about building a kernel (and, eventually the utilities) for my Fedora12 systems?
>  
> Thanks in advance,
>  
> -Bryson Lee
> <ATT00001.c>
Raghavendra Bhat | 2 Mar 04:29 2012
Picon

Re: venet ip question

Gerhard Gappmeier posts

>> problem is I run out of external IPs and want to use the external IPs of CT0 (eth1) for new containers.
>> Is this possible?

You may want to read this relevant doc at the OpenVZ Wiki

http://wiki.openvz.org/Using_NAT_for_VE_with_private_IPs

A similar issue was resolved here using this method, but you may have
to fine tune your run-level scripts to
make this rule persistent across reboots.

YMMV
--

-- 
Ragu Bhat
Help stop net censorship in India.
Censorship of Google, Yahoo, Facebook and Twitter has to stomped down!
Dave Lawrence | 2 Mar 15:23 2012

Set disk quota on arbitary mount point


Hello

If I do
mount -n -t simfs /somepath /vz/root/101/mnt/test -o /mnt/test
Then
vzctl exec 101 df

It tells me that the moint /mnt/test has 10GB of data.
How can I set the quotas on this mount point?

Specifically what I am trying to do is allow a container to access the
private area of another container eg
mount -n -t simfs /vz/private/200 /vz/root/101/mnt/test -o /mnt/200

CT200 would not be running in this example.  I would like to be able
to set the moint point at /mnt/200 to have the same disk quota limits
as are set for CT200.  This allows CT101 to act as a rescue shell for
CT200.  Currently my rescue shell works just fine, but only because
none of our users yet have more than 10GB of data.

An alternative approach would be if there was a way of mounting
/vz/root/200 without actually starting CT200, and then use a bind mount:
mount --bind /vz/root/200 /vz/root/101/mnt/200

I don't think mounting a container without starting it is currently
supported, but this would be a relatively straightforward change to
vzctl?  I could submit a patch if I get it working and you think it
would be a worthwhile feature.
Dave Lawrence | 2 Mar 15:41 2012

Re: Set disk quota on arbitary mount point


On 02/03/12 14:23, Dave Lawrence wrote:

> I don't think mounting a container without starting it is
> currently supported,

In fact it's totally supported
vzctl mount 200

I'm not sure where I got that idea from.
Ragu Bhat | 3 Mar 05:04 2012
Picon

Re: CT console feature


Kir announces:

>> Use vzctl attach CTID to attach to this container's console, and you
>> will be able to see all the messages CT init is writing to console

Real cool feature, thanks for your work on improving vzctl and friends.

--

-- 
Regards
Raghavendra Bhat
Help stop Net Censorship in India
Stomp down on Censorship of Google, Yahoo, Facebook and Twitter!
John Knight | 4 Mar 00:55 2012

New Linux distribution with OpenVZ features

Greetings,

CCT is announcing the general availability of CCT Enterprise Linux 6.2.   CCT Enterprise Linux is derived from upstream EL6 source packages with support for OpenVZ, KVM and LXC built in. 

We have added desktop features to our initial in-house builds (pre 6.2) so that it might also be suitable as a workstation/desktop for OpenVZ development, testing and hosting.  We're publicly making it available over the weekend for anyone to download and use indefinitely. 

If you require a more official support channel for this project, CCT offers this service with a support contract.  More information at http://www.classiccitytelco.com/

We originally built CCT EL to run our OpenVZ/PBX product called VoiceGrid and want to make the OS software available to everyone.  Easy to install and easy to restore. 

We love the OpenVZ project and invite you to try out this new distribution. 

Downloads (livedvd and netinstall installation media) are available here:  http://www.classiccitytelco.com/?page_id=488

Thanks!

--

John Knight
Classic City Telco LLC
Email: john-M27W7+ckyYZugXRYqdZ4B1aTQe2KTcn/@public.gmane.org | Main: (706) 995-0200
Direct: (706) 995-0201 | Mobile: (706) 255-9203

CCT Enterprise Linux 6 is released! Click here to learn more.

_______________________________________________
Users mailing list
Users@...
https://openvz.org/mailman/listinfo/users
Steven Crothers | 6 Mar 00:20 2012
Picon

Simple OpenVZ load average aggregation script

I needed something to show me load averages of each individual VE quickly.

Figured I would share it with the community in case someone else was looking for similar functionality.

https://gist.github.com/1981920

Disclaimer: Yes, I'm aware I probably don't follow bash best practices, however I don't really care since its under 50 lines anyway :)

--
Steven Crothers


_______________________________________________
Users mailing list
Users@...
https://openvz.org/mailman/listinfo/users
Kir Kolyshkin | 6 Mar 00:43 2012

Re: Simple OpenVZ load average aggregation script

If I am not mistaken (can't check right now), vzlist have a field for that.

On Mar 6, 2012 3:27 AM, "Steven Crothers" <steven.crothers <at> gmail.com> wrote:
I needed something to show me load averages of each individual VE quickly.

Figured I would share it with the community in case someone else was looking for similar functionality.

https://gist.github.com/1981920

Disclaimer: Yes, I'm aware I probably don't follow bash best practices, however I don't really care since its under 50 lines anyway :)

--
Steven Crothers

_______________________________________________
Users mailing list
Users-GEFAQzZX7r8dnm+yROfE0A@public.gmane.org
https://openvz.org/mailman/listinfo/users

_______________________________________________
Users mailing list
Users@...
https://openvz.org/mailman/listinfo/users
Martynas Venckus | 6 Mar 00:52 2012
Picon

Re: Simple OpenVZ load average aggregation script

vzlist -o veid,hostname,laverage

On 3/6/12, Kir Kolyshkin <kir@...> wrote:
> If I am not mistaken (can't check right now), vzlist have a field for that.
> On Mar 6, 2012 3:27 AM, "Steven Crothers"
<steven.crothers@...> wrote:
>
>> I needed something to show me load averages of each individual VE quickly.
>>
>> Figured I would share it with the community in case someone else was
>> looking for similar functionality.
>>
>> https://gist.github.com/1981920
>>
>> Disclaimer: Yes, I'm aware I probably don't follow bash best practices,
>> however I don't really care since its under 50 lines anyway :)
>>
>> --
>> Steven Crothers
>> steven.crothers@...
>>
>>
>> _______________________________________________
>> Users mailing list
>> Users@...
>> https://openvz.org/mailman/listinfo/users
>>
>>
>

Gmane