Jeff Dike <[EMAIL PROTECTED]> writes:
> However, the rest of that will be true (with some crossed fingers on
> the performance part since I have no numbers yet).
How much slower is UML than Xen anyway?
I've always found it pretty good.
By the way... I've been trying to setup a UML based startu
"Dave Pearson" <[EMAIL PROTECTED]> writes:
> Following up the discussion last week, I've made some time & am in the
> process of moving over to a JSPWiki UML Wiki site. The old wiki is at
> olduml.harlowhill.com, the new one at uml.harlowhill.com or
> uml.jfdi.org. Take your pick. I'll migrate the
Dave Pearson <[EMAIL PROTECTED]> writes:
> happy to stay on though - its hosted on my companies server & has plenty
> of grunt & bandwidth available. Just lacking a 'competent' admin!
Are you sure??
Everytime I try and use it I have interminable wait times - just to
download pages.
--
Nic Fer
Jonas Meyer <[EMAIL PROTECTED]> writes:
> However, under edgy eft, there is no inittab. It appears to have been
> phased out as part of the transition from sysvinit to upstart. Anyone
> got any ideas about how I might be able to screen into my machine and
> log in?
How about this?
Take the VM
Blaisorblade <[EMAIL PROTECTED]> writes:
> What do you think?
Move it please.
It's so slow that I can't even consider making
updates.
I would *really* like to contribute to a UML wiki. But I want to do it
without dealing with TCP timeouts if possible.
--
Nic Ferrier
http://www.tapse
Blaisorblade <[EMAIL PROTECTED]> writes:
> Ok, if I recall correctly (but I can't get anything to agree on this, I'm
> testing on an x86_64 box however sadly) this last mapping is in an earlier
> position (by one page) than it is usually, and UML uses up to the last page;
> having matching sett
Blaisorblade <[EMAIL PROTECTED]> writes:
>> > Beyond that, I've found your bug. You can cast a (struct sockaddr*) to a
>> > (struct sockaddr_in*), but your doing a worse cast; since ifr_addr is a
>> > struct sockaddr,
>>
>> I don't think I'm doing that:
>>
>> ifreq.ifr_addr
>>
>> is the same a
Blaisorblade <[EMAIL PROTECTED]> writes:
> On Friday 13 October 2006 05:33, Nic James Ferrier wrote:
>> I've patched the ipv4 stack so that it does proc registration of the
>> IP address for an iface.
> Take a look at /proc/net/if_inet6, generated by
I did take a
I know this is not UML specific... but maybe the kernel hackers here
can answer this anyway, since I'm doing it to make life with UMLs
easier.
I've patched the ipv4 stack so that it does proc registration of the
IP address for an iface.
Currently I'm just getting the details for eth0.
I'm using
Jeff Dike <[EMAIL PROTECTED]> writes:
> On Fri, Oct 13, 2006 at 12:08:57AM +0100, Nic James Ferrier wrote:
>> Ok. I didn't really mean that... I meant that there's a lot of spam
>> being delivered via the list. That's bad - the list should stop it as
>>
"Ryan Finnie" <[EMAIL PROTECTED]> writes:
>> Not if it boots off a different host it won't.
>
> You quoted everything in my message except the part where I
> essentially say ", unless it's on a different host".
Apologies. I didn't mean to misrepresent you.
> Besides, the automatic generation
Flavio <[EMAIL PROTECTED]> writes:
> Yes, of course, but sometimes our email-address is visible.
> I usually see my mail address when I'm watching a list Archive.
> So spammer can get it and use for their aims.
>
> but.. It's only my opinion.. I'm not a mailing-list expert.
Ok. I didn't really me
Flavio <[EMAIL PROTECTED]> writes:
> I think spam is very boring.. unfortunately we can do anything.
Nonsense. If we hosted the list somewhere else we could get built in
spam filtering.
--
Nic Ferrier
http://www.tapsellferrier.co.uk for all your tapsell ferrier needs
---
"Ryan Finnie" <[EMAIL PROTECTED]> writes:
> I took a simpler route for pseudo-random non-varying mac address
> generation. Here you go:
>
> echo $(hostname)${UMID} | md5sum | \
> awk '{print "00:08:93:"substr($1,1,2)":"substr($1,3,2)":"substr($1,5,2)}'
>
> This combines the umid you expect you as
Is the list archive broken?
There is also a *lot* of spam coming in here.
--
Nic Ferrier
http://www.tapsellferrier.co.uk for all your tapsell ferrier needs
-
Using Tomcat but need to do more? Need to support web services,
In my quest for the perfect network setup for a VM here is a piece of
code that will allocate mac addresses from a pool.
It's not that polished yet. So far you can only allocate based on the
pool:
fe:fd:12:aa:bb:cc
where aa:bb:cc are the pool. However, that is 17 million different mac
addresse
Blaisorblade <[EMAIL PROTECTED]> writes:
> Wait - I think this specific case does not need any work with dentries, it is
> different from the crash I described. I was giving generic suggestions -
> afterwards I looked at sys_getdents and found that this one was easier.
> Finish this patch this
Blaisorblade <[EMAIL PROTECTED]> writes:
>> but how hard can it
>> be?
>
> That's a lie, face it... it _is_ hard. Being a very smart programmer helps,
> but as you say documentation is lacking, so you become able to write good
> code when you're able to face lack of documentation.
I meant that
Blaisorblade <[EMAIL PROTECTED]> writes:
> That is not a bad attempt, and you show having some uncommon experience with
> kernel code. I'm not reviewing it line-by-line - I should check the APIs of
> VFS calls and usage examples, but I haven't the time.
Cheers. I've never written kernel code be
Blaisorblade <[EMAIL PROTECTED]> writes:
>> Is there a reason that can't be done?
> No, it can be done. Produce a working patch and then that will be ok, and I
> guess it will be appreciated (at least by me). It isn't IMHO that much
> trivial however - the code will be probably be similar but no
Jeff Dike <[EMAIL PROTECTED]> writes:
> Given that you need to find the dhcp client and count on it having the
> lease file open on a particular descriptor, I would call this not very
> nice.
/8->
I think I can alter mconsole_kern::mconsole_proc so that it returns
data for directories the same w
Blaisorblade <[EMAIL PROTECTED]> writes:
>> After that it should work.
> Note that I forgot it in some recent -bb/-bs releases, so you'll maybe need
> to
> take it from older ones.
I just tested it on 2.6.17.1 which is not so new now.
>> Thanks for pointing it out though. It's a really useful
=?iso-8859-1?Q?Ferm=EDn_Gal=E1n_M=E1rquez?= <[EMAIL PROTECTED]>
writes:
> but, in spite of the message, it seems not being working because the file
> "foo" doesn't get created in the virtual machine filesystem.
>
> Googling for "uml_mconsole exec", I found the following at
> http://www.user-mode-l
Gosh this is frustrating.
I'm trying to work out a better way of starting umls whilst using
dhcp address allocation.
Currently I approach this by having the uml fs ssh the ip address it
gets from dhcp to a known host.
Doing that requires a lot of infrastructure though. It's quite
fragile.
What
Has anyone tried this?
I just tried it and the booted UML kernel seems to hang after starting
VFS.
--
Nic Ferrier
http://www.tapsellferrier.co.uk for all your tapsell ferrier needs
-
Take Surveys. Earn Cash. Influence th
Blaisorblade <[EMAIL PROTECTED]> writes:
> For UML you should enable CONFIG_HOST_2G_2G, if UML is recent enough that it
> is used by SKAS0. With -bb1 you have more detailed settings
> (CONFIG_HOST_VMSPLIT_*).
Bingo!
Thanks.
--
Nic Ferrier
http://www.tapsellferrier.co.uk for all your tapsel
Hi UML people
I'm trying to run a uml kernel v 2.6.17.1 on my laptop.
The same kernel is working perfectly well on 2.6.12 host machine.
My laptop is a 2.6.16-2 machine.
When I turn SKAS mode on the kernel fails:
mapping mmap stub failed, errno = 12
Kernel panic - not syncing: start_usersp
"Itamar Reis Peixoto" <[EMAIL PROTECTED]> writes:
> anyone know a web-page to download UBUNTU 6.06 root filesystem
>
> or a good how to to create image.
You can use the debootstrap method with UBUNTU.
Something like:
mkdir myubuntu
debootstrap breezy myubuntu
dd if=/dev/zero of=./myub
Jeff Dike <[EMAIL PROTECTED]> writes:
> When you're DHCP-ing, you need to provide a MAC on the command line
> because that's the only way that the DHCP server can identify the
> client.
No it isn't. DHCP can arbitrate between clients based on MAC or:
- sent host name
- sent dhcp client identifie
Paolo Carpo <[EMAIL PROTECTED]> writes:
> I would like to know if exist any documents about how UML manage the CPU: how
> the CPU of the real PC is devided between the istances of the UML VMs. Is
> there a benchmark or something?
I've done some testing on (de)nice-ing the kernel thread processe
[EMAIL PROTECTED] writes:
> I beg to differ. Here is some mac-addresses which I have on my
> running umls and their respective ip:
>
> fe:fd:ac:10:2a:0d 172.16.42.13
> fe:fd:c0:a8:02:04 192.168.2.4
> fe:fd:ac:10:2a:0a 172.16.42.10
> fe:fd:c0:a8:02:02 192.168.2.2
>
> The paramet
[EMAIL PROTECTED] writes:
> In the host side of tap device the mac address seems to be random
> but it seems like in guest side the mac address is derived from
> first given ip-address and when using dhcp it's always 0.0.0.0 which
> causes it to be fe:fd:00:00:00:00. The problem is that I can't se
[EMAIL PROTECTED] writes:
> I call it with "eth2=tuntap,ext-test". I got the impression that it
> is enough to do that when I'm using preconfigured interface. At
> least there is example of that in Virtual Networking Tutorial. And
> besides ip-address would be little hard to set as I'm connecting
[EMAIL PROTECTED] writes:
> I have a problem with running ISC DHCP3 client in uml guest.
> Everytime dhcp-client runs the MAC-address is set to
> FE:FD:00:00:00:00 regardless what I order it to use.
Are you starting the guest with the correct tuntap specifier?
Specifying the mac on the guest
Has anyone considered this?
If cow files were just a little bit more portable it would be possible
to do version control with them.
This would be neat. Especially for complex software systems. It would
integrate the OS with the version control.
eg: need a particular version of a particular libra
Is there any point at all in having a swap file assigned to a UML
virtual machine? (other than testing swap)
Since the UML is obtaining it's memory from the host virtual memory I
figure that it is better to simply assign a larger memory to each UML
instance than to assign each UML a swap file.
Do
"Simon Matthews" <[EMAIL PROTECTED]> writes:
> Can anyone suggest a good way to start a UML instance from the host's
> initscripts? The host is a RedHat EL3 machine.
>
> It seems that the UML won't start if it can't attach to a terminal, so
> unless I can stop this behaviour I don't immediately s
")
Message-ID: <[EMAIL PROTECTED]>
MIME-Version: 1.0
Content-Type: text/plain; charset=us-ascii
"Daniel Lundberg" <[EMAIL PROTECTED]> writes:
> This might be a really, really stupid question, but I'm not in-tune
> with what UML really does so please excuse if so.
>
> I'm on an UML server via my I
")
Message-ID: <[EMAIL PROTECTED]>
MIME-Version: 1.0
Content-Type: text/plain; charset=us-ascii
"[EMAIL PROTECTED]" <[EMAIL PROTECTED]> writes:
> Hi,
>
> I'm starting a degree thesis about uml. I tryed to compile it
> using 2.6.16.18 kernel (this is the last, ya?).
> Compiled it I start
> with
")
Message-ID: <[EMAIL PROTECTED]>
MIME-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Nic James Ferrier <[EMAIL PROTECTED]> writes:
> My second question is: can uml_switch help me make tuntap's on the fly
> and connect them all together?
Never mind.
The c
")
Message-ID: <[EMAIL PROTECTED]>
MIME-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Sometime ago we had this conversation about how to do dynamic creation
of UMLs, particularly for laptop development environments.
Blaisorblade <[EMAIL PROTECTED]> writes:
> b) pre-setup a tap0 interf
")
Message-ID: <[EMAIL PROTECTED]>
MIME-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Blaisorblade <[EMAIL PROTECTED]> writes:
> On Tuesday 16 May 2006 23:39, Jason Lunz wrote:
>> [EMAIL PROTECTED] said:
>> > Even in the other solution (PPP over SSH), or in whichever one you can
>> > th
")
Message-ID: <[EMAIL PROTECTED]>
MIME-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Nic <[EMAIL PROTECTED]> writes:
> John Parsons <[EMAIL PROTECTED]> writes:
>
>> --- Nic <[EMAIL PROTECTED]> wrote:
>>>
>>> You could just set up VPNs between the VMs using PPP over SSH.
>>>
>>
>> I'm
43 matches
Mail list logo
