This is an archived copy of the Xen.org mailing list, which we have preserved to ensure that existing links to archives are not broken. The live archive, which contains the latest emails, can be found at http://lists.xen.org/
Home Products Support Community News


[Xen-devel] Maximum number of DomUs? (x86_64)

To: "xen-devel@xxxxxxxxxxxxxxxxxxx" <xen-devel@xxxxxxxxxxxxxxxxxxx>
Subject: [Xen-devel] Maximum number of DomUs? (x86_64)
From: Florian Kirstein <xenlist@xxxxxxxxxxxxxx>
Date: Sun, 9 Apr 2006 03:37:26 +0200
Delivery-date: Sat, 08 Apr 2006 18:37:46 -0700
Envelope-to: www-data@xxxxxxxxxxxxxxxxxx
List-help: <mailto:xen-devel-request@lists.xensource.com?subject=help>
List-id: Xen developer discussion <xen-devel.lists.xensource.com>
List-post: <mailto:xen-devel@lists.xensource.com>
List-subscribe: <http://lists.xensource.com/cgi-bin/mailman/listinfo/xen-devel>, <mailto:xen-devel-request@lists.xensource.com?subject=subscribe>
List-unsubscribe: <http://lists.xensource.com/cgi-bin/mailman/listinfo/xen-devel>, <mailto:xen-devel-request@lists.xensource.com?subject=unsubscribe>
Sender: xen-devel-bounces@xxxxxxxxxxxxxxxxxxx
User-agent: Mutt/

I just tried to push a quad-opteron server with 16GB of memory to the limit
and I reached it a bit earlier than expected. It seems that no matter
how much memory I assign to each DomU, and wether i use loopback or
phy/lvm devices for the storage, or no vif/storage at all, I can't
crate more than 107 DomUs (yes, that's not a real problem, I'm just
curious :) After No. 107 that I get(*):
Using config file "vm107".
Started domain VM107
Using config file "vm108".
Error: (12, 'Cannot allocate memory')
Using config file "vm109".
Error: (12, 'Cannot allocate memory')

but of course there's still memory left in Domain-0 to balloon out:

Name                              ID Mem(MiB) VCPUs State  Time(s)
Domain-0                           0     9215     4 r-----   319.3
VM1                                1       64     1 ------     0.8
VM10                              10       64     1 -b----     0.7
VM100                            100       64     1 -b----     0.4

When I assign 128MB per DomU exactly the same effect:
Name                              ID Mem(MiB) VCPUs State  Time(s)
Domain-0                           0     2372     4 r-----  2196.3
VM1                              129      128     1 -b----     0.4
VM10                             138      128     1 -b----     0.4
VM100                            228      128     1 -b----     0.2

So I wonder, which structure could run out of memory there? Is this a known
limitation? Didn't find much about the maximum number of DomUs
in the list archives... The Dom0 is running a CentOS4 x86_64 and I don't
even need a DomU image to reach this limit, network and diskless DomUs
just running into an "no root filesystem" panic being "preserve"d after
the crash have the same effect.

(:ul8er, r@y
(*) yes, I know, I could have used a vmid parameter instead of creating
100s of configfiles :)

Xen-devel mailing list

<Prev in Thread] Current Thread [Next in Thread>