headers
Antibozo | 1 Aug 01:24
Favicon

Re: drbd 8 primary/primary and xen migration on RHEL 5

On 2008-07-31 21:58, nathan <at> robotics.net wrote:
> I am running DRBD primary/primary on Centos 5.2 with CLVM and GFS with 
> no problems. The only issue I have with live migration is that the arp 
> takes 10 - 15 sec to get refreshed so you lose connectivity during that 
> time. I have the problem with 3.0ish xen on Centos 5.2 as well as xen 
> 3.2.1.

One can run a job on the vm to generate a packet every second or two to 
resolve this; ping in a loop should do it.

My scenario doesn't involve any clustered filesystem. I'm using phy: 
drbd devices as the backing for the vm, not files. As far as I 
understand things, a clustered filesystem shouldn't be necessary, as 
long as the drbd devices are in sync at the moment migration occurs.

But the question remains whether that condition is guaranteed, and I 
hope to hear from someone who knows the answer to that question...

> Anyway, other then the ARP issue, I have this working in production with 
> about two dozen DomUs.
> 
> Note: If you want to use LVM for xen rather then files on GFS/LVM/DRBD 
> you need to run the latest DRBD that supports max-bio-bvecs.

I'm actually running drbd on top of LVM. But I'll look into the 
max-bio-bvecs thing anyway out of curiosity.

Thanks for the reply.

> On Thu, 31 Jul 2008, Antibozo wrote:
(Continue reading)

Permalink | Reply | 3 comments |
headers
Antibozo | 1 Aug 01:43
Favicon

Re: drbd 8 primary/primary and xen migration on RHEL 5

On 2008-07-31 23:24, Antibozo wrote:
> On 2008-07-31 21:58, nathan <at> robotics.net wrote:
>> I am running DRBD primary/primary on Centos 5.2 with CLVM and GFS with 
>> no problems. The only issue I have with live migration is that the arp 
>> takes 10 - 15 sec to get refreshed so you lose connectivity during 
>> that time. I have the problem with 3.0ish xen on Centos 5.2 as well as 
>> xen 3.2.1.
> 
> One can run a job on the vm to generate a packet every second or two to 
> resolve this; ping in a loop should do it.

Quick follow-up:

arping -b -A *ip-address-of-vm*

seems to work pretty well. I get a 2-3 second dropout if I'm running 
this during live migration.

--

-- 
Jefferson Ogata : Internetworker, Antibozo
<ogata <at> antibozo.net>  http://www.antibozo.net/ogata/
Permalink | Reply | 2 comments |
headers
nathan | 1 Aug 01:58
Favicon
Gravatar

Re: drbd 8 primary/primary and xen migration on RHEL 5

On Thu, 31 Jul 2008, Antibozo wrote:

> On 2008-07-31 23:24, Antibozo wrote:
>> On 2008-07-31 21:58, nathan <at> robotics.net wrote:
>>> I am running DRBD primary/primary on Centos 5.2 with CLVM and GFS with no 
>>> problems. The only issue I have with live migration is that the arp takes 
>>> 10 - 15 sec to get refreshed so you lose connectivity during that time. I 
>>> have the problem with 3.0ish xen on Centos 5.2 as well as xen 3.2.1.
>> 
>> One can run a job on the vm to generate a packet every second or two to 
>> resolve this; ping in a loop should do it.
>
> Quick follow-up:
>
> arping -b -A *ip-address-of-vm*
>
> seems to work pretty well. I get a 2-3 second dropout if I'm running this 
> during live migration.

Odd, I am seeing 15 - 20 sec, how on earth does someone get 165ms that is 
talked about? I get the same delay if I am migrating over gig e or 20 gig 
infiniband. It also should not be my I/O subsystem as I am getting 400 MB 
writes and 550 MB reads.

><>
Nathan Stratton                                CTO, BlinkMind, Inc.
nathan at robotics.net                         nathan at blinkmind.com
http://www.robotics.net                        http://www.blinkmind.com
Permalink | Reply | 1 comment |
headers
jim burns | 1 Aug 02:14

Re: 3.1.x -> 3.2.1 and now Windows 2008 won't install

On Thu July 31 2008 6:11:17 am James Harper wrote:
> # xm list w2k8test32
> Name                                        ID   Mem VCPUs      State
> Time(s)
> w2k8test32                                 892   512     1     ----c-
> 112.3

Domain # 892?! I can't get much above 30-40 on F8 before I start getting the 
infamous 'hotplug scripts not working: vifx.y'. Rebooting dom0 restores proper 
functionality in my case.
Permalink | Reply | 1 comment |
headers
Joti Mail | 1 Aug 02:19
Picon

Exporting USB /USB-PCI-Device to DomU

Hello xen-users!

this evening I've been trying to get a Sinus 45 USB Dect Base-station to
run in a DomU to be usable with VBox and stuff.

The Sinus loads in the Dom0 fine with the gigaset and bas_gigaset
modules included in the kernel.

[   21.745129] gigaset: Hansjoerg Lipp <hjlipp <at> web.de>, Tilman Schmidt
<tilman <at> imap.cc>, Stefan Eilers
[   21.745274] gigaset: Driver for Gigaset 307x
[   21.797616] usb 2-1: gigaset_probe: Device matched (Vendor: 0x681,
Product: 0x1)
[   21.797802] usbcore: registered new interface driver bas_gigaset
[   21.797898] bas_gigaset: Tilman Schmidt <tilman <at> imap.cc>, Hansjoerg
Lipp <hjlipp <at> web.de>, Stefan Eilers
[   21.798014] bas_gigaset: USB Driver for Gigaset 307x

I've retrieved the PCI/Slot-Address of the USB-Hub the Sinus is
connected with and have added it to GRUB's menu.lst with pciback.hide

pciback.hide=(03:07.0) (00:12.1)
(The 03:07.0 is for a different device which works fine in a DomU)

00:12.1 USB Controller: ATI Technologies Inc Unknown device 4398
(prog-if 10 [OHCI])
         Subsystem: Elitegroup Computer Systems Unknown device 1b61
         Flags: bus master, 66MHz, medium devsel, latency 64, IRQ 16
         Memory at fe7fd000 (32-bit, non-prefetchable) [size=4K]
(Continue reading)

Permalink | Reply | 0 comments |
headers
jim burns | 1 Aug 02:39

Re: Re: [Xen-devel] State of Xen in upstream Linux

On Thu July 31 2008 1:54:38 pm Grant McWilliams wrote:
> > Fedora 9's kernel-xen package has been based on the mainline kernel from
> >
>  > the outset, but it is still packaged as a separate kernel.  kernel-xen
>  > has been dropped from rawhide (what will become Fedora 10), and all Xen
>  > support - both 32 and 64 bit - has been rolled into the main kernel
>  > package.
>
> Does this mean in the future all Fedora kernels will be Xen kernels? Is
> this wise? If I try to run VirtualBox on a Xen kernel the machine will
> reboot. If the Vbox module is loaded at runtime it will reboot forever.
> Yes, I know it's a Vbox issue but what about KVM. Can we run KVM on a Xen
> kernel?
>
> Or am I reading this completely wrong?

I'd imagine this is a question for the kernel dev guys. Linus seemed to think 
it wouldn't be a conflict, especially since xen still has to be enabled in the 
kernel config. Various distros may still provide two different kernels if they 
deem it necessary, regardless of what Fedora does.

(In fact, this is what F8 & F9 do now - the xen kernel doesn't just have all 
the xen features besides the subset offered by pvops: the non-xen kernel 
doesn't even have xen enabled. In my tests with Fedora 2.6.25, you have to 
enable xen, set xen_netfront and xen_blkfront to modules from builtin, and try 
to boot off the vmlinux file (which has elf extensions) instead of the vmlinuz 
file, and it still won't boot - around the time it loads xen_blkfront, just 
before mounting root.)
Permalink | Reply | 0 comments |
headers
Joti Mail | 1 Aug 02:57
Picon

[Fwd: Exporting USB /USB-PCI-Device to DomU]

Hello,

some news to it:
I tried to manually unbind the driver (ohci_hcd) from the usb hub's slot 
and then tried to rebind it with Xen. That worked. After a restart of 
the DomU the USB-Hub and the Gigaset were recognized.

In that case i'ld rather regard the problem as solved.

Regards,
Johannes

-------- Original-Nachricht --------
Betreff: Exporting USB /USB-PCI-Device to DomU
Datum: Fri, 01 Aug 2008 02:19:14 +0200
Von: Joti M.
An: xen-users <at> lists.xensource.com

Hello xen-users!

this evening I've been trying to get a Sinus 45 USB Dect Base-station to
run in a DomU to be usable with VBox and stuff.

The Sinus loads in the Dom0 fine with the gigaset and bas_gigaset
modules included in the kernel.

[   21.745129] gigaset: Hansjoerg Lipp <hjlipp <at> web.de>, Tilman Schmidt
<tilman <at> imap.cc>, Stefan Eilers
[   21.745274] gigaset: Driver for Gigaset 307x
[   21.797616] usb 2-1: gigaset_probe: Device matched (Vendor: 0x681,
(Continue reading)

Permalink | Reply | 0 comments |
headers
James Harper | 1 Aug 03:15
Picon

RE: 3.1.x -> 3.2.1 and now Windows 2008 won't install

> On Thu July 31 2008 6:11:17 am James Harper wrote:
> > # xm list w2k8test32
> > Name                                        ID   Mem VCPUs      State
> > Time(s)
> > w2k8test32                                 892   512     1     ----c-
> > 112.3
> 
> Domain # 892?! I can't get much above 30-40 on F8 before I start getting
> the infamous 'hotplug scripts not working: vifx.y'. Rebooting dom0 
> restores proper functionality in my case.

Name                              ID   Mem VCPUs      State   Time(s)
Domain-0                           0  1001     1     r-----  64236.8
sfg-dc                           883  1024     1     -b----    844.5
sfg-exch                         911  1024     1     r-----   6683.6
sfg-sql                          885  1024     1     ------    514.8
smtp2                              2   512     1     -b----   4984.0
w2k3test32                       905   512     1     -b----    534.4
w2k3test64                       910   512     1     -b----    117.0

all of the domains are working fine, it's only the w2k8test32 one that keeps crashing during install. I was
up to well over 3000 the other day before I rebooted, and it was only then that I started having problems. It
seems that every time a domain ends it leaves a certain about of cruft in the xenstore database, and
eventually there is enough cruft in there that 'xenstore -ls' causes a kernel bug, and 'xm create' takes
minutes rather than milliseconds to do anything.

The reason my numbers get that high is that I've been running 'xm save', and 'xm restore' in a loop testing the
gplpv drivers, and it routinely gets up to over 500 before it crashes. I think I've fixed all the crashes now though.

James
(Continue reading)

Permalink | Reply | 0 comments |
headers
Ramirez, Laura L | 1 Aug 03:58
Picon
Favicon

Migrating an HVM domain to localhost


Hi

I have tried to migrate a Linux HVM domain using both xen-3.2 and xen-unstable and the migrate just hangs
waiting for the domain to suspend.

I saw a previous posting on this list with the same behavior a while back, but I didn't see a reply.

Just wondering if there is some setup for an HVM domain that I am missing?

Thanks,

laura
Permalink | Reply | 0 comments |
headers
Admin Admin | 1 Aug 05:59
Picon
Favicon

How to 'mkinitrd' of CentOS (domU) on Ubuntu (dom0)?

Hello,

What (where) is the best way to get the xen enabled
CentOS 5.2 vmlinuz and initrd images for use as a
guest domain (domU) on Ubuntu host domain (dom0).

These are for use with post install of domU, which is
setup using xen-create-image.

Thanks
Permalink | Reply | 0 comments |

Gmane