[Users] Abysmal network performance inside hypervisor node

Sander Grendelman sander at grendelman.com
Wed Dec 11 13:51:04 UTC 2013


Then I'm also out of ideas.

You could take al look at the options in /etc/vdsm/vdsm.conf
( information at /usr/share/doc/vdsm-*/vdsm.conf.sample )

On Wed, Dec 11, 2013 at 2:48 PM, Markus Stockhausen
<stockhausen at collogia.de> wrote:
>> Von: sander.grendelman at gmail.com [sander.grendelman at gmail.com]&quot; im Auftrag von &quot;
>> Gesendet: Mittwoch, 11. Dezember 2013 14:43
>> An: Markus Stockhausen
>> Cc: ovirt-users
>> Betreff: Re: [Users] Abysmal network performance inside hypervisor node
>>
>> I'm just wondering if the tested interfaces are bridged because I've
>> seen some issues with
>> network througput and bridged interfaces on my local system (F19).
>>
>> Basically, if an IP is configured on the bridge itself ( in oVirt this
>> is the case if the network
>> is configured as a VM network ) latency goes up and throughput goes down.
>>
>> Can you rule this one out by using an unbridged interface?
>
> I see. But my storage infiniband network is working without a bridge.
> That was the mounting network device of my initial mail. So with
> (ovirtmgmt) and without bridge (infiniband) I have the same problem.
>
> [root at colovn01 ~]# ifconfig ib1
> ib1: flags=4163<UP,BROADCAST,RUNNING,MULTICAST>  mtu 2044
>         inet 10.10.30.1  netmask 255.255.255.0  broadcast 10.10.30.255
>         inet6 fe80::223:7dff:ff94:d3fe  prefixlen 64  scopeid 0x20<link>
> Infiniband hardware address can be incorrect! Please read BUGS section in ifconfig(8).
>         infiniband 80:00:00:49:FE:80:00:00:00:00:00:00:00:00:00:00:00:00:00:00  txqueuelen 256  (InfiniBand)
>         RX packets 3575120  bytes 7222538528 (6.7 GiB)
>         RX errors 0  dropped 10  overruns 0  frame 0
>         TX packets 416942  bytes 29156648 (27.8 MiB)
>         TX errors 0  dropped 0 overruns 0  carrier 0  collisions 0
>
> Markus



More information about the Users mailing list