[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: [Xen-users] unstable network problem

  • To: xen-users@xxxxxxxxxxxxxxxxxxx
  • From: Mogens Valentin <monz@xxxxxxxxx>
  • Date: Mon, 08 Aug 2005 23:53:23 +0200
  • Delivery-date: Mon, 08 Aug 2005 21:41:08 +0000
  • List-id: Xen user discussion <xen-users.lists.xensource.com>

Ernst Bachmann wrote:
On Monday 08 August 2005 15:34, Tim Durack wrote:

I must be doing something wrong:

Have set up a basic Xen system, Dom0 + one Xen0 guest. Configured for
a routed network.

Ping Dom0 -> guest works.
Ping guest -> Dom0 works.

nmap Dom0 -> guest reports port 22 open.

Ssh Dom0 -> guest hangs.
ssh guest -> Dom0 hangs. Same behaviour. Watching a packet capture
shows traffic passing but with retransmits and poor performance.

If I build a gre tunnel between Dom0 and guest, everything works fine,
with the exception that I must generate some traffic from the guest

I have no iptables rules configured, so it's not that. Looks like
there is maybe some issue with tcp traffic between domains.

Looks like a bug to me, but I'm posting to the user list first.

There was a long thread about this problem maybe one or two weeks ago...
I think the problem was that the virtual eth devices had packet queueing disabled (txqueuelen:0 on ifconfig), and had a 4k size max. ring buffer for transfering packets between domains.
Whenever that buffer would run full, packets are dropped...

A too-small network buffer should never be able to hang a system; then again, it's unclear what Tim means by 'hang'.

Kind regards,
Mogens Valentin

Xen-users mailing list



Lists.xenproject.org is hosted with RackSpace, monitoring our
servers 24x7x365 and backed by RackSpace's Fanatical Support®.