Date: Fri, 20 Jun 2014 11:54:55 +0200 From: =?ISO-8859-1?Q?Roger_Pau_Monn=E9?= <roger.pau@citrix.com> To: Karl Pielorz <kpielorz_lst@tdx.co.uk>, <freebsd-xen@freebsd.org> Subject: Re: re. pr188261 - any way I can investigate further? Message-ID: <53A404EF.4070400@citrix.com> In-Reply-To: <7B689F2C8C07F01AD552CA61@Mail-PC.tdx.co.uk> References: <7B689F2C8C07F01AD552CA61@Mail-PC.tdx.co.uk>
next in thread | previous in thread | raw e-mail | index | archive | help
On 18/06/14 15:11, Karl Pielorz wrote: > > Hi, > > I submitted a PR a while ago (pr188261) - or now, Bugzilla > <https://bugs.freebsd.org/bugzilla/show_bug.cgi?id=188261>; > > I've updated this recently - as I found doing 'ifconfig xn0 -txcsum' on > a "client" will fix the issue, for just that client - but I'm still > unable to fix the 'router' VM in such a way as all clients don't need a > fix. > > I have a test system setup with all this now (and all of 9.2-S, 10-S, > and 11-Current have the issue) - is anyone able to point me in the right > direction of what I can do to investigate this further? (i.e. debug > things, packet captures, etc. etc.) > > We could really do with getting this fixed - we use FreeBSD as routers > on a few systems - as as they move to Xen we end up having to either > setup 2 pools (and be very, very careful about what machines we put on > each) in order to avoid this - or face running them in HVM mode for > which the performance is very, very poor. I suggest that you try to bring this to freebsd-net, I certainly don't have the network skills to know what's going on, maybe someone can provide more insight there which could help track this down (I can certainly help as much as possible with the Xen part). If not I will try to get a look, but I cannot tell when is that going to happen, I've got my plate quite full right now. Roger.
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?53A404EF.4070400>