On 2010/12/10 12:18, Brynet wrote:
quoted text
> Mark Kettenis wrote:
> > Here's an attempt to fix a potential MCLGETI issue with vr(4) similar
> > to what I recently fixed fro re(4).  Unfortunately I don't have any
> > vr(4) hardware myself, so I need some hep testing this.
> > 
> > Things to look for are:
> > 
> > 1. Does this diff have any effect on throughput (packets/s, bits/s).
> > 
> > 2. Does this fix any problems with the interface if you blast it with
> >    packets?
> > 
> > 3. Does livelock mitigation actualy work?
> > 
> > Thanks,
> > 
> > Mark
> 
> Hi Mark,
> 
> Did you receive any feedback for this? is there an easy way to replicate
> the problems others are seeing, as my systems seem stable.

i was hoping to have tested this already but i just had to burn the
couple of hours i had "spare" on getting some ipsec boxes to start
talking to each other again. maybe next week.

to replicate, on a fast nearby machine:

pkg_add netrate
netblast ip.addr.of.vr 1234 5 10

i typically saw hangs in a fraction of a second on a vr/geode system.
to recover connect in on another nic or the console and ifconfig vrX down
then ifconfig vrX up.