On Wed, Nov 10, 2010 at 01:04:28PM +0200, Eugene Perevyazko wrote:
>
> I have a router running RELENG_7 with two dual-port igbs - igb0 and igb1 are
> on
> 82575 on intel s5520ur mb and igb2 and igb3 are on 82576 on ET dual-port card.
> 82576 is in 8x slot.
> Main traffic flows
On Thu, Nov 11, 2010 at 12:49:52PM +0200, Eugene Perevyazko wrote:
> On Thu, Nov 11, 2010 at 01:47:02AM +0100, Ivan Voras wrote:
> > On 11/10/10 12:04, Eugene Perevyazko wrote:
> >
> > >Tried 2 queues and 1 queue per iface, neither hitting cpu limit.
> >
> >
On Thu, Nov 11, 2010 at 01:47:02AM +0100, Ivan Voras wrote:
> On 11/10/10 12:04, Eugene Perevyazko wrote:
>
> >CPU is e5...@2.4ghz, 8 cores, irqs bound to different cores skipping HT
> >ones.
>
> Unless you need the CPU cores for other tasks on the server, they won
= network
subclass = ethernet
cap 01[40] = powerspec 3 supports D0 D3 current D0
cap 05[50] = MSI supports 1 message, 64 bit, vector masks
cap 11[70] = MSI-X supports 10 messages in map 0x1c enabled
cap 10[a0] = PCI-Express 2 endpoint max data 256(512) link x4(x4)
--
Eugene
stick.
I've even made a crontab entry to replace corrupted file.
So the problem seems to be probabilistic, depending on amount of transferred
data.
--
Eugene Perevyazko
___
freebsd-net@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/
eue eats 100% of one core while other 3 ng_queues sit idle.
I'm going to set queueing on all ng_nat's hooks to see if it will help.
--
Eugene Perevyazko
___
freebsd-net@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-
"? What kind of DoS is it about?
--
Eugene Perevyazko
___
freebsd-net@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-net
To unsubscribe, send any mail to "freebsd-net-unsubscr...@freebsd.org"
e lack of virtual/physical memory - the host has some "free" and
lots of "Inact" memory and uses no swap space yet.
Can it be related to those sysctls?
vm.kmem_size_scale: 3
vm.kmem_size_max: 335544320
vm.kmem_size_min: 0
vm.kmem_si
arly correlates with network load for the host,
which is mainly doing dummynet and ng_nat.
How can I find what to tune in this case?
--
Eugene Perevyazko
___
freebsd-net@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-net
To u
ire in 20 seconds.
If I set arp with "arp -S host etheraddr" - it will be really permanent.
It looks like with "-s" the flag of incompleteness isn't cleared.
Is this a bug or a feature?
--
Eugene Perevyazko
___
freebs
root access to the host with this NIC if you have time/wish.
--
Eugene Perevyazko
___
freebsd-net@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-net
To unsubscribe, send any mail to "freebsd-net-unsubscr...@freebsd.org"
On Wed, Aug 26, 2009 at 12:48:56PM +0300, Eugene Perevyazko wrote:
> On Wed, Aug 26, 2009 at 12:39:16PM +0300, Eugene Perevyazko wrote:
> > On Tue, Aug 25, 2009 at 11:25:53AM -0700, Pyun YongHyeon wrote:
> > >
> > > Try attached patch and let me know how it goes on
On Wed, Aug 26, 2009 at 12:39:16PM +0300, Eugene Perevyazko wrote:
> On Tue, Aug 25, 2009 at 11:25:53AM -0700, Pyun YongHyeon wrote:
> >
> > Try attached patch and let me know how it goes on your box.
> > You can see statistics counters maintained in driver with sysctl.
d media:
media autoselect
media 1000baseSX mediaopt full-duplex
media 1000baseSX
media none
--
Eugene Perevyazko
___
freebsd-net@freebsd.org mailing list
http://lists.freebsd.org/m
Hello.
On Tue, Aug 25, 2009 at 02:46:49PM +0300, Eugene Perevyazko wrote:
> > >
> > > ATM there is no easy/clean way to pass driver specific data to mii
> > > layer in FreeBSD so e1000phy(4) incorrectly thinks it found copper
> > > PHY. Marvell PHYs
Hello Pyun.
On Tue, Aug 25, 2009 at 11:38:57AM +0300, Eugene Perevyazko wrote:
> [...]
> >
> > ATM there is no easy/clean way to pass driver specific data to mii
> > layer in FreeBSD so e1000phy(4) incorrectly thinks it found copper
> > PHY. Marvell PHYs seem to h
Hello Pyun.
I apologise for long delay, we had 3-day state holidays here. :)
On Fri, Aug 21, 2009 at 03:19:32PM -0700, Pyun YongHyeon wrote:
> On Fri, Aug 21, 2009 at 05:20:39PM +0300, Eugene Perevyazko wrote:
[...]
>
> > Then the driver recognizes the card:
> >
> > ms
media 10baseT/UTP
media none
The adapter is definitely not broken, I've tested it on Win and there it works
normally. Also on Win it has no speed/duplex settings, while msk driver allows
to change media.
I've also tried Marvell's binary if_myk dr
/handbook/ipsec.html which
> is pretty darn good)?
I use IPSec tunnels without gif interface on freebsd, don't know if it will
work with it. I declare policy in /etc/ipsec.conf, and use racoon
(ports/security/ipsec-tools) to do all the rest. It's pretty simple
rminate dead connections.
And why don't you try an mpd (/usr/ports/net/mpd) that is perfectly suited
to manage hundreds and thousands of simultaneous ppp (and PPPoE in particular)
connections? It's fast, flexible and easy to setup both as ser
On Fri, Jan 16, 2009 at 01:21:15PM -0800, Chuck Swiger wrote:
> On Jan 16, 2009, at 3:50 AM, Eugene Perevyazko wrote:
> >On Fri, Jan 16, 2009 at 12:20:21PM +0300, Alexey Ivanov wrote:
> >>Is there any command identical to:
> >> iptables -A INPUT -p tcp -m tcp -dpo
be a week or two later,
when I will have some free time.
Eugene Perevyazko
___
freebsd-net@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-net
To unsubscribe, send any mail to "freebsd-net-unsubscr...@freebsd.org"
look into this problem in a bit more
> detail.
Thanks for your attention, Christian, but please consider that it's possible
to mix tagged and untagged frames on interface and such DLT_ will put an
artificial limit on this while technically there's no such li
On Thu, May 15, 2008 at 12:21:23PM -0400, Sam Leffler wrote:
> Eugene Perevyazko wrote:
> > Is there a way to inject full-mtu vlan-tagged frame through BPF on parent
> >device?
> > I'd like not to open several tenths of BPFs for each vlanN if it's
> > possib
t;ifconfig fxp0 -vlanmtu" also makes no difference.
Is there a way to inject full-mtu vlan-tagged frame through BPF on parent
device?
I'd like not to open several tenths of BPFs for each vlanN if it's possible.
I've tried on bge also with same results, so that is
25 matches
Mail list logo