100Mbit network performance - again

[Andrew P.]

On 7/27/05, Michael C. Shultz <ringworm01 at gmail.com> wrote:
> On Tuesday 26 July 2005 16:48, Andrew P. wrote:
> > On 7/27/05, Michael C. Shultz <ringworm01 at gmail.com> wrote:
> > > On Tuesday 26 July 2005 16:00, Andrew P. wrote:
> > > > Hello all!
> > > >
> > > > I remember being able to reach 11-12Mbytes/s between two Win95
> > > > workstations with NE2000 $10 NIC's installed, connected via BNC cable.
> > > > I am now able to reach 11-12Mbytes/s between all kinds of Windows
> > > > 2000/XP machines with all kinds of cheapest 100Mbit ethernet hardware.
> > > >
> > > > But I have never ever exceeded 8-9Mbytes/s between a Windows machine
> > > > and a FreeBSD box - _never_. Be it Samba, different ftp/http servers,
> > > > different FreeBSD versions (4.x/5.x), with ipfw enabled or disabled,
> > > > etc., - the speed always hovers around 7-8Mb/s. I know it's not
> > > > critical, I know I should've upgraded to Gigabit hardware long ago,
> > > > but is there something wrong?
> > > >
> > > > I tried different linux distros, but they all seem to be even slower.
> > > > Wazzup?..
> > > >
> > > > Thanks,
> > > > Andrew P.
> > >
> > > Here is the "ifconfig" output from a machine that has one nic set at
> > > 10Mbit/half duplex and one at 100Mbit full duplex. how does it compare
> > > with your system?
> > >
> > > xl0: flags=8843<UP,BROADCAST,RUNNING,SIMPLEX,MULTICAST> mtu 1500
> > >         options=1<RXCSUM>
> > >         inet6 fe80::210:4bff:fe70:4fb0%xl0 prefixlen 64 scopeid 0x1
> > >         inet 71.102.0.97 netmask 0xffffff00 broadcast 71.102.0.255
> > >         ether 00:10:4b:70:4f:b0
> > >         media: Ethernet autoselect (10baseT/UTP)
> > >         status: active
> > > xl1: flags=8843<UP,BROADCAST,RUNNING,SIMPLEX,MULTICAST> mtu 1500
> > >         options=1<RXCSUM>
> > >         inet 192.168.1.1 netmask 0xffffff00 broadcast 192.168.1.255
> > >         inet6 fe80::210:4bff:fe0a:7cbc%xl1 prefixlen 64 scopeid 0x2
> > >         ether 00:10:4b:0a:7c:bc
> > >         media: Ethernet 100baseTX <full-duplex>
> > >         status: active
> >
> > Well, if that really matters to you:
> > (freebsd 5.4)
> > vr0: flags=8843<UP,BROADCAST,RUNNING,SIMPLEX,MULTICAST> mtu 1500
> >         inet6 fe80::20f:3dff:feca:c494%vr0 prefixlen 64 scopeid 0x1
> >         inet 192.168.17.217 netmask 0xffffff00 broadcast 192.168.17.255
> >         ether 00:0f:3d:ca:c4:94
> >         media: Ethernet autoselect (100baseTX <full-duplex>)
> >         status: active
> > rl0: flags=8843<UP,BROADCAST,RUNNING,SIMPLEX,MULTICAST> mtu 1500
> >         options=8<VLAN_MTU>
> >         inet 192.168.17.1 netmask 0xffffff00 broadcast 192.168.17.255
> >         ether 00:40:f4:8d:a7:f8
> >         media: Ethernet autoselect (100baseTX <full-duplex>)
> >         status: active
> 
> Same netmask for two different segments of the same class C network? How's it
> work with one segment disconnected?
> 
> -Mike
> 
> 
> > rl1: flags=8843<UP,BROADCAST,RUNNING,SIMPLEX,MULTICAST> mtu 1500
> >         options=8<VLAN_MTU>
> >         ether 00:40:f4:8d:9c:af
> >         media: Ethernet autoselect (100baseTX <full-duplex>)
> >         status: active
> > (fedora core 4)
> > eth0      Link encap:Ethernet  HWaddr 00:E0:81:2F:04:3E
> >           inet addr:193.233.5.13  Bcast:193.233.5.63  Mask:255.255.255.192
> >           inet6 addr: fe80::2e0:81ff:fe2f:43e/64 Scope:Link
> >           UP BROADCAST RUNNING MULTICAST  MTU:1500  Metric:1
> >           RX packets:123946466 errors:0 dropped:0 overruns:0 frame:0
> >           TX packets:176380358 errors:0 dropped:0 overruns:0 carrier:0
> >           collisions:0 txqueuelen:1000
> >           RX bytes:42267471987 (39.3 GiB)  TX bytes:197116022761 (183.5
> > GiB) Interrupt:177
> >
> > Andrew P.
> 

Actually vr0 and rl0 are on different boxes :)

Andrew P.