Hello,
recently I've installed 7.0 and now I'm observing strange thing with my
Internet connection.
Access to some sites may be VERY slow or doesn't work at all with different
kinds of timeout
messages or without messages at all (but other sites works fine). For example,
firefox may say
"Transferring data from..." message and then "The connection to
the server was reset while
the page was loading", lynx just says "HTTP/1.1 200 OK" and
displays no data, etc.
I'm also using Linux and Windows XP on the same machine, and the same sites
are loaded very
quickly. Note I'm not using a proxy.
My NIC is:
mskc0: <Marvell Yukon 88E8053 Gigabit Ethernet> port 0xc800-0xc8ff mem
0xcdefc000-0xcdefffff irq 17 at device 0.0 on pci2
msk0: <Marvell Technology Group Ltd. Yukon EC Id 0xb6 Rev 0x01> on mskc0
msk0: Ethernet address: 00:11:2f:37:d5:19
miibus0: <MII bus> on msk0
e1000phy0: <Marvell 88E1111 Gigabit PHY> PHY 0 on miibus0
e1000phy0: 10baseT, 10baseT-FDX, 100baseTX, 100baseTX-FDX, 1000baseTX-FDX, auto
mskc0: [FILTER]
And network configuration is:
msk0: flags=8843<UP,BROADCAST,RUNNING,SIMPLEX,MULTICAST> metric 0 mtu 1500
options=19a<TXCSUM,VLAN_MTU,VLAN_HWTAGGING,VLAN_HWCSUM,TSO4>
ether 00:11:2f:37:d5:19
inet 192.168.1.2 netmask 0xffffff00 broadcast 192.168.1.255
media: Ethernet autoselect (100baseTX <full-duplex>)
status: active
lo0: flags=8049<UP,LOOPBACK,RUNNING,MULTICAST> metric 0 mtu 16384
inet 127.0.0.1 netmask 0xff000000
msk0 is DHCP-configured from Linux router, which is connected to ISP via PPPoE:
eth0 Link encap:Ethernet HWaddr 00:01:C0:02:F5:45
UP BROADCAST RUNNING MULTICAST MTU:1500 Metric:1
...
eth1 Link encap:Ethernet HWaddr 00:01:C0:03:0A:66
inet addr:192.168.1.1 Bcast:192.168.1.255 Mask:255.255.255.0
UP BROADCAST RUNNING MULTICAST MTU:1500 Metric:1
...
ppp0 Link encap:Point-to-Point Protocol
inet addr:X.X.X.X P-t-P:10.0.0.1 Mask:255.255.255.255
UP POINTOPOINT RUNNING NOARP MULTICAST MTU:1492 Metric:1
...
Here eth0 is a "transport layer" for ppp0, which is connected to ISP,
and eth1 is a local
network interface. The Linux router does a full NAT for all machines from
192.168.1.0.
I've tried to reduce MTU to match router's ppp0 setting with
"ifconfig msk0 mtu 1492" and even
down it to 1480, but it doesn't help.
Any ideas ?
Dmitry
Dmitry Antipov wrote:> Hello, > > recently I've installed 7.0 and now I'm observing strange thing with my > Internet connection.Meaning you had earlier versions installed that worked? If so, did configuration change?> I've tried to reduce MTU to match router's ppp0 setting with "ifconfig > msk0 mtu 1492" and even > down it to 1480, but it doesn't help.I cannot help you directly but here's some ideas: - Try ridiculously low MTUs - Do you have a firewall somewhere in the network? Maybe it's improperly configured. I remember seeing similar weird behaviour when I accidentally disabled all ICMP traffic. -------------- next part -------------- A non-text attachment was scrubbed... Name: signature.asc Type: application/pgp-signature Size: 189 bytes Desc: OpenPGP digital signature Url : http://lists.freebsd.org/pipermail/freebsd-stable/attachments/20080229/9e9ddc7f/signature.pgp
Jeremy Chadwick
2008-Feb-29 15:47 UTC
7.0 - slow/unstable Internet access via Linux router
On Fri, Feb 29, 2008 at 02:06:38PM +0300, Dmitry Antipov wrote:> recently I've installed 7.0 and now I'm observing strange thing with my > Internet connection. > Access to some sites may be VERY slow or doesn't work at all with different > kinds of timeout > messages or without messages at all (but other sites works fine). For > example, firefox may say > "Transferring data from..." message and then "The connection to the server > was reset while > the page was loading", lynx just says "HTTP/1.1 200 OK" and displays no > data, etc.Sounds like TCP stack breakage, and not so much an MTU problem. I read many months ago that some others having this problem solved it by disabling RFC1323 extensions (default is on), which is a little odd, but it worked for a couple people. Try doing "sysctl net.inet.tcp.rfc1323=0" and see if the problem goes away. If it does, put tcp_extensions="no" in /etc/rc.conf. I can't reproduce this behaviour, though, on my own setup at home (using em(4) gigE NICs on the BSD box, and Broadcom BCM47xx 10/100 NICs on the Linux box (a WRT54GL). -- | Jeremy Chadwick jdc at parodius.com | | Parodius Networking http://www.parodius.com/ | | UNIX Systems Administrator Mountain View, CA, USA | | Making life hard for others since 1977. PGP: 4BD6C0CB |
On Fri, 29 Feb 2008, Dmitry Antipov wrote:> Hello, > > recently I've installed 7.0 and now I'm observing strange thing with > my Internet connection. Access to some sites may be VERY slow or > doesn't work at all with different kinds of timeout messages or > without messages at all (but other sites works fine). For example, > firefox may say "Transferring data from..." message and then "The > connection to the server was reset while the page was loading", lynx > just says "HTTP/1.1 200 OK" and displays no data, etc. > > I'm also using Linux and Windows XP on the same machine, and the same > sites are loaded very quickly. Note I'm not using a proxy. > > My NIC is: > > mskc0: <Marvell Yukon 88E8053 Gigabit Ethernet> port 0xc800-0xc8ff mem > 0xcdefc000-0xcdefffff irq 17 at device 0.0 on pci2 > msk0: <Marvell Technology Group Ltd. Yukon EC Id 0xb6 Rev 0x01> on mskc0 > msk0: Ethernet address: 00:11:2f:37:d5:19 > miibus0: <MII bus> on msk0 > e1000phy0: <Marvell 88E1111 Gigabit PHY> PHY 0 on miibus0 > e1000phy0: 10baseT, 10baseT-FDX, 100baseTX, 100baseTX-FDX, 1000baseTX-FDX, > auto > mskc0: [FILTER]*snip*> Any ideas ?I saw this commit[1] to HEAD this morning concerning msk(4) and 88E8053. It may not be related. Sean 1. http://lists.freebsd.org/pipermail/cvs-src/2008-February/087879.html -- scf@FreeBSD.org
Jeremy Chadwick wrote:> Sounds like TCP stack breakage, and not so much an MTU problem. > > I read many months ago that some others having this problem solved it by > disabling RFC1323 extensions (default is on), which is a little odd, but > it worked for a couple people. Try doing "sysctl > net.inet.tcp.rfc1323=0" and see if the problem goes away. If it does, > put tcp_extensions="no" in /etc/rc.conf. > > I can't reproduce this behaviour, though, on my own setup at home (using > em(4) gigE NICs on the BSD box, and Broadcom BCM47xx 10/100 NICs on the > Linux box (a WRT54GL).Is it required to have 'options INET6' even if I'm not using any IPv6 connectivity ? Also I have occasional 'mskc0: Uncorrectable PCI Express error' messages, which is a known (http://www.freebsd.org/cgi/query-pr.cgi?pr=kern/119613) problem... Thanks, Dmitry