On Wed, 20 Nov 2002 10:44:26 -0500 Mark Weaver <[EMAIL PROTECTED]> wrote:
> Pierre Fortin wrote: > > On Wed, 20 Nov 2002 08:04:34 -0500 (EST) daRcmaTTeR > > <[EMAIL PROTECTED]> wrote: > > > > > >> On Tue, 19 Nov 2002, Pierre Fortin wrote: > >> > >> > >>> On Tue, 19 Nov 2002 13:29:43 -0500 Mark Weaver > >>> <[EMAIL PROTECTED]> wrote: > >>> > >>> > >>>> Hi List, > >>>> > >>>> has anyone else noticed the network connectivity in Mandrake > >>>> 9.0 goes dead periodically? It's very annoying and it seems to > >>>> be linked to Shorewall firewall system. > >>>> > >>>> Mark > >>> > >>> Any chance you're experiencing the ssh "pause"... I use ssh > >>> extensively and experience pauses ONLY when modem connected > >>> (never when LAN connected) to the server. It doesn't affect all > >>> sessions; I can continue to work on another session while one or > >>> more are stalled... I rarely connect between 2 modem connected > >>> machines; but it's on my list to watch for that possible > >>> scenario... > >>> > >>> Pierre > >> > >> Hi Pierre, > >> > >> No. This is happening at work on the LAN. I don't use 9 on my > >> server at home. "it" and my old machine don't get along at all very > >> well. the network troubles with that machine running 9.0 are > >> terrible. The interruptions are across the spectrum from web server > >> to ssh. > >> > > > > > > In that case, some more details might help... during the outage of > > course :^) The usual: ifconfig, route -n, ... > > > > Is the problem on a "schedule" (at specific times or rate)? > > > > Anything else happening at the time? Other symptoms/anomolies? Other > > users affected? > > > > Cheers, Pierre > > nothing discernable, which makes it rather a pain to track. One thing > i've noticed is that if I restart the network service, OR just restart > the shorewall service the problem clears up. the shorewall service > restart may, or may not be related. In the short time I was able to get > 9.0 running on my home server I noticed the same behavior for a time. > > As far as I've noticed there doesn't appear to be any regular interval. > the events seem to be random. Other then losing connectivity for a time, > > and it does clear (unclog) after a time, things return to normal. > > here's the information from the ifconfig and route -n output: > > eth0 Link encap:Ethernet HWaddr 00:60:97:B9:59:E3 > inet addr:205.216.60.162 Bcast:205.216.60.255 > Mask:255.255.255.0 > UP BROADCAST RUNNING MULTICAST MTU:1500 Metric:1 > RX packets:4836553 errors:0 dropped:0 overruns:0 frame:0 > TX packets:417172 errors:0 dropped:0 overruns:0 carrier:0 > collisions:0 txqueuelen:100 > RX bytes:488489887 (465.8 Mb) TX bytes:357181688 (340.6 Mb) > Interrupt:14 Base address:0xec80 > > eth1 Link encap:Ethernet HWaddr 00:60:08:2E:15:CF > inet addr:205.216.60.39 Bcast:205.216.60.255 > Mask:255.255.255.0 UP BROADCAST RUNNING MULTICAST MTU:1500 > Metric:1 RX packets:4953429 errors:0 dropped:0 overruns:0 > frame:0 TX packets:257085 errors:0 dropped:0 overruns:0 > carrier:0 collisions:0 txqueuelen:100 > RX bytes:509489013 (485.8 Mb) TX bytes:193906058 (184.9 Mb) > Interrupt:9 Base address:0xe8c0 > > lo Link encap:Local Loopback > inet addr:127.0.0.1 Mask:255.0.0.0 > UP LOOPBACK RUNNING MTU:16436 Metric:1 > RX packets:99456 errors:0 dropped:0 overruns:0 frame:0 > TX packets:99456 errors:0 dropped:0 overruns:0 carrier:0 > collisions:0 txqueuelen:0 > RX bytes:15233458 (14.5 Mb) TX bytes:15233458 (14.5 Mb) > > [root@download root]# route -n > Kernel IP routing table > Destination Gateway Genmask Flags Metric Ref Use > Iface > 205.216.60.0 0.0.0.0 255.255.255.0 U 0 0 0 > eth1 205.216.60.0 0.0.0.0 255.255.255.0 U 0 0 > 0 eth1 > 127.0.0.0 0.0.0.0 255.0.0.0 U 0 0 0 > lo 0.0.0.0 205.216.60.1 0.0.0.0 UG 0 0 > 0 eth0 OUCH!!! TWO interfaces on the same subnet! Are you really sure what your network is doing? Just a peek at the route table shows two subnets (same) on eth1 while the default route is out eth0... Since you're apparently not routing between eth0 & eth1 (bridging?), try shutting down one interface... my guess is that your problems will not occur unless you bring it back online... If you have ethereal, try sniffing all interfaces... the RX packets/bytes on the interfaces is almost identical; but eth0 seems to carry more TX traffic (about 2:1)... sounds like some subtle pathing problems may be biting you... HTH, Pierre
Want to buy your Pack or Services from MandrakeSoft? Go to http://www.mandrakestore.com
