Le mardi 18 septembre 2007 19:33, Michael a écrit : > Hello everyone! > > I Installed Debian 4.0 (Etch) on my notebook Toshiba Satellite A105 S2091 a > couple of weeks ago and I've been facing this extremely annoying problem > since then: my internet connection falls, for absolutely any apparent > reason, and only returns after completely restarting the system. Nothing > else works! I've already tried the internet service from somewhere else, > restarting my modem, unplugging and replugging the wires, the GNOME > applications, the `ifdown eth0' and `ifup eth0' commands on a shell and > several combinations of these. > > After discussing this on the Brasilian users list, someone pointed out the > possibility of a problem with the driver for my card. The output of `lspci' > gives me, among several other lines, > > ` > 02:07.0 Ethernet controller: Realtek Semiconductor Co., Ltd. > RTL-8139/8139C/8139C+ (rev 10) > ' > > what, I guess, indicates that Debian can see my card. After Googling a > little bit I got to > http://www.linuxquestions.org/hcl/showproduct.php?product=482, where I > learned that this card uses module `8139too'. But it seems like it is > already loaded, as indicated the output output of `lsmod': > ` > ntbkmichael:/home/michael# lsmod > Module Size Used by > rfcomm 34584 0 > l2cap 21760 5 rfcomm > bluetooth 46020 4 rfcomm,l2cap > ppdev 8676 0 > parport_pc 32132 0 > lp 11012 0 > parport 33256 3 ppdev,parport_pc,lp > ipv6 226272 8 > button 6672 0 > ac 5188 0 > battery 9636 0 > dm_snapshot 15552 0 > dm_mirror 19152 0 > dm_mod 50232 2 dm_snapshot,dm_mirror > loop 15048 0 > joydev 9088 0 > pcmcia 34140 0 > firmware_class 9600 1 pcmcia > snd_hda_intel 17332 1 > snd_hda_codec 137856 1 snd_hda_intel > snd_pcm_oss 38368 0 > snd_mixer_oss 15200 1 snd_pcm_oss > rtc 12372 0 > snd_pcm 68676 3 snd_hda_intel,snd_hda_codec > - Masquer le texte des messages précédents - > ,snd_pcm_oss > snd_timer 20996 1 snd_pcm > snd 47012 8 > snd_hda_intel,snd_hda_codec,snd_pcm_oss,snd_mixer_oss,snd_pcm,snd_timer > soundcore 9248 1 snd > yenta_socket 24460 1 > snd_page_alloc 9640 2 snd_hda_intel,snd_pcm > psmouse 35016 0 > serio_raw 6660 0 > pcspkr 3072 0 > rsrc_nonstatic 11840 1 yenta_socket > pcmcia_core 36852 3 pcmcia,yenta_socket,rsrc_nonstatic > i2c_piix4 8140 0 > shpchp 33024 0 > pci_hotplug 28704 1 shpchp > i2c_core 19680 1 i2c_piix4 > ati_agp 8652 0 > agpgart 29896 1 ati_agp > tsdev 7520 0 > evdev 9088 1 > ext3 119240 1 > jbd 52456 1 ext3 > mbcache 8356 1 ext3 > ide_cd 36064 0 > cdrom 32544 1 ide_cd > 8139cp 21920 0 > usbhid 37248 0 > sd_mod 19040 3 > 8139too 25120 0 > mii 5344 2 8139cp,8139too > atiixp 6064 0 [permanent] > ehci_hcd 28136 0 > ohci_hcd 18276 0 > sata_sil 11464 2 > usbcore 112644 4 usbhid,ehci_hcd,ohci_hcd > libata 89396 1 sata_sil > scsi_mod 124168 2 sd_mod,libata > generic 4868 0 [permanent] > ide_core 110504 3 ide_cd,atiixp,generic > thermal 13608 0 > processor 28840 1 thermal > fan 4804 0 > ' > > Does anybody have a clue of what might be happening? The excerpt of the > output of `grep eth0 /var/log/kern.log' referring to one of the many > sessions in which I've eventually lost my connection was > > ` > Sep 16 15:57:16 ntbkmichael kernel: eth0: RealTek RTL8139 at 0xa000, > 00:a0:d1:38:4f:b6, IRQ 201 > Sep 16 15:57:16 ntbkmichael kernel: eth0: Identified 8139 chip type > 'RTL-8100B/8139D' > Sep 16 15:57:16 ntbkmichael kernel: eth0: link up, 100Mbps, full-duplex, > lpa 0x45E1 > Sep 16 16:09:27 ntbkmichael kernel: NETDEV WATCHDOG: eth0: transmit timed > out > Sep 16 16:09:30 ntbkmichael kernel: eth0: Transmit timeout, status 0c 0005 > c07f media 00. > Sep 16 16:09:30 ntbkmichael kernel: eth0: Tx queue start entry 6562 dirty > entry 6558. > Sep 16 16:09:30 ntbkmichael kernel: eth0: Tx descriptor 0 is 0008a042. > Sep 16 16:09:30 ntbkmichael kernel: eth0: Tx descriptor 1 is 0008a042. > Sep 16 16:09:30 ntbkmichael kernel: eth0: Tx descriptor 2 is 0008a042. > (queue head) > Sep 16 16:09:30 ntbkmichael kernel: eth0: Tx descriptor 3 is 0008a042. > Sep 16 16:09:30 ntbkmichael kernel: eth0: link up, 100Mbps, full-duplex, > lpa 0x45E1 > Sep 16 16:09:42 ntbkmichael kernel: NETDEV WATCHDOG: eth0: transmit timed > out > Sep 16 16:09:45 ntbkmichael kernel: eth0: Transmit timeout, status 0c 0005 > c07f media 00. > Sep 16 16:09:45 ntbkmichael kernel: eth0: Tx queue start entry 4 dirty > entry 0. > Sep 16 16:09:45 ntbkmichael kernel: eth0: Tx descriptor 0 is 0008a042. > (queue head) > Sep 16 16:09:45 ntbkmichael kernel: eth0: Tx descriptor 1 is 0008a04b. > Sep 16 16:09:45 ntbkmichael kernel: eth0: Tx descriptor 2 is 0008a04b. > Sep 16 16:09:45 ntbkmichael kernel: eth0: Tx descriptor 3 is 0008a042. > Sep 16 16:09:45 ntbkmichael kernel: eth0: link up, 100Mbps, full-duplex, > lpa 0x45E1 > Sep 16 16:09:57 ntbkmichael kernel: NETDEV WATCHDOG: eth0: transmit timed > out > Sep 16 16:10:00 ntbkmichael kernel: eth0: Transmit timeout, status 0c 0005 > c07f media 00. > Sep 16 16:10:00 ntbkmichael kernel: eth0: Tx queue start entry 4 dirty > entry 0. > Sep 16 16:10:00 ntbkmichael kernel: eth0: Tx descriptor 0 is 0008a04b. > (queue head) > Sep 16 16:10:00 ntbkmichael kernel: eth0: Tx descriptor 1 is 0008a04b. > Sep 16 16:10:00 ntbkmichael kernel: eth0: Tx descriptor 2 is 0008a05d. > Sep 16 16:10:00 ntbkmichael kernel: eth0: Tx descriptor 3 is 0008a05d. > Sep 16 16:10:00 ntbkmichael kernel: eth0: link up, 100Mbps, full-duplex, > lpa 0x45E1 > ' > > > Thank you very much for anything that would help me elucidate this mystery! > > > Kindest regards, > Michael
It's not that I have a great knowledge, but in your kernel.log I saw the expression WATCHDOG, that means (if I don't make a mistake) you have a count down that stop your connection. I don't know how to configure that, but for me the easiest way is to recompile the kernel, disabling the watchdog option. Usually we don't need this option. friendly, Serge.
