| View previous topic :: View next topic |
| Author |
Message |
ericp
n00b
Joined: 13 May 2006
Posts: 9
|
 Posted: Sat May 13, 2006 3:56 pm Post subject: eth0 dissapears from ifconfig after 10seconds [Solved] |
 |
|
Hi folks, relatively new here...
System configuration
==============
Machine running 2006.0 with two NICs.
Via-Rhine and 3Com
Linksys Router
On the linux system, eth0 = internal (to router), eth1 = directconnect to internet (cable modem)
I have absolutely no problems with eth1 (right now, knock on wood)
My problem exists with eth0.
When I perform an
the result is
| Code: |
eth1 Link encap:Ethernet HWaddr 00:02:89:D8:F0:A9
inet addr:xx.xx.xx.xx Bcast:255.255.255.255 Mask:255.255.252.0
UP BROADCAST NOTRAILERS RUNNING MULTICAST MTU:1500 Metric:1
RX packets:211353 errors:0 dropped:0 overruns:0 frame:0
TX packets:8546 errors:0 dropped:0 overruns:0 carrier:0
collisions:0 txqueuelen:1000
RX bytes:13545030 (12.9 Mb) TX bytes:2156892 (2.0 Mb)
Interrupt:10 Base address:0xdc00
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:6 errors:0 dropped:0 overruns:0 frame:0
TX packets:6 errors:0 dropped:0 overruns:0 carrier:0
collisions:0 txqueuelen:0
RX bytes:354 (354.0 b) TX bytes:354 (354.0 b)
|
However, if I do a:
I get:
| Code: |
eth0 Link encap:Ethernet HWaddr 00:50:04:82:F0:CF
BROADCAST MULTICAST MTU:1500 Metric:1
RX packets:6488 errors:0 dropped:0 overruns:0 frame:0
TX packets:9688 errors:0 dropped:0 overruns:0 carrier:0
collisions:0 txqueuelen:1000
RX bytes:711335 (694.6 Kb) TX bytes:1293952 (1.2 Mb)
Interrupt:11 Base address:0x2000
eth1 Link encap:Ethernet HWaddr 00:02:89:D8:F0:A9
inet addr:xx.xx.xx.xx Bcast:255.255.255.255 Mask:255.255.252.0
UP BROADCAST NOTRAILERS RUNNING MULTICAST MTU:1500 Metric:1
RX packets:213469 errors:0 dropped:0 overruns:0 frame:0
TX packets:8557 errors:0 dropped:0 overruns:0 carrier:0
collisions:0 txqueuelen:1000
RX bytes:13672226 (13.0 Mb) TX bytes:2158718 (2.0 Mb)
Interrupt:10 Base address:0xdc00
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:6 errors:0 dropped:0 overruns:0 frame:0
TX packets:6 errors:0 dropped:0 overruns:0 carrier:0
collisions:0 txqueuelen:0
RX bytes:354 (354.0 b) TX bytes:354 (354.0 b)
|
If I do dhcpd eth0 (or a /etc/init.d/net.eth0 start), it does pick up the IP address from router (192.168.1.99, it's assigned by the MAC address), but after about 10-30seconds, it "loses" the IP address and it's done.
| Code: |
home conf.d # /etc/init.d/net.eth0 restart
* Stopping eth0
* Bringing down eth0
* Shutting down eth0 ... [ ok ]
* Starting eth0
* Bringing up eth0
* dhcp
* Running dhcpcd ... [ ok ]
* eth0 received address 192.168.1.99
home conf.d #
|
result of lspci:
| Code: |
00:00.0 Host bridge: VIA Technologies, Inc. VT82C693A/694x [Apollo PRO133x] (rev c4)
00:01.0 PCI bridge: VIA Technologies, Inc. VT82C598/694x [Apollo MVP3/Pro133x AGP]
00:07.0 ISA bridge: VIA Technologies, Inc. VT82C686 [Apollo Super South] (rev 40)
00:07.1 IDE interface: VIA Technologies, Inc. VT82C586A/B/VT82C686/A/B/VT823x/A/C PIPC Bus Master IDE (rev 06)
00:07.2 USB Controller: VIA Technologies, Inc. VT82xxxxx UHCI USB 1.1 Controller (rev 16)
00:07.3 USB Controller: VIA Technologies, Inc. VT82xxxxx UHCI USB 1.1 Controller (rev 16)
00:07.4 Bridge: VIA Technologies, Inc. VT82C686 [Apollo Super ACPI] (rev 40)
00:09.0 Ethernet controller: VIA Technologies, Inc. VT6105 [Rhine-III] (rev 8b)
00:0b.0 Ethernet controller: 3Com Corporation 3c905B 100BaseTX [Cyclone] (rev 30)
01:00.0 VGA compatible controller: ATI Technologies Inc Rage 128 RF/SG AGP
|
result of lsmod:
| Code: |
Module Size Used by
via_rhine 19916 -
3c59x 39536 -
|
Contents of /etc/conf.d/net
| Code: |
config_eth0=( "dhcp" )
dhcp_eth0="nodns nontp nonis"
config_eth1=( "dhcp" )
dhcp_eth1="nodns nontp nonis"
|
I'm really scratching my head on this one --- the reason I'm going this direction is that I want the home PC's (windows) to be behind the firewall router, but I want to have my server machine "straddling" the networks. I've a feeling that what I'm trying to do is not going to work, and that I'm setting myself up for a sort of routing confusion.
If performed the suggestions posted in https://forums.gentoo.org/viewtopic-t-461440.html, but alas no go.
If you need any additional information, please feel free to request, and I"ll do my best to reply.
Thanks in advance!
Eric.
Last edited by ericp on Tue May 16, 2006 2:29 am; edited 1 time in total |
|
| Back to top |
|
 |
dgaffuri
Advocate
Joined: 05 Jun 2005
Posts: 2078
Location: Italy
|
| Posted: Sat May 13, 2006 4:05 pm Post subject: |
|
|
Do you have some error in dmesg when eth0 goes down?
_________________
Adopt an unanswered post
If you feel that your problem has been solved please edit the top post and add [solved] to the subject |
|
| Back to top |
|
|
thepustule
Apprentice
Joined: 22 Feb 2004
Posts: 212
Location: Toronto, Canada
|
| Posted: Sat May 13, 2006 4:46 pm Post subject: |
|
|
| whenever eth0 disappears, does it also disappear from ifconfig -a? |
|
| Back to top |
|
|
ericp
n00b
Joined: 13 May 2006
Posts: 9
|
| Posted: Sat May 13, 2006 5:09 pm Post subject: |
|
|
| Quote: |
Do you have some error in dmesg when eth0 goes down?
|
Hmm...
Missed this before:
| Code: |
<snip>
IRQ routing conflict for 0000:00:07.2, have irq 10, want irq 11
IRQ routing conflict for 0000:00:07.3, have irq 10, want irq 11
3c59x: Donald Becker and others. www.scyld.com/network/vortex.html
0000:00:0b.0: 3Com PCI 3c905B Cyclone 100baseTx at e0852000. Vers LK1.1.19
via-rhine.c:v1.10-LK1.2.0-2.6 June-10-2004 Written by Donald Becker
PCI: Found IRQ 10 for device 0000:00:09.0
eth1: VIA Rhine III at 0x1dc00, 00:02:89:d8:f0:a9, IRQ 10.
eth1: MII PHY found at address 1, status 0x786d advertising 05e1 Link 45e1.
|
I thought that DMESG was only a bootup log --- I"m guessing I'm wrong? (wish there were timestamps  )
| Quote: |
whenever eth0 disappears, does it also disappear from ifconfig -a?
|
Nope, if I do an ifconfig -a, I get as I posted above:
| Code: |
eth0 Link encap:Ethernet HWaddr 00:50:04:82:F0:CF
BROADCAST MULTICAST MTU:1500 Metric:1
RX packets:6488 errors:0 dropped:0 overruns:0 frame:0
TX packets:9688 errors:0 dropped:0 overruns:0 carrier:0
collisions:0 txqueuelen:1000
RX bytes:711335 (694.6 Kb) TX bytes:1293952 (1.2 Mb)
Interrupt:11 Base address:0x2000
eth1 Link encap:Ethernet HWaddr 00:02:89:D8:F0:A9
inet addr:xx.xx.xx.xx Bcast:255.255.255.255 Mask:255.255.252.0
UP BROADCAST NOTRAILERS RUNNING MULTICAST MTU:1500 Metric:1
RX packets:213469 errors:0 dropped:0 overruns:0 frame:0
TX packets:8557 errors:0 dropped:0 overruns:0 carrier:0
collisions:0 txqueuelen:1000
RX bytes:13672226 (13.0 Mb) TX bytes:2158718 (2.0 Mb)
Interrupt:10 Base address:0xdc00
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:6 errors:0 dropped:0 overruns:0 frame:0
TX packets:6 errors:0 dropped:0 overruns:0 carrier:0
collisions:0 txqueuelen:0
RX bytes:354 (354.0 b) TX bytes:354 (354.0 b)
|
which I assume (probably incorrectly) that the drivers are loaded, and the hardware detected successfully?
Any other files you wish to see?
Thanks again!
Eric |
|
| Back to top |
|
|
dgaffuri
Advocate
Joined: 05 Jun 2005
Posts: 2078
Location: Italy
|
| Posted: Sat May 13, 2006 5:38 pm Post subject: |
|
|
| ericp wrote: | | I thought that DMESG was only a bootup log --- I"m guessing I'm wrong? (wish there were timestamps ) |
dmesg shows the kernel message buffer, which is saved (with time stamps) by the syslog deamon, usually in /var/log/messages.
| ericp wrote: | | Code: |
<snip>
IRQ routing conflict for 0000:00:07.2, have irq 10, want irq 11
IRQ routing conflict for 0000:00:07.3, have irq 10, want irq 11
3c59x: Donald Becker and others. www.scyld.com/network/vortex.html
0000:00:0b.0: 3Com PCI 3c905B Cyclone 100baseTx at e0852000. Vers LK1.1.19
via-rhine.c:v1.10-LK1.2.0-2.6 June-10-2004 Written by Donald Becker
PCI: Found IRQ 10 for device 0000:00:09.0
eth1: VIA Rhine III at 0x1dc00, 00:02:89:d8:f0:a9, IRQ 10.
eth1: MII PHY found at address 1, status 0x786d advertising 05e1 Link 45e1.
|
|
This is at boot time. Do you have 3c59x built as a module? If so try to remove it and modprobe again, and check dmesg last lines.
_________________
Adopt an unanswered post
If you feel that your problem has been solved please edit the top post and add [solved] to the subject |
|
| Back to top |
|
|
ericp
n00b
Joined: 13 May 2006
Posts: 9
|
| Posted: Sat May 13, 2006 6:00 pm Post subject: |
|
|
| dgaffuri wrote: |
This is at boot time. Do you have 3c59x built as a module? If so try to remove it and modprobe again, and check dmesg last lines. |
Hmm...I was going to say that the NICs are build as modules, but upon closer inspection in my module.autoload.d/kernel-2.6, it may not be --- there are no lines for the via-rhine/3x59x in it (actually, that entire file is commented out)...I'm assuming I've built them into the kernel? I honestly don't remember 
If so, should I "module"ize them? |
|
| Back to top |
|
|
dgaffuri
Advocate
Joined: 05 Jun 2005
Posts: 2078
Location: Italy
|
| Posted: Sat May 13, 2006 10:14 pm Post subject: |
|
|
| ericp wrote: | | Hmm...I was going to say that the NICs are build as modules, but upon closer inspection in my module.autoload.d/kernel-2.6, it may not be --- there are no lines for the via-rhine/3x59x in it (actually, that entire file is commented out)...I'm assuming I've built them into the kernel? I honestly don't remember |
If you've coldplug or a recent udev modules are loaded even if not in autoload.d. You may check with modprobe -l.
| ericp wrote: | | If so, should I "module"ize them? |
Not at all, I was just asking to check kernel messages on modprobe.
_________________
Adopt an unanswered post
If you feel that your problem has been solved please edit the top post and add [solved] to the subject |
|
| Back to top |
|
|
ericp
n00b
Joined: 13 May 2006
Posts: 9
|
| Posted: Sat May 13, 2006 10:20 pm Post subject: |
|
|
results of modprobe -l:
| Code: |
/lib/modules/2.6.15-gentoo-r1/kernel/net/ipv4/netfilter/iptable_raw.ko
/lib/modules/2.6.15-gentoo-r1/kernel/net/ipv4/netfilter/ipt_NOTRACK.ko
/lib/modules/2.6.15-gentoo-r1/kernel/lib/libcrc32c.ko
/lib/modules/2.6.15-gentoo-r1/kernel/drivers/usb/misc/phidgetservo.ko
/lib/modules/2.6.15-gentoo-r1/kernel/drivers/usb/misc/cytherm.ko
/lib/modules/2.6.15-gentoo-r1/kernel/drivers/usb/input/touchkitusb.ko
/lib/modules/2.6.15-gentoo-r1/kernel/drivers/scsi/sata_sx4.ko
/lib/modules/2.6.15-gentoo-r1/kernel/drivers/scsi/sata_sis.ko
/lib/modules/2.6.15-gentoo-r1/kernel/drivers/scsi/ipr.ko
/lib/modules/2.6.15-gentoo-r1/kernel/drivers/scsi/dpt_i2o.ko
/lib/modules/2.6.15-gentoo-r1/kernel/drivers/net/via-rhine.ko
/lib/modules/2.6.15-gentoo-r1/kernel/drivers/net/s2io.ko
/lib/modules/2.6.15-gentoo-r1/kernel/drivers/net/dummy.ko
/lib/modules/2.6.15-gentoo-r1/kernel/drivers/net/3c59x.ko
/lib/modules/2.6.15-gentoo-r1/kernel/drivers/base/firmware_class.ko
|
I do have these lines repeating over and over again in DMESG:
| Code: |
IRQ routing conflict for 0000:00:07.2, have irq 10, want irq 11
IRQ routing conflict for 0000:00:07.3, have irq 10, want irq 11
|
Could it be really as simple as an IRQ conflict between the NIC cards? I'm more a windows person, so I know how to check/change via the Control Panel, but is there something that I can do on the console side in Linux to help rectify the IRQ conflict?
I really appreciate the assistance 
eric |
|
| Back to top |
|
|
dgaffuri
Advocate
Joined: 05 Jun 2005
Posts: 2078
Location: Italy
|
| Posted: Sat May 13, 2006 10:47 pm Post subject: |
|
|
| ericp wrote: | results of modprobe -l: | Code: |
....
/lib/modules/2.6.15-gentoo-r1/kernel/drivers/net/via-rhine.ko
....
/lib/modules/2.6.15-gentoo-r1/kernel/drivers/net/3c59x.ko
.... |
|
They're modules
| ericp wrote: | I do have these lines repeating over and over again in DMESG:
| Code: |
IRQ routing conflict for 0000:00:07.2, have irq 10, want irq 11
IRQ routing conflict for 0000:00:07.3, have irq 10, want irq 11
|
Could it be really as simple as an IRQ conflict between the NIC cards? I'm more a windows person, so I know how to check/change via the Control Panel, but is there something that I can do on the console side in Linux to help rectify the IRQ conflict? |
Could be, may you post lspci and cat /proc/interrupts output?
_________________
Adopt an unanswered post
If you feel that your problem has been solved please edit the top post and add [solved] to the subject |
|
| Back to top |
|
|
ericp
n00b
Joined: 13 May 2006
Posts: 9
|
| Posted: Sun May 14, 2006 4:19 am Post subject: |
|
|
lspci:
| Code: |
00:00.0 Host bridge: VIA Technologies, Inc. VT82C693A/694x [Apollo PRO133x] (rev c4)
00:01.0 PCI bridge: VIA Technologies, Inc. VT82C598/694x [Apollo MVP3/Pro133x AGP]
00:07.0 ISA bridge: VIA Technologies, Inc. VT82C686 [Apollo Super South] (rev 40)
00:07.1 IDE interface: VIA Technologies, Inc. VT82C586A/B/VT82C686/A/B/VT823x/A/C PIPC Bus Master IDE (rev 06)
00:07.2 USB Controller: VIA Technologies, Inc. VT82xxxxx UHCI USB 1.1 Controller (rev 16)
00:07.3 USB Controller: VIA Technologies, Inc. VT82xxxxx UHCI USB 1.1 Controller (rev 16)
00:07.4 Bridge: VIA Technologies, Inc. VT82C686 [Apollo Super ACPI] (rev 40)
00:09.0 Ethernet controller: VIA Technologies, Inc. VT6105 [Rhine-III] (rev 8b)
00:0b.0 Ethernet controller: 3Com Corporation 3c905B 100BaseTX [Cyclone] (rev 30)
01:00.0 VGA compatible controller: ATI Technologies Inc Rage 128 RF/SG AGP
|
Looks like the DMESG messages I was getting are in regards to the USB controllers and the 3COM driver fighting over the same IRQ  .
cat /proc/interrupts:
| Code: |
CPU0
0: 12518026 XT-PIC timer
2: 0 XT-PIC cascade
7: 0 XT-PIC parport0
10: 3344247 XT-PIC uhci_hcd:usb1, uhci_hcd:usb2, eth1
14: 38653 XT-PIC ide0
15: 42 XT-PIC ide1
NMI: 0
LOC: 0
ERR: 0
MIS: 0
|
I believe my BIOS has one of those "Let the OS determine PnP" --- would it be worth my while to manually assign IRQs to the different slots? Part of me also wants to move the offending NIC (3Com) to another PCI slot and see if that clears up any IRQ conflicts |
|
| Back to top |
|
|
dgaffuri
Advocate
Joined: 05 Jun 2005
Posts: 2078
Location: Italy
|
| Posted: Sun May 14, 2006 9:13 am Post subject: |
|
|
Are you booting with noacpi? May you post your kernel command line? I don't see interrupts enabled for your eth0, while from dmesg it seems that it uses IRQ 11. I'm really just guessing, I'm not an expert in this.
_________________
Adopt an unanswered post
If you feel that your problem has been solved please edit the top post and add [solved] to the subject |
|
| Back to top |
|
|
ericp
n00b
Joined: 13 May 2006
Posts: 9
|
| Posted: Mon May 15, 2006 2:00 pm Post subject: |
|
|
| Quote: |
re you booting with noacpi? May you post your kernel command line? I don't see interrupts enabled for your eth0, while from dmesg it seems that it uses IRQ 11. I'm really just guessing, I'm not an expert in this.
|
Sorry about the delay --- I haven't had a chance to touch the machine yesterday, and I'm at work today, so anything I can do "physically" would be later today.
One thing I have noticed (since I was running remotely) is that if I bring up the net.eth0, the system "locks" up when viewing it remotely.
By locks up, I mean the following. I'm currently using SSH to administer the machine from work. As soon as I execute /etc/init.d/net.eth0 start, eth1 quits responding/updating (eth1 is my external NIC, which sits on the internet directly). As soon as eth0 goes away again, eth1 starts functioning again.
I'm definately thinking there's a conflict with this. I'm going to try to move the network card to another PCI slot tonight when I get home, and hope for the best
Will update once I get home.
Thanks again for the assistance.
Eric. |
|
| Back to top |
|
|
ericp
n00b
Joined: 13 May 2006
Posts: 9
|
| Posted: Tue May 16, 2006 1:34 am Post subject: |
|
|
I checked my command line, and I"m not booting with "noacpi"....I did add it to the config, but that made all network devices undetectable upon reboot --- so I changed it back..
I also played around with NIC position in the chassis, as well as swapping out both the VIA-Rhine and 3c59x cards with another 3c59x...
All produce nothing 
Another thing of note, I went ahead and set up the "culprit" NIC to a static IP address, and all networking dropped...had to reboot the machine (lost my SSH connection).
Is it possible that the way I'm physically connected, that's my problem is simply a routing thing, and Linux has the smarts to automatically disable the offending connection?
For those that are still following along, my topology looks something like this (pardon the bad ASCII drawing )
| Code: |
_______________
Internet/Cable Modem <----> Switch <----> Router <----> eth0 | 192.168.x.y |
^ | Linux Machine |
\-----------------------> eth1 |_<outside IP>__|
|
Thanks to all that have replied so far --- I hope to get this resolved if it kills me   |
|
| Back to top |
|
|
ericp
n00b
Joined: 13 May 2006
Posts: 9
|
| Posted: Tue May 16, 2006 2:14 am Post subject: |
|
|
Ok! Got some progress --- it looks like it's definately a routing issue..
In my resolv.conf, I had:
nameserver <external ISP nameserver1>
nameserver <external ISP nameserver2>
What *seemed* to be happening, was that when eth0 came online, it would get the internal IP address, but would try to resolve the IP with the external ISP nameservers, as opposed to the nameserver hosted by the router.
As soon as I added/changed one of the nameservers in the resolv.conf to be 192.168.1.1 (the ip address of the router) everything worked.
I then had to be stupid and reboot, and not modify my /etc/conf.d/net so now it was overwritten with the eth1 information again 
From what I've read in the net.example, the DHCP will by default, overwrite the nameserver information, unless I put the "nodns" information in the config.
(Someone please correct me)
If anyone sees a flaw in this, please let me know --- is there any special order to bring up my adapters, or is it as long as I have the right options,k I"m good......
Thanks again
e. |
|
| Back to top |
|
|
ericp
n00b
Joined: 13 May 2006
Posts: 9
|
| Posted: Tue May 16, 2006 2:28 am Post subject: |
|
|
Finally, I can put [SOLVED] on this
It turned out to be a routing issue (previous post of mine was a correct assumption)
For those that "may" get the same problem as me:
Here is my resolv.conf file:
| Code: |
# Generated by dhcpcd for interface eth0
nameserver 192.168.1.1
nameserver <external nameserver1>
nameserver <external nameserver2>
domain mydomain.com
|
Where
external nameserver1 & 2 are my ISPs DNS servers
192.168.1.1 is the nameserver of my internal network
Here is my /etc/conf.d/net file:
| Code: |
config_eth1=( "192.168.1.2 netmask 255.255.255.0 brd 192.168.1.255" )
routes_eth1=( "default gw 192.168.1.1" )
config_eth0=( "dhcp" )
dhcp_eth0="nodns"
|
The "nodns" part is what prevented the resolv.conf file from being overwritten.
I should have RTFMed a little earlier I think
Thanks to all that assisted, I feel a little better now that I muddled my way through this |
|
| Back to top |
|
|
|
|
You cannot post new topics in this forum
You cannot reply to topics in this forum
You cannot edit your posts in this forum
You cannot delete your posts in this forum
You cannot vote in polls in this forum
|
|
Networking & Security
