LinuxQuestions.org
Download your favorite Linux distribution at LQ ISO.
Go Back   LinuxQuestions.org > Forums > Linux Forums > Linux - Networking
User Name
Password
Linux - Networking This forum is for any issue related to networks or networking.
Routing, network cards, OSI, etc. Anything is fair game.

Notices


Reply
  Search this Thread
Old 02-25-2004, 11:49 AM   #1
artao
LQ Newbie
 
Registered: Feb 2004
Distribution: Mandrake 9.2
Posts: 22

Rep: Reputation: 15
network sporadic and slow IF it starts


First off, let me say I'm fairly new to Linux inasmuch as i've 'used' it (or Unix) on and off for years but this (Mandrake 9.2 on a Tyan Tomcat IV w/dual P200 and 128M RAM, SB Live!, 4G HD, SMC 9432TX netcard, Elsa Gloria Synergy 8M video) is my first install. I've tried Red Hat on the same box, as well as several CD bootable distros--though I've yet to try Knoppix which I've been reading is good--but they never installed/ran. I'm finding configuration in the Linux world strange and irritating, but I'm sure I'll grow to love it.
anyhow
My internet connection is cable, routed through a D-Link DI-604, which also has an XP box connected to it which works fine.
during boot I get 'bringing up interface eth0: failed'
once in Linux, logged in as root since i'm trying to get this hardware to work. Linux 'sees' the netcard for what it is, though the device is down. I've run some commands for which the output is below, including ifconfig which shows only lo. Once I do a dhclient eth0 I get an IP, but still no web, and ifconfig shows eth0. However, I can ping out, say .. to google, though it's really really slow. I can ping the router or the XP box, again slowly. Also, there are lots of time-outs. The machine can ping itself just fine. After a very short while, generally less than a minute, what connection it had is again gone. All but a self-ping gives network unreachable, and ifconfig again shows only lo as it did prior to dhclient. One other thing, watching the lights on the router (gotta love blinky lights!!) the light for the Linux box connection kinda pulses oooonnnnnn--offff--onn-off--on--offffffffffffffff--on--offfff etc...
I have tried a different cable (a couple actually) as well as connecting directly to the cable-modem, as well as using the cable from the linux box on the XP box (which worked fine).
I've been beating my head against this for a couple days now, and spent much much time searching these boards and those boards and that mailing list over there ...... ahhhhhhh!!!!! SO CLOSE!@!@@!!! Please help

I suppose I should also mention that I installed Samba right away because I (of course) want to do windows file-sharing, and it's my understanding that Linux likes it better if "it's all there right away" so to speak. I'm about ready to re-format and re-install ...

here's the outputs of some commands:

ifconfig before dhclient eth0 and after it looses the connection:

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:136 errors:0 dropped:0 overruns:0 frame:0
TX packets:136 errors:0 dropped:0 overruns:0 carrier:0
collisions:0 txqueuelen:0
RX bytes:9946 (9.7 Kb) TX bytes:9946 (9.7 Kb)



dhclient eth0

Internet Software Consortium DHCP Client V3.0.1rc11
Copyright 1995-2002 Internet Software Consortium.
All rights reserved.
For info, please visit [URL removed]
Listening on LPF/eth0/00:e0:29:07:d7:1e
Sending on LPF/eth0/00:e0:29:07:d7:1e
Sending on Socket/fallback
DHCPREQUEST on eth0 to 255.255.255.255 port 67
DHCPREQUEST on eth0 to 255.255.255.255 port 67
DHCPREQUEST on eth0 to 255.255.255.255 port 67
DHCPDISCOVER on eth0 to 255.255.255.255 port 67 interval 3
DHCPDISCOVER on eth0 to 255.255.255.255 port 67 interval 6
DHCPDISCOVER on eth0 to 255.255.255.255 port 67 interval 6
DHCPOFFER from 192.168.0.1
DHCPREQUEST on eth0 to 255.255.255.255 port 67
DHCPACK from 192.168.0.1
bound to 192.168.0.101 -- renewal in 246825 seconds.

odd how it kept requesting an IP after it was offered one, no? I've seen it go on longer than this example, with several offers.



ifconfig after this

eth0 Link encap:Ethernet HWaddr 00:E0:29:077:1E
inet addr:192.168.0.101 Bcast:192.168.0.255 Mask:255.255.255.0
UP BROADCAST RUNNING MULTICAST MTU:1500 Metric:1
RX packets:450 errors:0 dropped:0 overruns:0 frame:0
TX packets:323 errors:0 dropped:0 overruns:0 carrier:0
collisions:0 txqueuelen:100
RX bytes:135772 (132.5 Kb) TX bytes:35947 (35.1 Kb)
Interrupt:19 Base address:0x6100

lo Link encap:Local Loopback
UP LOOPBACK RUNNING MTU:16436 Metric:1
RX packets:276 errors:0 dropped:0 overruns:0 frame:0
TX packets:276 errors:0 dropped:0 overruns:0 carrier:0
collisions:0 txqueuelen:0
RX bytes:34767 (33.9 Kb) TX bytes:34767 (33.9 Kb)



and here's the output of lsmod

Module Size Used by Not tainted
nls_iso8859-1 3484 1 (autoclean)
nls_cp850 4284 1 (autoclean)
vfat 12012 1 (autoclean)
fat 39320 0 (autoclean) [vfat]
autofs4 11924 2 (autoclean)
emu10k1 67852 1
sound 75040 0 [emu10k1]
soundcore 6916 0 [emu10k1 sound]
ac97_codec 16056 0 [emu10k1]
nfsd 76336 8 (autoclean)
af_packet 16264 7 (autoclean)
ide-floppy 16224 0 (autoclean)
ide-tape 49712 0 (autoclean)
ide-cd 34052 0 (autoclean)
cdrom 33056 0 (autoclean) [ide-cd]
floppy 56988 2
epic100 15116 1 (autoclean)
mii 3996 0 (autoclean) [epic100]
supermount 84480 2 (autoclean)
rtc 10028 0 (autoclean)
ext3 63024 2
jbd 42760 2 [ext3]


finally here's the only thing if seen ifup eth0 report

Determining IP information for eth0... failed; no link present. Check cable?

and it seems like any connection I may have had I lose after this command.


One last thing, here's my /etc/sysconfig/network-scripts/ifcfg-eth0

DEVICE=eth0
BOOTPROTO=dhcp
NETMASK=255.255.255.0
ONBOOT=yes
NEEDHOSTNAME=yes


uh .... i guess that's it for now, sorry about the length, just trying to be thorough .... B-)
 
Old 02-25-2004, 12:08 PM   #2
bnice
Member
 
Registered: Feb 2004
Location: Sacramento, CA
Distribution: Slack 9.1, slackware-current
Posts: 284

Rep: Reputation: 30
try this at the command line:

route add default gw 192.168.0.1


your driver module is epic100 and seems to load OK, but is it being loaded at boot or when you call dhclient?
 
Old 02-25-2004, 12:44 PM   #3
artao
LQ Newbie
 
Registered: Feb 2004
Distribution: Mandrake 9.2
Posts: 22

Original Poster
Rep: Reputation: 15
fresh boot. immediately upon root login that command returns:

SIOCADDRT: Network is unreachable


after doing dhclient eth0 then the 'route ...' command, I get:

SIOCADDRT: File exists



I don't know when epic100 loads, but on boot eth0 ALWAYS fails to start 'bringing up interface eth0: failed' ... I do have eth0 set to load on startup


one other thing ..... do I need to/how do I set a hostname for the Linux box. I thought I did at install, but it still just refers to itself as localhost. And what's the difference between Host name and Zeroconf Host name. I notice here and now that for my default profile under LAN configuration in the Mandrake Configuration Utility, when I hit the button 'Configure hostname', the window that comes up says the Zeroconf Hose name is krsna (what I set it as during netcard install) but the DNS server says 127.0.0.1 .... that doesn't seem right. isn't the router the DNS server for this?? and the router's address is 192.168.0.1 on the local net .... hmmmm .... and i shouldn't have to set a gateway, right??
dang it!!! XP just works when i plug it into the router!! why is Linux having this hangup?? During install, the Mandrake Update phase, it never complained about not being able to connect to the internet, so I assume it was able to at that point .... hmmmmm
 
Old 02-25-2004, 12:49 PM   #4
artao
LQ Newbie
 
Registered: Feb 2004
Distribution: Mandrake 9.2
Posts: 22

Original Poster
Rep: Reputation: 15
I just did that command again after it lost the connection ...

SIOCADDRT: Network is unreachable

again. Just updating.
 
Old 02-25-2004, 01:18 PM   #5
bnice
Member
 
Registered: Feb 2004
Location: Sacramento, CA
Distribution: Slack 9.1, slackware-current
Posts: 284

Rep: Reputation: 30
let's backtrack a little

did you run `drakconnect` to set up your network?
 
Old 02-25-2004, 01:35 PM   #6
artao
LQ Newbie
 
Registered: Feb 2004
Distribution: Mandrake 9.2
Posts: 22

Original Poster
Rep: Reputation: 15
if you mean the network wizard thing, yeah. (I just typed drakconnect into a console, and the network wizard thing came up). I've run it several times. Both choosing LAN and choosing Cable, and choosing both together.
I've tried removing the card and booting linux to make it go away, then putting it back and booting again. I've tried checking 'Enable ACPI' and 'Force no APIC' in the drakboot bootloader configurator ...... and more .....

right now my network/internet configuration is all autodetect, LAN, dhcp .... i'll be glad to do it again if you have any suggestions
 
Old 02-25-2004, 04:24 PM   #7
bnice
Member
 
Registered: Feb 2004
Location: Sacramento, CA
Distribution: Slack 9.1, slackware-current
Posts: 284

Rep: Reputation: 30
Sorry to be so long in replying... power failure.

you should have a file in /etc/sysconfig/network-scripts named ifcfg-eth0. This is mine:

Code:
DEVICE=eth0
BOOTPROTO=static
IPADDR=192.168.0.5
NETMASK=255.255.255.0
NETWORK=192.168.0.0
BROADCAST=192.168.0.255
ONBOOT=yes
MII_NOT_SUPPORTED=yes
make sure the line reads ONBOOT=yes
the other stuff'll be different since you use DHCP
 
Old 02-25-2004, 06:56 PM   #8
artao
LQ Newbie
 
Registered: Feb 2004
Distribution: Mandrake 9.2
Posts: 22

Original Poster
Rep: Reputation: 15
here's what mine looks like

DEVICE=eth0
BOOTPROTO=dhcp
ONBOOT=yes
NETMASK=255.255.255.0
NEEDHOSTNAME=yes

there is also a ifcfg-eth0 in /etc/netprofile/profiles/default/files/etc/sysconfig/network-scripts .... is this something I should be concerned about? why is it here as well? also it's different by one line, that is, the NETMASK line isn't there. Should I change it to match the file in /etc/sysconfig/network-scripts ?
i've tried taking out that last line, seems to me it shouldn't be there, but with no result. However, instinct tells me that shouldn't be there ... unless it means it needs to REQUEST a hostname .... should that be there??? I don't think i can set the router to assign a hostname.
And what's that MII_NOT_SUPPORTED?? Is that something i SHOULD have in mine? it shows up when i do an lsmod (see above), but what is it?

LONG!! way to go to learn Linux ....
 
Old 02-25-2004, 07:44 PM   #9
artao
LQ Newbie
 
Registered: Feb 2004
Distribution: Mandrake 9.2
Posts: 22

Original Poster
Rep: Reputation: 15
something else i just noticed while browsing through /boot/config .... there's a line that reads (under the Ethernet (10 or 100Mbit) section)

CONFIG_NET_VENDOR_3COM=y
and also
CONFIG_NET_VENDOR_SMC=y

could that be causing some kind of conflict? I have no 3com devices in the machine. Is the SMC using a 3com driver (the epic100 or mii)??

time to take a break for awhile, been at this all day .... well, for the last 24 hours, literally.

l8r
 
Old 02-26-2004, 11:13 AM   #10
bnice
Member
 
Registered: Feb 2004
Location: Sacramento, CA
Distribution: Slack 9.1, slackware-current
Posts: 284

Rep: Reputation: 30
Let's check some more files:

This is where you set your hostname

/etc/sysconfig/network
Code:
HOSTNAME=gandalf.shire.org
NETWORKING=yes
GATEWAY=192.168.0.1     #router IP
nameserver stuff

/etc/resolv.conf
Code:
search shire.org
nameserver 192.168.0.1  #router IP
you may also set hostname here, Should match with /etc/sysconfig/network

/etc/hosts
Code:
127.0.0.1		  localhost
192.168.0.5		gandalf.shire.org gandalf
CONFIG_NET_VENDOR_3COM=y means support for 3COM cards is included in your kernel
you don't need this.

MII_NOT_SUPPORTED=yes This is specific to my card (Intel EtherExpress) since epic100 is loading this module, it probably needs it.

A peek at the relevant section of dmesg would be helpful, example:
Code:
eepro100.c:v1.09j-t 9/29/99 Donald Becker http://www.scyld.com/network/eepro100.html
eepro100.c: $Revision: 1.36 $ 2000/11/17 Modified by Andrey V. Savochkin <saw@saw.sw.com.sg> and others
eth0: 0000:02:08.0, 00:07:E9:94:B5:0F, I/O at 0xdc00, IRQ 11.
  Board assembly 000000-000, Physical connectors present: RJ45
  Primary interface chip i82555 PHY #1.
  General self-test: passed.
  Serial sub-system self-test: passed.
  Internal registers self-test: passed.
  ROM checksum self-test: passed (0x04f4518b).
you're probably learning more about networking than you *ever* wanted to know!
 
Old 02-26-2004, 04:46 PM   #11
jludwig
Member
 
Registered: Feb 2004
Distribution: FC6
Posts: 32

Rep: Reputation: 15
I had the same problem with my test network at my old job testing a XP box
It turns out that XP can spew junk all over the net when it first starts.
This will shut many routers/switches/etc down because the firmware in the transciever chips
will turn off if there is sporatic noise.
Something line a sniffer may be able to determine if this is what is happening.
 
Old 02-26-2004, 06:54 PM   #12
artao
LQ Newbie
 
Registered: Feb 2004
Distribution: Mandrake 9.2
Posts: 22

Original Poster
Rep: Reputation: 15
hmmmm ... my config files look all screwy. Maybe it'd be best to just reformat/reinstall. I musta run the networking wizard 10 times or so. I s'pose I could try another netcard, I've got an old Winbond 10/100 somewhere. Anyhow, here's what I get when I check the config files you mentioned:

/etc/sysconfig/network
NETWORKING=yes
GATEWAY=192.168.0.1

/etc/resolv.conf
search blu local
nameserver 127.0.0.1
nameserver 192.168.0.1

/etc/hosts
127.0.0.1 localhost


and the relevant section of dmesg .... uh, which part is that There's some crazy-looking stuff in there ... I figure you mean the part about the net card

epic100.c:v1.11 1/7/2001 Written by Donald Becker <becker@scyld.com>
http://www.scyld.com/network/epic100.html
(unofficial 2.4.x kernel port, version 1.11+LK1.1.14, Aug 4, 2002)
epic100(00:11.0): MII transceiver #3 control 3000 status 7809.
epic100(00:11.0): Autonegotiation advertising 01e1 link partner 0001.
eth0: SMSC EPIC/100 83c170 at 0x6100, IRQ 19, 00:e0:29:07:d7:1e.

also this

eth0: Setting full-duplex based on MII #3 link partner capability of 45e1.
eth0: Setting half-duplex based on MII #3 link partner capability of 0001.
eth0: Setting full-duplex based on MII #3 link partner capability of 45e1.
eth0: Setting half-duplex based on MII #3 link partner capability of 0001.

and then this goes on for awhile, many lines, sectore keeps increasing ... hmmmmm

SCSI subsystem driver Revision: 1.00
udf: registering filesystem
UDF-fs DEBUG lowlevel.c:57:udf_get_last_session: XA disk: no, vol_desc_start=32
UDF-fs DEBUG super.c:1420:udf_read_super: Multi-session=0
UDF-fs DEBUG super.c:413:udf_vrs: Starting at sector 16 (2048 byte sectors)
hdc: command error: status=0x51 { DriveReady SeekComplete Error }
hdc: command error: error=0x50
end_request: I/O error, cmd 0 dev 16:00 (hdc), sector 64
UDF-fs DEBUG super.c:1160:udf_check_valid: Failed to read byte 32768. Assuming open disc. Skipping validity check
hdc: command error: status=0x51 { DriveReady SeekComplete Error }
hdc: command error: error=0x50
end_request: I/O error, cmd 0 dev 16:00 (hdc), sector 1081264
hdc: command error: status=0x51 { DriveReady SeekComplete Error }
hdc: command error: error=0x50

and at the very start was this:

cpu: 1, clocks: 664672, slice: 221557
CPU1<T0:664672,T1:221552,D:6,S:221557,C:664672>
checking TSC synchronization across CPUs: passed.

does that mean linux thinks the CPUs are 66mhz?? they're both P200MMX. I also noticed in another file yesterday (forget which) a section that seemed to indicate MMX support wasn't going.

learning more about networking that I wanted to? learning more about networking than I knew there was to learn ... 8-|
 
Old 02-26-2004, 07:01 PM   #13
artao
LQ Newbie
 
Registered: Feb 2004
Distribution: Mandrake 9.2
Posts: 22

Original Poster
Rep: Reputation: 15
Talking

not only am i not really sure WHAT a sniffer is ..... i wouldn't know what to do with one, or how to interpret what it told me
 
Old 02-27-2004, 12:49 AM   #14
artao
LQ Newbie
 
Registered: Feb 2004
Distribution: Mandrake 9.2
Posts: 22

Original Poster
Rep: Reputation: 15
well ... how about that!!! Swapped out the SMC netcard for some generic card with a Winbond chipset and violla!! network is up!!! yay!!!


now to get XP and Linux talking ...
 
  


Reply


Thread Tools Search this Thread
Search this Thread:

Advanced Search

Posting Rules
You may not post new threads
You may not post replies
You may not post attachments
You may not edit your posts

BB code is On
Smilies are On
[IMG] code is Off
HTML code is Off



Similar Threads
Thread Thread Starter Forum Replies Last Post
Network fails when X starts Vie Linux - Wireless Networking 0 07-13-2005 12:24 PM
Software starts slow ... :( Quatsch_Z Slackware 17 05-01-2005 08:05 AM
Fluxbox starts slow/X11 keeps CPU running at 9x% kaon Slackware 0 02-10-2005 11:28 AM
sporadic network problems on direct connection pfunk Linux - Networking 0 10-14-2004 03:09 PM
Change hostname in X, sendmail starts slow Guru3 Red Hat 4 03-19-2004 03:44 PM

LinuxQuestions.org > Forums > Linux Forums > Linux - Networking

All times are GMT -5. The time now is 02:15 PM.

Main Menu
Advertisement
My LQ
Write for LQ
LinuxQuestions.org is looking for people interested in writing Editorials, Articles, Reviews, and more. If you'd like to contribute content, let us know.
Main Menu
Syndicate
RSS1  Latest Threads
RSS1  LQ News
Twitter: @linuxquestions
Facebook: linuxquestions Google+: linuxquestions
Open Source Consulting | Domain Registration