Void's Forums

by **dishawjp** » Tue Mar 22, 2005 7:20 pm

Hi All,

Still trying to get my FC2 boot to connect properly to the internet. Still can't get a connection to a DNS server. I've been looking at dhclient and mDNSResponder thinking that one of them may be the issue, but really don't know much about either.

I do know that when I boot to FC3, one of my messages at boot is "starting mDNSResponder." I don't get this when booting FC2. I googled mDNSResponder and it seems to refer to an Apple "rendezVous" thing so I don't know if it's even related. Dunno if that's even related to this issue.

I do have dhclient, which may be something related to getting DHCP going installed on both OS's, but a "# chkconfig --list | grep dhclient" shows nothing on either system. I do get a return from a "# chkconfig --list | grep mDNSResponder" on my FC3 boot. Could this be related to the problem?

For more info on this whole issue, please look at:

viewtopic.php?t=1423

Thanks in advance for any assistance. This is not a "critical" issue since I use my FC3 boot for most things, but am interested in figuring out what went wrong and how to set it right.

Jim Dishaw

by **Void Main** » Tue Mar 22, 2005 8:22 pm

Nope, mDNSResponder is in no way related. It's one of the first things I turn off (among many many other things). dhclient is not a service so you will not see it when "checkconfig". Your "ifup" script calls this which is called from your "network" service. See /etc/sysconfig/network-scripts/ifup. For comparison, here's what dhclient has entered into my /var/log/messages when I booted my laptop this evening:

Code: Select all: Mar 22 18:17:47 localhost dhclient: DHCPREQUEST on ath0 to 255.255.255.255 port 67 Mar 22 18:17:55 localhost dhclient: DHCPREQUEST on ath0 to 255.255.255.255 port 67 Mar 22 18:17:55 localhost dhclient: DHCPACK from 192.168.0.3 Mar 22 18:17:55 localhost NET: /sbin/dhclient-script : updated /etc/resolv.conf Mar 22 18:17:55 localhost dhclient: bound to 192.168.0.112 -- renewal in 38651 seconds.

Notice the ACK from my DHCP server (my server is 192.168.0.3). I don't ever see you getting an acknowledgement which leads me to believe that your interface might not actually be coming up. I would suggest you first try and configure your interface with a static address, gateway, and DNS settings. Then try and ping other local IP addresses. If you can't then it's an interface issue and not a DHCP issue. This is probably driver related. Do you see any driver errors in your "dmesg" command output?

Of course my log messages are from FC3. I don't have an FC2 installation around anymore, but things should be roughly the same.

by **dishawjp** » Wed Mar 23, 2005 7:17 pm

Voidmain,

I think I'm getting closer, thanks to you. I should have checked dmesg a long time ago.

It does appear to be loading the drivers OK:

Code: Select all: Linux Tulip driver version 1.1.13 (May 11, 2002) tulip0: MII transceiver #1 config 3000 status 7829 advertising 01e1. divert: allocating divert_blk for eth0 eth0: Lite-On 82c168 PNIC rev 32 at 0x2284cf00, 02:00:06:E3:3E:9C, IRQ 11. divert: freeing divert_blk for eth0 ip_tables: (C) 2000-2002 Netfilter core team ip_conntrack version 2.1 (4095 buckets, 32760 max) - 296 bytes per conntrack Linux Tulip driver version 1.1.13 (May 11, 2002) tulip0: MII transceiver #1 config 1000 status 782d advertising 01e1. divert: allocating divert_blk for eth0 eth0: Lite-On 82c168 PNIC rev 32 at 0x2290ef00, 02:00:06:E3:3E:9C, IRQ 11.

But after bringing eth0 up, I get this in dmesg:

Code: Select all: eth0: Setting full-duplex based on MII#1 link partner capability of 45e1. eth0: no IPv6 routers present

My guess would be that this, the "no IPv6 routers present" which is killing me. I did google this for a few minutes, but didn't get a clear understanding of what it even means, much less a solution.

Any thoughts? Thanks again for all your help.

Jim

by **Void Main** » Wed Mar 23, 2005 7:32 pm

The ipv6 message should not be a problem. You can turn off ipv6 by putting this in your interface config:

IPV6INIT=no

But it won't hurt anything by being enabled. Did you try setting a static address? Do you have another local machine that you can try and ping? Are you running a different kernel now than before when it was working?

by **dishawjp** » Thu Mar 24, 2005 6:41 am

Hi Voidmain,

Do you have another local machine that you can try and ping?

No. My wife is having my "computer room" redone. I'm lucky to be able to find anything right now. Right now the only machine I have set up is this one, and that's in one of my daughters' bedrooms. When she gets back from college this afternoon, I don't know what I'll do. Most of my Linux books, disks and stuff are piled on her bed. Anyway, I don't know if this would use eth0, but I was able to ping 127.0.0.1. But that may just use lo and have nothing to do with eth0.

Are you running a different kernel now than before when it was working?

No, it's running the same kernel. Actually, since it can't connect I haven't been able to update anything. The only changes since it last worked are:
1) Changed fstab to remove a reference to a swap partition on hdc which is no longer there. I removed that partition when I reinstalled FC3 and both now use the swap partition on hdd. FC3 is installed on hdc, and FC2 is installed on hdd.
2) I had a Linksys wireless router which died. Until I get around to replacing that, I am connecting without a router.
3) I did have to manually edit grub after reinstalling FC3 to get FC2 to boot, but grub is located on hdc.
4) I have 3 different FC3 kernels installed, but FC2 still has the original kernel installed. I did do an update of the FC3 kernel between the time that FC2 was working and now, but no changes to the FC2 kernel.

That's all I can think of. It has to be something fairly basic and be something to do with dhcpcd or something similar since it is affecting both ppp0 and eth0. My modem is an external modem (USR) and I can hear it connect, see all the right lights flash and it is talking with my ISP, it just fails to get to a DNS server.

Did you try setting a static address?

Not quite sure how to do this. I do have valid DNS server addresses in my /etc/resolv.conf, if that's what you mean.

Thanks again for all help on this.

Jim

by **Void Main** » Thu Mar 24, 2005 6:48 am

dishawjp wrote:Hi Voidmain,

Do you have another local machine that you can try and ping?

No. My is having my "computer room" redone. Right now the only machine I have set up is this one, and that's in one of my daughters' bedrooms. I don't know if this would use eth0, but I was able to ping 127.0.0.1. But that may just use lo and have nothing to do with eth0.

Nope, nothing to do with eth0.

Are you running a different kernel now than before when it was working?

2) I had a Linksys wireless router which died. Until I get around to replacing that, I am connecting without a router.

Aha! You were probably getting DHCP from your wireless router before and now it's gone.... On both your FC2 and FC3 look at your /var/lib/dhcp/dhclient-*.leases file(s). Your interface name will be part of the file name. This contains your DHCP client lease info. What is contained in those files?

by **dishawjp** » Thu Mar 24, 2005 1:13 pm

On both your FC2 and FC3 look at your /var/lib/dhcp/dhclient-*.leases file(s). Your interface name will be part of the file name. This contains your DHCP client lease info. What is contained in those files?

For FC3, it's:

Code: Select all: [dishawjp@eunix dhcp]$ cat dhclient-eth0.leases lease { interface "eth0"; fixed-address 24.58.88.19; option subnet-mask 255.255.248.0; option dhcp-lease-time 86400; option routers 24.58.88.1; option dhcp-message-type 5; option dhcp-server-identifier 10.100.32.1; option domain-name-servers 24.92.226.176,24.92.226.52,24.92.226.40,24.92.226.97; option broadcast-address 255.255.255.255; option domain-name "twcny.rr.com"; renew 4 2005/3/24 11:53:30; rebind 4 2005/3/24 22:00:49; expire 5 2005/3/25 01:00:49; } lease { interface "eth0"; fixed-address 24.58.88.19; option subnet-mask 255.255.248.0; option routers 24.58.88.1; option dhcp-lease-time 68113; option dhcp-message-type 5; option domain-name-servers 24.92.226.176,24.92.226.52,24.92.226.40,24.92.226.97; option dhcp-server-identifier 10.100.32.1; option broadcast-address 255.255.255.255; option domain-name "twcny.rr.com"; renew 5 2005/3/25 02:26:55; rebind 5 2005/3/25 11:38:55; expire 5 2005/3/25 14:00:50; } [dishawjp@eunix dhcp]$

for FC2, it's:

Code: Select all: [root@eunix dhcp]# pwd /mnt/var/lib/dhcp [root@eunix dhcp]# cat dhclient-eth0.leases lease { interface "eth0"; fixed-address 192.168.1.101; option subnet-mask 255.255.255.0; option routers 192.168.1.1; option dhcp-lease-time 86400; option dhcp-message-type 5; option domain-name-servers 24.92.226.52,24.92.226.48,24.92.226.176; option dhcp-server-identifier 192.168.1.1; option domain-name "twcny.rr.com"; renew 5 2004/12/24 12:49:47; rebind 5 2004/12/24 22:08:55; expire 6 2004/12/25 01:08:55; } [root@eunix dhcp]#

That looks to be the problem. If I change the "fixed-address" line and/or the "option routers" line, will that take care of the problem, or should I just copy the entire file from FC3 to FC2?

Thanks!

Jim

by **Void Main** » Thu Mar 24, 2005 5:50 pm

No, copying the file won't help because this is autogenerated by dhclient but I can tell a lot from the two files. I can tell that you used to be getting your DHCP configuration from your wireless router and now that it's gone you have to get your DHCP configuration from your provider. Your FC3 side seems to be doing this but your FC2 side hasn't actually renewed it's IP address since Christmas Eve. Is December 24th really the last last time you were able to use it? If you have used FC2 since that date then it appears you already had the problem. I seem to recall helping someone else with a similar DHCP issue and I believe it was on FC2. Let me search my forums and see what I can find, sounds too familiar. If I recall that one was some sort of incompatibility between the providers DHCP server and dhclient.

EDIT: I can't find it, must have been on another forum. At any rate, you could try configuring your machine statically with the information from your FC3 dhcpd.leases file. And you are sure you aren't running iptables with a configuration that could be blocking DHCP requests?

# service iptables stop
# ifup eth0

by **dishawjp** » Fri Mar 25, 2005 2:51 pm

This just keeps getting weirder...

Your FC3 side seems to be doing this but your FC2 side hasn't actually renewed it's IP address since Christmas Eve. Is December 24th really the last last time you were able to use it? If you have used FC2 since that date then it appears you already had the problem

That's probably about right. I had some time then, reinstalled FC3 and managed to hose my grub while at it. Since the reinstalled FC3 was working fine and I was just able to mount the FC2 partitions to get any files I needed from there, I wasn't really too worried about it. Then last week, I had some spare time, repaired grub and found that FC2 would not connect. It's really not too important to get FC2 fixed, but I have learned quite a bit from this...ummm... exercise... and want to see it through if I can.

No, copying the file won't help because this is autogenerated by dhclient but I can tell a lot from the two files.

Not that I doubted you, but I did try copying the contents of my FC3 /var/lib/dhcp/dhclient-eth0.leases, and of course it didn't change a thing.

And you are sure you aren't running iptables with a configuration that could be blocking DHCP requests?

You had suggested this earlier and I tried it then because it made a lot of sense, but for the record...

Code: Select all: [root@eunix root]# service iptables stop Flushing firewall rules: [ OK ] Setting chains to policy ACCEPT: filter [ OK ] Unloading iptables modules: [ OK ] [root@eunix root]# ifup eth0 Determining IP information for eth0... failed. [root@eunix root]#

I seem to recall helping someone else with a similar DHCP issue and I believe it was on FC2. Let me search my forums and see what I can find, sounds too familiar. If I recall that one was some sort of incompatibility between the providers DHCP server and dhclient.

It could have been a total coincidence that I stopped using FC2 at the same time I did an update to it and put a bad version on dhcp client on it. Would it make any sense to get a current version of the dhcpcd .rpm file for FC3 and install it on FC2? I haven't checked versions yet, but if that's what's knocking me out, it might be worth a try. Or would I be better off downloading and installing a current dhclient .rpm for FC2?

Once again thanks for all your help!

Jim

by **dishawjp** » Fri Mar 25, 2005 7:55 pm

I did a little more poking around, and perhaps some of the following will be helpful.

First I brought up eth0 as usual:

Code: Select all: [dishawjp@eunix dishawjp]$ net Determining IP information for eth0... failed.

Then looked to see if eth0 had actually come up:

Code: Select all: [dishawjp@eunix dishawjp]$ /sbin/ifconfig eth0 Link encap:Ethernet HWaddr 02:00:06:E3:3E:9C inet6 addr: fe80::6ff:fee3:3e9c/64 Scope:Link UP BROADCAST RUNNING MULTICAST MTU:1500 Metric:1 RX packets:15159 errors:2 dropped:0 overruns:0 frame:0 TX packets:26 errors:3 dropped:0 overruns:0 carrier:3 collisions:0 txqueuelen:1000 RX bytes:911351 (889.9 Kb) TX bytes:7016 (6.8 Kb) Interrupt:11 Base address:0xef00 lo Link encap:Local Loopback inet6 addr: ::1/128 Scope:Host UP LOOPBACK RUNNING MTU:16436 Metric:1 RX packets:14 errors:0 dropped:0 overruns:0 frame:0 TX packets:14 errors:0 dropped:0 overruns:0 carrier:0 collisions:0 txqueuelen:0 RX bytes:2528 (2.4 Kb) TX bytes:2528 (2.4 Kb)

So it looked like it was up, just no DNS server contacted. Then I switched to root and ran dhclient:

Code: Select all: [dishawjp@eunix dishawjp]$ su - Password: [root@eunix root]# dhclient Internet Systems Consortium DHCP Client V3.0.1rc14 Copyright 2004 Internet Systems Consortium. All rights reserved. For info, please visit http://www.isc.org/products/DHCP sit0: unknown hardware address type 776 /sbin/dhclient-script: configuration for sit0 not found. sit0: unknown hardware address type 776 Listening on LPF/sit0/ Sending on LPF/sit0/ Listening on LPF/eth0/02:00:06:e3:3e:9c Sending on LPF/eth0/02:00:06:e3:3e:9c Listening on LPF/lo/ Sending on LPF/lo/ Sending on Socket/fallback receive_packet failed on sit0: Network is down DHCPDISCOVER on lo to 255.255.255.255 port 67 interval 3 DHCPDISCOVER on sit0 to 255.255.255.255 port 67 interval 6 send_packet: Network is down DHCPDISCOVER on eth0 to 255.255.255.255 port 67 interval 3 DHCPDISCOVER on lo to 255.255.255.255 port 67 interval 4 DHCPDISCOVER on eth0 to 255.255.255.255 port 67 interval 5 DHCPDISCOVER on sit0 to 255.255.255.255 port 67 interval 16 send_packet: Network is down DHCPDISCOVER on lo to 255.255.255.255 port 67 interval 8 DHCPDISCOVER on eth0 to 255.255.255.255 port 67 interval 11 DHCPDISCOVER on lo to 255.255.255.255 port 67 interval 11 DHCPDISCOVER on eth0 to 255.255.255.255 port 67 interval 11 DHCPDISCOVER on sit0 to 255.255.255.255 port 67 interval 21 send_packet: Network is down DHCPDISCOVER on lo to 255.255.255.255 port 67 interval 21

I killed this with a <ctrl>-C after seeing the above. Does this point to anything helpful? I googled sit0 and found that it stands for "simple internet transition" or something like that. And that it has something to do with IPv4 and IPv6 stuff, but I don't really understand any of that. Then I went through my /sbin/dhclient-script, but was unable to find the pattern "sit0" in there. Would the "sit0 unknown hardware address" complaint be referring to something having to do with the router which is no longer there? If so, what should I change?

Does any of this shed any more light on the problem? ANyway, thanks for listening and your help so far.

Jim

by **Void Main** » Sat Mar 26, 2005 9:05 am

Put IPV6INIT=no in your eth0 interface config file and do that same thing again. I see over 20,000 recieve packets and only 26 transmit packets. It would appear you are getting traffic from somewhere, "ethereal" would tell you that. Again, I would take all the settings from your dhcp leases file and set them statically (IP address, netmask, and gateway). Then I would try and ping and traceroute some IP addresses. You are getting link lights on your NIC and on your switch port right? Does "ethtool eth0" give you any more information about your link status?

Void's Forums

mDNSResponder/dhclient

mDNSResponder/dhclient

Who is online