After update, Red interface not getting dhcp ip address from ISP

Getting Started with IPFire Updates Trouble
1

Working device updated to 198 and now Red is not getting an IP. Downloaded and did a fresh install with the same problem (tried 198, 197 and 195, the problem persists). The device has 4 ports, I’ve swapped red and green ports and cables, the red interface never gets an IP.

If I plug red into a different local DHCP server it will get an IP.

Searching through posts someone mentioned forcing MTU to 1500, that did not work.

00:25:47 dhcpcd[23557] : dhcpcd-10.2.4 starting
00:25:47 dhcpcd[23560] : DUID 00:01:00:01:30:a5:85:5b:00:90:27:xx:xx:xx
00:25:47 dhcpcd[23560] : red0: IAID 27:xx:xx:xx
00:25:49 dhcpcd[23560] : red0: soliciting a DHCP lease
00:25:51 dhcpcd[23560] : red0: offered 192.168.1.106 from 192.168.1.4
00:25:51 dhcpcd[23560] : red0: probing address 192.168.1.106/24
00:25:56 dhcpcd[23560] : red0: leased 192.168.1.106 for 86400 seconds
00:25:56 dhcpcd[23560] : red0: adding route to 192.168.1.0/24
00:25:56 dhcpcd[23560] : red0: adding default route via 192.168.1.4
00:34:39 dhcpcd[23560] : red0: carrier lost
00:34:39 dhcpcd[23560] : red0: deleting route to 192.168.1.0/24
00:34:39 dhcpcd[23560] : red0: deleting default route via 192.168.1.4
00:34:58 dhcpcd[23560] : red0: carrier acquired
00:34:58 dhcpcd[23560] : red0: IAID 27:xx:xx:xx
00:34:58 dhcpcd[23560] : red0: rebinding lease of 192.168.1.106
00:35:03 dhcpcd[23560] : red0: DHCP lease expired
00:35:05 dhcpcd[23560] : red0: soliciting a DHCP lease
00:37:31 dhcpcd[25885] : sending signal ALRM to pid 23559
00:37:31 dhcpcd[25885] : waiting for pid 23559 to exit
00:37:31 dhcpcd[23560] : received SIGALRM, releasing
00:37:31 dhcpcd[23560] : red0: removing interface
00:37:33 dhcpcd[23560] : main: control_stop: No such file or directory
00:37:33 dhcpcd[23560] : dhcpcd exited
00:37:33 dhcpcd[26178] : dhcpcd-10.2.4 starting
00:37:33 dhcpcd[26181] : DUID 00:01:00:01:30:a5:85:5b:00:90:27:xx:xx:xx
00:37:34 dhcpcd[26181] : red0: waiting for carrier
00:37:36 dhcpcd[26181] : red0: carrier acquired
00:37:36 dhcpcd[26181] : red0: IAID 27:xx:xx:xx
00:37:38 dhcpcd[26181] : red0: soliciting a DHCP lease
00:38:34 dhcpcd[26181] : timed out
00:39:40 dhcpcd[26560] : sending signal ALRM to pid 26180
00:39:40 dhcpcd[26560] : waiting for pid 26180 to exit
00:39:40 dhcpcd[26181] : received SIGALRM, releasing
00:39:40 dhcpcd[26181] : red0: removing interface
00:39:42 dhcpcd[26181] : main: control_stop: No such file or directory
00:39:42 dhcpcd[26181] : dhcpcd exited

00:37:33 kernel: 8021q: adding VLAN 0 to HW filter on device red0
00:37:36 kernel: e1000e 0000:01:00.0 red0: NIC Link is Up 1000 Mbps Full Duplex, Flow Control: Rx/Tx
00:39:42 kernel: e1000e 0000:01:00.0 red0: NIC Link is Down
00:39:42 kernel: 8021q: adding VLAN 0 to HW filter on device red0
00:39:45 kernel: e1000e 0000:01:00.0 red0: NIC Link is Up 1000 Mbps Full Duplex, Flow Control: Rx/Tx

2

Below is a link to a similar thread.

Maybe you will find some hints.

Regards

3

I saw the thread about e1000e unit hang, but I’m not getting that error in the logs. Also replaced the cable. The red interface can get an IP address from a local DHCP server, and has no problems running for several days.

4

Hall @red_chaos

Welcome to the IPFire community

This shows that the dhcp communication between the IPFire dhcpcd program and your ISP was working normally at 00:25

A dhcp lease was requested, one was offered and it was accepted.

Then at 00:34

the carrier signal from your ISP was lost for 19 seconds. Then it was acquired again and the existing lease rebinded.

Then at 00:35 that lease expired and a new one was requested, although normally I would expect it to just be renewed by your ISP.

Then 2.5 minutes later the dhcpcd program was stopped. Either you rebooted at that time or manually restarted dhcpcd.

Here dhcpcd is restarting, the carrier was acquired and a dhcp lease requested but this time your ISP did not respond with anything and so after 60 seconds the dhcpcd program, timed out.

Then a minute later

the dhcpcd program is stopped again.

Can you confirm that these are from you re-booting the system or manually restarting dhcpcd from the console command line?

The dhcpcd communication seems to be occurring correctly but there seems to have been some issue with the connection to your ISP as they are not responding back with the offer of a lease.

Either your ISP has a line problem in their connection to you, in which case that should work again after some time, or they have modified their dhcp server software to not be RFC compliant.

Worth trying would be to disable the Rapid Commit option as described in the documentation here:-

https://www.ipfire.org/docs/installation/step5#setting-red-interface-address

If that then works then your ISP has updated their dhcp server software to not properly respond to the allowed option of rapid commit. If not supported the RFC defines that it should just be ignored but there do seem to be quite a few ISP’s who have software that just stops responding if they Rapid Commit traffic is sent.

If with that option disabled you still have a problem getting an IP leased then there would seem to be a problem at your ISP.

3 Likes
5

yes I was restarting dhcpd here. I disabled Rapid Commit and it still failed.

I’ll contact my ISP. Thank you for your time looking into this.