Bug 53700

Summary: pump doesnt work with RoadRunner, violates RFC
Product: [Retired] Red Hat Linux Reporter: Mario Lorenz <ml>
Component: pumpAssignee: Elliot Lee <sopwith>
Severity: high Docs Contact:
Priority: medium    
Version: 7.1CC: mello, olivier.baudron
Target Milestone: ---   
Target Release: ---   
Hardware: All   
OS: Linux   
Fixed In Version: Doc Type: Bug Fix
Doc Text:
Story Points: ---
Clone Of: Environment:
Last Closed: 2001-09-17 10:13:21 UTC Type: ---
Regression: --- Mount Type: ---
Documentation: --- CRM:
Verified Versions: Category: ---
oVirt Team: --- RHEL 7.3 requirements from Atomic Host:
Cloudforms Team: ---

Description Mario Lorenz 2001-09-15 18:32:17 UTC
From Bugzilla Helper:
User-Agent: Mozilla/4.77 [de] (X11; U; Linux 2.4.5-SGI_XFS_1.0.1 i686)

Description of problem:
This is basically a "reopen" of #52436
The problem is that pump is violating the DHCP RFC by not sending
DHCPDISCOVER (and probably other packets as well) with a source
IP of (as required by the RFC) but with the IP of other interfaces
initialized earlier (in my case it used an 192.168.1.x IP from the internal
RR seems to have implemented IP source filters or IP source verification so
192.168.x sourced IP's will not be accepted (hence not 
given to the DHCP server).
Using dhcpcd, things work fine as dhcpcd properly uses as
source IP until it has given its IP from the DHCP server.
I have tried all versions of pump up to Rawhide (0.8.11-7)

Version-Release number of selected component (if applicable):

How reproducible:

Steps to Reproduce:
1. Get computer with two NIC's
2. ifconfig eth0  (for example)
3. ifconfig eth1 up           (so you can do step 4)
4. tcpdump -eni eth1           (to watch. Better use tethereal if
5. /sbin/pump -i eth1
6 (optional) /sbin/dhcpcd eth1  (to compare...)

Actual Results:  tcpdump shows DHCP packets on eth1 (DHCPDISCOVER). They
bear a source
IP of

Expected Results:  DHCP RFC mandates a source IP of

Additional info:

Comment 1 Olivier Baudron 2001-09-17 10:13:17 UTC
See also bug 23052 for further infos (same bug).

Comment 2 Mario Lorenz 2001-09-17 18:49:46 UTC
Indeed, its the same bug as 23052. Probably 52436 should now also be marked
duplicate to that bug. 

*** This bug has been marked as a duplicate of 23052 ***