Bug 29349

Summary: Incorrect ICMP ECHO source address on reply
Product: [Retired] Red Hat Linux Reporter: John Bass <jbass>
Component: kernelAssignee: Michael K. Johnson <johnsonm>
Status: CLOSED NOTABUG QA Contact: Brock Organ <borgan>
Severity: medium Docs Contact:
Priority: medium    
Version: 7.0   
Target Milestone: ---   
Target Release: ---   
Hardware: i386   
OS: Linux   
Whiteboard:
Fixed In Version: Doc Type: Bug Fix
Doc Text:
Story Points: ---
Clone Of: Environment:
Last Closed: 2001-02-25 09:36:44 UTC Type: ---
Regression: --- Mount Type: ---
Documentation: --- CRM:
Verified Versions: Category: ---
oVirt Team: --- RHEL 7.3 requirements from Atomic Host:
Cloudforms Team: --- Target Upstream Version:
Embargoed:

Description John Bass 2001-02-25 09:36:41 UTC 
01:15:58.747749 < Client1.domain.test > Server.domain.test: icmp: echo
request (DF) 
01:15:58.747808 > Server.domain.test > Client1.domain.test: icmp: echo
reply
01:16:13.587872 < Client1.domain.test > Server.domain.test: icmp: echo
request (DF) 
01:16:13.587936 > Server.domain.test > Client1.domain.test: icmp: echo
reply
01:16:16.598712 < Client1.domain.test > Client2.domain.test: icmp: echo
request (DF)
01:16:16.599321 < Client1.domain.test > Client2.domain.test: icmp: echo
request
01:16:16.631662 > Client2.domain.test > Client1.domain.test: icmp: echo
reply
01:16:16.663876 > Client2.domain.test > Client1.domain.test: icmp: echo
reply
01:16:18.579882 > arp who-has Client1.domain.test tell Server.domain.test
(x:xx:xx:xx:xx:xx)
01:16:18.585470 < arp reply Client1.domain.test is-at 0:40:96:30:d5:46
(x:xx:xx:xx:xx:xx)
01:16:22.677906 < Client1.domain.test > Server.domain.test: icmp: echo
request (DF)
01:16:22.677972 > Server.domain.test > Client1.domain.test: icmp: echo
reply

The echo request from Client1 were for some reason replied to with a source
address of another client (Client2) instead of the Server.

I just noticed this debugging the ntpd problem previously reported, so I'm
not sure if it's a kernel or tcpdump problem ... I suspect kernel since
tcp dump didn't have any reason to resolve Client2's IP or name otherwise.
Comment 1 Michael K. Johnson 2001-02-28 00:06:43 UTC 
I see matching requests, does not look like a bug to me.