Bug 592854

Summary:

Should not keep send release after got a reply message.

Product:

[Fedora] Fedora

Reporter:

Yang Ren <ryang>

Component:

dhcp

Assignee:

Jiri Popelka <jpopelka>

Status:

CLOSED NOTABUG

QA Contact:

Fedora Extras Quality Assurance <extras-qa>

Severity:

medium

Docs Contact:

Priority:

low

Version:

CC:

jpopelka, llim

Target Milestone:

---

Target Release:

---

Hardware:

All

OS:

Linux

Whiteboard:

Fixed In Version:

Doc Type:

Bug Fix

Doc Text:

Story Points:

---

Clone Of:

Environment:

Last Closed:

2010-05-18 09:04:15 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:

Attachments:

Description	Flags
Wireshark packet capture screenshot	none

Description Yang Ren 2010-05-17 08:06:25 UTC

Description of problem:
I got a failed in a tahi test.
After server send a reply for release message. Still got a release message.
I'm not sure it's retransmit problem. Just ask your opinion here.

http://10.66.65.86/DHCPv6_Self_Test_P2_1_1_0_client_fedora/rfc3315/58.html

First release message in 12:53:42
Got second release in 12:53:42
ElapsedTime is 109
After this release message test suite send reply to client in 12:53:43
And also in 12:53:43 got third release message it's elapsedtime is 334


Version-Release number of selected component (if applicable):
dhclient-4.1.1-16.fc12.x86_64

How reproducible:
Always

Steps to Reproduce:
1.Client release current address 
2.server send reply between the second and the third release message
  
Actual results:
Client keep send release message

Expected results:
Client stop send release message

Additional info:

Comment 1 Jiri Popelka 2010-05-17 10:04:24 UTC

Created attachment 414503 [details]
Wireshark packet capture screenshot

Hi Yang,
I went through the log, but don't know why the client sent the third message.

I was testing the situation here and everything looks correct.
I'm attaching the wireshark packet capture screenshot
from my test, where you can see:
1) In time 0.009438 client sends first Release message.
2) No Reply msg comes from server for the Release.
3) In time 1.007082 (approx. 1000ms after the first Release) client sends second Release msg.
4) In time 1.015104 comes Reply with NoBinding StatusCode.
5) No more Release messages are sent.

The only strange thing I see in log from your test are the values of ElapsedTime.
The value of ElapsedTime in the second Release should be around 1000(ms), but there's value 109.
The value of ElapsedTime in the third Release should be around 3000(ms), but there's value 334.

Comment 2 Yang Ren 2010-05-18 02:31:22 UTC

I read the verify point of this test. 

"If addresses are released but the Reply from a DHCP server is lost,
the client will retransmit the Release message, and the server may
respond with a Reply indicating a status of NoBinding.  Therefore,
the client does not treat a Reply message with a status of NoBinding
in a Release message exchange as if it indicates an error."


This reply message with status code 3 nobinding. Is it the problem?

Comment 3 Yang Ren 2010-05-18 02:32:34 UTC

Oh I see the screen shot it's also with status code 3. I'll try to find other fail reason.

Comment 4 Yang Ren 2010-05-18 03:27:19 UTC

I find the problem. The time is not correct.
The first release sent in 12:52:27
Second release sent in 12:52:28
Third release sent in 12:52:30

But Send DHCPv6 Reply Message: SERVER1 --> NUT at 12:53:43

So actually client send 6 release messages before got reply. 

So I think this is a test suite problem. Just close it.