Index: [Article Count Order] [Thread]

Date: Tue, 1 Apr 2008 12:35:12 +0900
From: Mitsuru Chinen <mitch@linux.vnet.ibm.com>
Subject: [dhcptest:00266] Re: Tests for Maximum Retransmission Time
To: dhcptest@tahi.org
Cc: Hiroshi MIYATA <miyata@tahi.org>
Message-Id: <20080401123512.f237c117.mitch@linux.vnet.ibm.com>
In-Reply-To: <DC2C6061-922A-4004-B0B4-ED0D784E1148@tahi.org>
References: <20080310205322.9ab32c78.mitch@linux.vnet.ibm.com>	<20080311203232.6670d3f5.mitch@linux.vnet.ibm.com>	<D49285CA-D185-4232-8E09-195F5781920A@tahi.org>	<DC2C6061-922A-4004-B0B4-ED0D784E1148@tahi.org>
X-Mail-Count: 00266

Miyata-san,

On Tue, 1 Apr 2008 11:31:05 +0900
Hiroshi MIYATA <miyata@tahi.org> wrote:

> Chinen-san,
> 
> Could you tell me which version of test tool are you using?

All of my patches are for the latest one, DHCPv6 Self Test version 1.0.8.

Best Regards,
----
Mitsuru Chinen <mitch@linux.vnet.ibm.com>

> Thanks,
> 
> ....miyata
> 
> On 2008/03/12, at 0:45, Hiroshi MIYATA wrote:
> 
> > Chinen-san,
> >
> > Thanks for your update.
> > I will examine it.
> >
> > Sorry for my late response.
> >
> > ...miyata
> >
> >
> > On 2008/03/11, at 20:32, Mitsuru Chinen wrote:
> >
> >> Hi there,
> >>
> >> On Mon, 10 Mar 2008 20:53:22 +0900
> >> Mitsuru Chinen <mitch@linux.vnet.ibm.com> wrote:
> >>
> >>> I think the following tests have a bug.
> >>>
> >>> rfc3315/
> >>>  #32 Part D : Maximum Retransmission Time of solicit message
> >>>  #37 Part C : Maximum Retransmission Time of Request messages
> >>>  #41 Part C : Maximum Retransmission Time of Request messages
> >>>  #46 Part C : Maximum Retransmission Time of Renew message, T1 and  
> >>> T2 non-zero
> >>>  #50 Part C : Maximum Retransmission Time of Rebind message, T1  
> >>> and T2 non-zero
> >>> rfc3736/
> >>>  #13 Part C : Maximum Retransmission Time of Information-request  
> >>> message
> >>>
> >>> Regarding RT, Section 14 in RFC3315 says:
> >>>
> >>>>  RT for each subsequent message transmission is based on the  
> >>>> previous
> >>>>  value of RT:
> >>>>
> >>>>     RT = 2*RTprev + RAND*RTprev
> >>>>
> >>>>  MRT specifies an upper bound on the value of RT (disregarding the
> >>>>  randomization added by the use of RAND).  If MRT has a value of 0,
> >>>>  there is no upper limit on the value of RT.  Otherwise:
> >>>>
> >>>>     if (RT > MRT)
> >>>>        RT = MRT + RAND*MRT
> >>>
> >>> However, the test doesn't recalculate $timeout, expected RT, when
> >>> timeout is bigger than MRT.
> >>
> >> Excuse me, I sent wrong patch in previous mail.
> >> Here's the correct one.
> >>
> >> Best Regards,
> >> ----
> >> Mitsuru Chinen <mitch@linux.vnet.ibm.com>
> >>
> >>
> >> diff --git a/rfc3315/C_RFC3315_17_1_2_SolRetransMRT.seq b/rfc3315/ 
> >> C_RFC3315_17_1_2_SolRetransMRT.seq
> >> index 26183e7..658aa4c 100755
> >> --- a/rfc3315/C_RFC3315_17_1_2_SolRetransMRT.seq
> >> +++ b/rfc3315/C_RFC3315_17_1_2_SolRetransMRT.seq
> >> @@ -126,6 +126,11 @@ while($mrtflag < $MAXCOUNT){
> >> 	}
> >> 	$initflag=1;
> >>
> >> +	# If $timeout is bigger than MRT, expected RT is MRT + RAND * MRT
> >> +	if($timeout > $MRT){
> >> +		$timeout = (1 + $RAND) * $MRT;
> >> +	}
> >> +
> >> #
> >> #----------- Wait for next retransmission
> >> #
> >> diff --git a/rfc3315/C_RFC3315_18_1_1_ReqRetransMRT.seq b/rfc3315/ 
> >> C_RFC3315_18_1_1_ReqRetransMRT.seq
> >> index a0d25ce..41df69f 100755
> >> --- a/rfc3315/C_RFC3315_18_1_1_ReqRetransMRT.seq
> >> +++ b/rfc3315/C_RFC3315_18_1_1_ReqRetransMRT.seq
> >> @@ -141,6 +141,11 @@ while($mrtflag < $MAXCOUNT){
> >> 	}
> >> 	$initflag=1;
> >>
> >> +	# If $timeout is bigger than MRT, expected RT is MRT + RAND * MRT
> >> +	if($timeout > $MRT){
> >> +		$timeout = (1 + $RAND) * $MRT;
> >> +	}
> >> +
> >> #
> >> #----------- Wait for next retransmission
> >> #
> >> diff --git a/rfc3315/C_RFC3315_18_1_2_CnfRetransMRT.seq b/rfc3315/ 
> >> C_RFC3315_18_1_2_CnfRetransMRT.seq
> >> index 279248c..4d449d4 100755
> >> --- a/rfc3315/C_RFC3315_18_1_2_CnfRetransMRT.seq
> >> +++ b/rfc3315/C_RFC3315_18_1_2_CnfRetransMRT.seq
> >> @@ -193,6 +193,11 @@ while($mrtflag < $MAXCOUNT){
> >> 	}
> >> 	$initflag=1;
> >>
> >> +	# If $timeout is bigger than MRT, expected RT is MRT + RAND * MRT
> >> +	if($timeout > $MRT){
> >> +		$timeout = (1 + $RAND) * $MRT;
> >> +	}
> >> +
> >> #
> >> #----------- Wait for next retransmission
> >> #
> >> diff --git a/rfc3315/C_RFC3315_18_1_3_RenRetransMRT.seq b/rfc3315/ 
> >> C_RFC3315_18_1_3_RenRetransMRT.seq
> >> index fcced3a..dd56be9 100755
> >> --- a/rfc3315/C_RFC3315_18_1_3_RenRetransMRT.seq
> >> +++ b/rfc3315/C_RFC3315_18_1_3_RenRetransMRT.seq
> >> @@ -186,6 +186,11 @@ while($mrtflag < $MAXCOUNT){
> >> 	}
> >> 	$initflag=1;
> >>
> >> +	# If $timeout is bigger than MRT, expected RT is MRT + RAND * MRT
> >> +	if($timeout > $MRT){
> >> +		$timeout = (1 + $RAND) * $MRT;
> >> +	}
> >> +
> >> #
> >> #----------- Wait for next retransmission
> >> #
> >> diff --git a/rfc3315/C_RFC3315_18_1_4_RebRetransMRT.seq b/rfc3315/ 
> >> C_RFC3315_18_1_4_RebRetransMRT.seq
> >> index 617351c..0947e4c 100755
> >> --- a/rfc3315/C_RFC3315_18_1_4_RebRetransMRT.seq
> >> +++ b/rfc3315/C_RFC3315_18_1_4_RebRetransMRT.seq
> >> @@ -183,6 +183,11 @@ while($mrtflag < $MAXCOUNT){
> >> 	}
> >> 	$initflag = 1;
> >>
> >> +	# If $timeout is bigger than MRT, expected RT is MRT + RAND * MRT
> >> +	if($timeout > $MRT){
> >> +		$timeout = (1 + $RAND) * $MRT;
> >> +	}
> >> +
> >> #
> >> #----------- Wait for next retransmission
> >> #
> >> diff --git a/rfc3736/C_RFC3736_INfoRetransMRT.seq b/rfc3736/ 
> >> C_RFC3736_INfoRetransMRT.seq
> >> index c532a4f..0f9cd4c 100755
> >> --- a/rfc3736/C_RFC3736_INfoRetransMRT.seq
> >> +++ b/rfc3736/C_RFC3736_INfoRetransMRT.seq
> >> @@ -106,6 +106,12 @@ while(check_equal($interval,$INF_MAX_RT,$RAND)){
> >> 	}else{
> >> 		$timeout = (1 + $RAND) * $IRT;
> >> 	}
> >> +
> >> +	# If $timeout is bigger than MRT, expected RT is MRT + RAND * MRT
> >> +	if($timeout > $INF_MAX_RT){
> >> +		$timeout = (1 + $RAND) * $INF_MAX_RT;
> >> +	}
> >> +
> >> 	($retinf2,%inf2) = wait_for_information_request($IF0, $timeout + 1);
> >> 	$nextime = getReceivedtime(\%inf2);
> >> 	if($retinf2 != 0){
> >>
> >
> >
> 
>