[Swan-dev] tests failing due to retransmissions

[Antony Antony]

On Wed, Sep 27, 2017 at 12:38:08PM -0400, D. Hugh Redelmeier wrote:
> I get a lot of them.  This isn't good or useful.
> 
> I ran the following script:
> 
> 	for i in testing/pluto/*/OUTPUT/*.diff ; do
> 		if [ ! -s "$i" ] ; then
> 			: no difference
> 		elif grep '^[-+][0-9]' "$i" | grep -v "retransmission" >/dev/null ; then
> 			: echo $i not just retransmission
> 		else
> 			echo $i just RETRANSMISSION
> 		fi
> 	done
> 
> Today I got 55 lines listing diff files containing only retransmission
> messages.  That means 55 extra files that I have to examine
> pointlessly.
> 
> What should be done?
> 
> Should we calm the hair-trigger on retransmission?

I haven't figured out pattern in "retransmission *" yet.

Another line that show up in several (20+ output files) IKEv1 tests as diff 
is:

"discarding duplicate packet; already STATE_MAIN_I3"

I wonder how useful is this line, in a typical user case. Why not change to 
a debug line?

ipsec auto --up con <conn> output should be minimal. Is this line really 
necessary here?

If we remove this line noise in ikev1 could be a bit less.  A similar line 
in ikev2, is is in state.c  I see it 6-10 tests.

"discarding packet received during asynchronous work" a

-antony