From: pimjager <pim@pimjager.nl>
Date: Sun, 15 Nov 2015 19:27:05 +0000 (+0100)
Subject: added discussion on not testing checksums
X-Git-Url: https://git.martlubbers.net/?a=commitdiff_plain;h=429e007fac682ed91353a951d41efaa1f23df07b;p=tt2015.git

added discussion on not testing checksums
---

diff --git a/a2/1cases.tex b/a2/1cases.tex
index 85064bc..d30a937 100644
--- a/a2/1cases.tex
+++ b/a2/1cases.tex
@@ -73,7 +73,7 @@ user permissions.
 	\caption{Preflight checklist\label{tbl:preflight}}
 \end{longtable}
 
-\subsection{Testing of SUT}
+\subsection{Testing of SUT} \label{sec:testingofsut}
 The SUT is a series of services for other computer programs with no end-user
 facing interface. Therefore the SUT will be tested solely by calling it's
 services through various automated scripts. An automated test suite will be
diff --git a/a2/2results.tex b/a2/2results.tex
index 4e685ed..eaa1731 100644
--- a/a2/2results.tex
+++ b/a2/2results.tex
@@ -66,36 +66,22 @@ Table~\ref{tbl:preflightresults} and Table~\ref{tbl:testcasesresults} this is
 the case. Therefore the SUT has PASSED/FAILED this test.
 %This failure is due to the failure of the test case(s) 1,2,3,n as specified
 %where the SUT failed to behave in the expected way by... %behavior dat SUT wel
-%%liet zien
+%%liet zien 
 
-%
-%\textbf{TODO: DEZE TABEL WEL UPDATEN!!}
-%
-%\begin{table}
-%\begin{tabularx}{\linewidth}{| l | X|}
-%	\hline
-%Check 1 & Get the SUT in a workable state. \\\hline
-%Course of action & \begin{enumerate}
-%	\item Import the VirtualBox image into VirtualBox.
-%	\item Boot the vm.
-%	\item Verify the SUT booted succesfully and the network module(s) are loaded.
-%\end{enumerate} \\\hline
-%Passed & \textit{\st{Yes}/No} \\\hline \hline
-%Check 2 & Verify the SUT is complete. \\\hline
-%Course of action & \begin{enumerate}
-%	\item Use the course of action for \emph{Check 1} in order to boot the SUT.
-%	\item Verify the loopback device exists using \emph{ifconfig}
-%	\item Verify the \emph{echo-server} is present on the system.
-%	\item Verify \emph{Scrappy} is present on the system.
-%	\item Verify all scripts used for testing are present on the system.
-%\end{enumerate} \\\hline
-%Passed & \textit{\st{Yes}/No} \\\hline \hline
-%Check 3 & Initialize the testing environment. \\\hline
-%Course of action & \begin{enumerate}
-%	\item Boot the SUT (see \emph{Check 1}).
-%	\item Setup iptables by executing \emph{./scripts/setup/iptables.sh}
-%	\item Setup ... by executing \emph{./scripts/setup/...}
-%\end{enumerate} \\\hline
-%Passed & \textit{\st{Yes}/No} \\\hline
-%\end{tabularx}
-%\end{table}
+\subsection{Notes on results}
+Due to severe difficulties with implementing the corruption of a checksum any
+tests which mention corruption of the checksum, that is, all tests testing
+the different partitions 4a, 4b, 4c, 5a and 5b as mentioned in 
+Section~\ref{sec:testingofsut}, are always done with a valid checksum. These
+tests always fall in partition 4a and 5a. 
+
+These difficulties arise because TCP-checksum computation is something which is
+done
+on the hardware level or somewhere otherwise quite far from the software
+implementation of anything calling the TCP-stack. Due to this, changing the
+checksum of the segments send from the test script to the SUT is not possible
+without writing raw data to the network sockets, which would result
+in high changes of manual errors.
+
+Because of this these partitions are not tested and the SUT is not tested for
+the requirement that it should disregard segments with invalid checksums. 
\ No newline at end of file