[Reproducible-commits] [presentations] 01/01: more WIP
Holger Levsen
holger at moszumanska.debian.org
Sat Jan 30 23:18:32 UTC 2016
This is an automated email from the git hooks/post-receive script.
holger pushed a commit to branch master
in repository presentations.
commit c9ccb0709162850d99f66dfdb3042a5c77da3ec4
Author: Holger Levsen <holger at layer-acht.org>
Date: Sun Jan 31 00:18:25 2016 +0100
more WIP
---
.../2016-01-31-FOSDEM16-Reproducible-ecosystem.tex | 81 ++++++++++++----------
2016-01-31-FOSDEM16/notes | 16 +++++
2 files changed, 62 insertions(+), 35 deletions(-)
diff --git a/2016-01-31-FOSDEM16/2016-01-31-FOSDEM16-Reproducible-ecosystem.tex b/2016-01-31-FOSDEM16/2016-01-31-FOSDEM16-Reproducible-ecosystem.tex
index 55be74b..f16b54d 100644
--- a/2016-01-31-FOSDEM16/2016-01-31-FOSDEM16-Reproducible-ecosystem.tex
+++ b/2016-01-31-FOSDEM16/2016-01-31-FOSDEM16-Reproducible-ecosystem.tex
@@ -346,6 +346,18 @@ and some hints where this might be going…}
\end{center}
\end{frame}
+ \begin{itemize}
+ \item \texttt{https://reproducible-builds.org/docs}
+ \item Lunar's talk from CCCamp 2015 also on
+ \texttt{https://media.ccc.de}
+ \begin{tikzpicture}[remember picture]
+ \node[shift={(-1.05\paperwidth, -0.3\paperheight)},at=(current page.south east)] {
+ \includegraphics[width=0.83\textwidth]{images/cccamp2015_lunar_random.png}
+ };
+ \end{tikzpicture}
+ \end{itemize}
+\end{frame}
+
\section{Common ressources}
\begin{frame}
@@ -373,17 +385,6 @@ and some hints where this might be going…}
\begin{frame}
\frametitle{Documentation about common problems}
- \begin{itemize}
- \item \texttt{https://reproducible-builds.org/docs}
- \item Lunar's talk from CCCamp 2015 also on
- \texttt{https://media.ccc.de}
- \begin{tikzpicture}[remember picture]
- \node[shift={(-1.05\paperwidth, -0.3\paperheight)},at=(current page.south east)] {
- \includegraphics[width=0.83\textwidth]{images/cccamp2015_lunar_random.png}
- };
- \end{tikzpicture}
- \end{itemize}
-\end{frame}
\begin{frame}
\frametitle{\texttt{SOURCE\_DATE\_EPOCH}}
@@ -454,7 +455,7 @@ and some hints where this might be going…}
\item Continuously testing Debian \texttt{testing}, \texttt{unstable} and
\texttt{experimental}
\item Also testing: coreboot, OpenWrt, NetBSD, FreeBSD,
- Arch Linux, Fedora and soon FDroid and Guix too
+ Arch Linux, Fedora and soon F-Droid and Guix too
\item<2> 230 jenkins jobs running on 22 hosts
\item<2> 42 scripts with a total of 4k lines of Python and 6k lines of Bash
Shell
@@ -850,7 +851,7 @@ Build-Environment:
\frametitle{Status FreeBSD}
\begin{itemize}
\item \texttt{https://tests.r-b.org/freebsd}
- \item base system not yet reproducible
+ \item base system not yet reproducible, but almost there
\item 63\% of 15k ports were reproducible in 2013 already, their wiki says
\item tests maintained by h01ger
\item recreating the build env: ?
@@ -878,22 +879,6 @@ Build-Environment:
\end{tikzpicture}
\end{frame}
-\begin{frame}
- \frametitle{Status Guix}
- \begin{itemize}
- \item I have little idea
- \item talk yesterday
- \item recreating the build env: by design
- \item user verification tool: yes! (Guix challange)
- \item<2> next: check your inbox, h01ger!
- \end{itemize}
- \begin{tikzpicture}[remember picture,overlay]
- \node[shift={(-0.15\paperwidth, 0.2\paperheight)},at=(current page.south east)] {
- \includegraphics[height=0.33\paperheight]{images/guix.png}
- };
- \end{tikzpicture}
-\end{frame}
-
\begin{frame}
\frametitle{Status Fedora}
@@ -902,8 +887,8 @@ Build-Environment:
\item maintained by Dhiru Kholia and h01ger
\item rpm repo available by Dhiru, but still \textbf{0\% reproducible}
\item rpm format includes build time and build host and…
- \item recreating the build env: unaddressed
- \item next: 24+rawhide, first reproducible rpm
+ \item recreating the build env: koji
+ \item next: 24+rawhide, first reproducible rpm, use koji
\item next: get more people involved
\end{itemize}
\begin{tikzpicture}[remember picture,overlay]
@@ -929,6 +914,30 @@ Build-Environment:
\end{tikzpicture}
\end{frame}
+\begin{frame}
+ \frametitle{Status F-Droid}
+ \begin{itemize}
+ \item not yet: \texttt{https://tests.r-b.org/f-droid}
+ \item maintained by Hans-Christoph Steiner and h01ger
+ \item work has just begun…
+ \end{itemize}
+\end{frame}
+
+\begin{frame}
+ \frametitle{Status Guix}
+ \begin{itemize}
+ \item I still have little idea, sadly missed the talk yesterday
+ \item recreating the build env: by design
+ \item user verification tool: yes! (Guix challange)
+ \item<2> next: check your inbox, h01ger!
+ \end{itemize}
+ \begin{tikzpicture}[remember picture,overlay]
+ \node[shift={(-0.15\paperwidth, 0.2\paperheight)},at=(current page.south east)] {
+ \includegraphics[height=0.33\paperheight]{images/guix.png}
+ };
+ \end{tikzpicture}
+\end{frame}
+
\placelogotrue
@@ -963,7 +972,7 @@ Build-Environment:
\begin{itemize}
\item Probably 100,000 new files per Debian suite; 50\% increase per suite
\item Mirrors would not be happy, so should not go there
- \item We'll need more files when we have detached signatures
+ \item<2> We'll need more files with detached signatures…
\item<2>{Revoking signatures?}
\item<2>{...}
\end{itemize}
@@ -986,7 +995,9 @@ Build-Environment:
\item Individuelly signed checksums (think web of trust) could work in the
Debian case (we have a gpg web of trust), but IMO won't scale.
\item<2> { We'll probably could use systematic rebuilders, run by large organisations
- (ACLU, CCC, CERN, DECIX, DESY, Deutsche Bank, EDF, EON, Greenpeace, NASA, NSA, XYZ).}
+ (ACLU, CCC, CERN, Deutsche Bank, EDF, EON, Greenpeace, NASA, NSA, XYZ).}
+ \item Fedora rebuilds Debian, Debian rebuilds OpenSUSE, OpenSUSE rebuilds
+ NetBSD, etc…
\end{itemize}
\end{frame}
@@ -1010,8 +1021,8 @@ Build-Environment:
\item We still want more (arm(64)) cores!
\item<2-6> We want to test other architectures!
\item<3-6> We want to test other projects!
- \item<4-6> We want more people looking at the results!
- \item<5-6> We want more people contributing code for their projects!
+ \item<4-6> We want more people contributing code for their projects!
+ \item<5-6> We want more people looking at the results!
\item<6> We don't want to build twice and test against what we built, but rather
the binaries distributed by these projects (if any)
\end{itemize}
diff --git a/2016-01-31-FOSDEM16/notes b/2016-01-31-FOSDEM16/notes
index ab78d57..296c413 100644
--- a/2016-01-31-FOSDEM16/notes
+++ b/2016-01-31-FOSDEM16/notes
@@ -3,6 +3,22 @@ demo: PTH=$(mktemp -d); OPTH=$PWD; P=giftrans; cp ${P}_* $PTH/; cd $PTH ; dpkg-s
really show live demo there? :)
+koji records the build
+ can be used to recreate it, in theory, in practice this needs documentation / be done
+rpm container has
+ build date
+ build host
+ signature
+ signature changes between development and release
+
+whye xplain why bit identical (eg cdroms)
+all of it:
+
+mention gsoc, new people
+
+emphasize this is too much for me…
+add thanks slide for all the people working on it in other projects
+
go through paper plans
--
Alioth's /usr/local/bin/git-commit-notice on /srv/git.debian.org/git/reproducible/presentations.git
More information about the Reproducible-commits
mailing list