Commit Graph

342 Commits (8bdbbfb099ecac884a901572d126b0fe47de5d95)

Author SHA1 Message Date
RaspberryPiFan 8bdbbfb099 Added stage 5 to the text (#353) 2019-11-27 16:11:20 +00:00
RaspberryPiFan d7c1ae21ca New types of images. (#352) 2019-11-26 15:49:49 +00:00
Leander 5456c3801f Explicitly check if postrun script exists in BASE_DIR (#351) 2019-11-26 15:03:47 +00:00
Leander 5e19033387 Remove accidental trailing brace (#348) 2019-11-18 14:38:54 +00:00
Leander 7f143a185e Export base directory before loading config file (#347) 2019-11-18 13:32:16 +00:00
snoe925 a449c75fac OS X does not have realpath (#342)
Check that realpath is present.  OS X doesn't have realpath.  The RPi builds on Docker for Mac if you increase the VM size. I am using 256G for my image.  See the Docker for Mac preferences.
2019-11-11 18:23:58 +00:00
Serge Schneider 95ba042c83 Add hostname to /etc/hosts 2019-11-06 14:33:07 +00:00
Rene Klootwijk d30343f626 Added hostname parameter 2019-11-05 17:21:33 +00:00
Serge Schneider 19c0188798 Ensure debootstrap runs through 'sh' rather than 'bash'
This fixes #338 and fixes #294
2019-10-30 13:31:07 +00:00
Matthijs Kooijman 336ff2107e Document how to get values for the localization values 2019-10-14 10:55:10 +01:00
Serge Schneider 8026d9dc8f Update changelog 2019-10-14 10:54:20 +01:00
Serge Schneider 80d486687e Update release notes 2019-09-25 19:26:16 +01:00
Serge Schneider 5780006ecd Update release notes 2019-09-25 15:06:49 +01:00
Matthijs Kooijman 99f702f0ff Use parted for partitioning (#285)
Previously, fdisk was used by sending commands into its stdin, which is
not very robust (since it heavily relies on the interactive prompts
offered by fdisk as well as the default values it offers, which seem
prone to changing in future version).

It seems likely that in the past, fdisk was easier than parted since it
provides default values that make it easier to create adjacent
partitions, without precalculating all positions in the script. However
now that partitions are manually being aligned, all data must be
calculated anyway.

This commit changes the partition generation to use parted rather than
fdisk. For this, it rewrites various calculations and renames variables
to be easier to read as well. All values are now in number of bytes,
rather than mixing bytes and sectors.

This commit also makes makes sure that the boot partition and root
partition are always adjacent (previously the root partition was aligned
without also rounding the boot partition size, leaving some empty space
in between).

As a side effect of using parted, this also causes the "bootcode" part
of the MBR to be filled with some default x86 bootcode. This is totally
irrelevant for booting the Raspberry Pi, but it does prevent triggering
a bug in parted. When using parted to change the partition table (e.g.
when resizing the root partition on first boot by raspi-config's
init_resize.sh), the disk identifier would be changed due to this bug,
which would change the PARTUUID of all partitions. The init_resize.sh
script would work around this by updating the PARTUUID in e.g. fstab,
but that's fragile at best.  This commit prevents the bug from
triggering and keeps the disk identifier the same.

See https://debbugs.gnu.org/35714 for details about this parted bug.

This commit fixes #284.
2019-09-25 13:46:38 +01:00
Serge Schneider 946f164006 stage3: Mousepad used as simple text editor instead of leafpad 2019-09-24 15:57:09 +01:00
Serge Schneider 6538d5babf Update release notes 2019-09-23 15:30:02 +01:00
fpicalausa 652780757b Update config.txt to replace lirc-rpi with gpio-ir (#328)
The example for enabling IR transmission in `/boot/config.txt` is still using the deprecated `lirc-rpi` overlay. 
The documentation in `/boot/overlays/README` indicates that this overlay has been deprecated in favor of `gpio-ir` / `gpio-ir-tx`.

This updates the actual config.txt to suggest `gpio-ir` instead of `lirc-rpi`.
2019-09-23 10:11:58 +01:00
Serge Schneider d1ed4a2982 Include rpi-eeprom 2019-09-18 14:53:35 +01:00
Russ Kubes 00c22ab57e Updated export-image to not depend on fake-hwclock and hardlink. (#326) 2019-08-27 18:16:24 +01:00
Serge Schneider e8460beb3b Revert "export-image: Remove interfaces.dpkg-old"
This reverts commit a016561600.
2019-08-14 15:54:50 +01:00
Serge Schneider 77833ba929 cmdline.txt: remove dwc_otg.lpm_enable=0 2019-08-14 15:32:05 +01:00
Serge Schneider c49261d756 stage2: install pciutils 2019-08-14 15:32:05 +01:00
Serge Schneider a016561600 export-image: Remove interfaces.dpkg-old 2019-08-14 15:28:12 +01:00
Serge Schneider c1ff2b1cac stage4: rename thonny package 2019-08-14 14:58:54 +01:00
Serge Schneider 28f344144c stage5: add scratch3 2019-08-14 14:58:54 +01:00
Joshua Bussdieker d1f7a38905 Support regional config settings (#323) 2019-08-14 11:26:55 +01:00
Christophe Hurpeau 8110eb4891 Remove unused interfaces file (#322) 2019-08-11 16:50:11 +01:00
Kevin Stone 18b945ffb2 Moved git-hash determination to passed in build variable (#273) 2019-08-11 16:44:00 +01:00
Hugo Hromic 920e22bdc5 Ensure that the configuration file is an absolute path in Docker build (#306)
* Use `&&` instead of `;` in Docker pipeline

* In case of error, `&&` does not continue execution

* Silence shellcheck warning

* SC2086: Double quote to prevent globbing and word splitting.

* Ensure that the configuration file is an absolute path in Docker build

The specific problem is in commit 2ddd7c1, where the passed config file
(using the `-c` option) is now mounted inside the container using the
`--volume src:dest:opt` Docker option.

The problem is that Docker requires absolute paths for mounting single
files inside the container, otherwise it silently tries to mount a volume
name instead as an empty directory. Therefore the Docker build no longer
works with the following invocation forms (relative config-paths):

    ./build-docker.sh -c myconfig
    /path/to/build-docker.sh -c myconfig   # also doesn't work

This commit uses `realpath` (included in coreutils) in the Docker build
script to ensure that the passed configuration file is always an
absolute path before passing it to Docker.
2019-07-30 12:38:26 +01:00
Jens Hilligsøe 5436273ec7 Make build_docker.sh portable again (#308)
Last commit made the script break on macOS.

From `man sed` (On Linux):

```
       -E, -r, --regexp-extended

              use extended regular expressions in the script (for portability use POSIX -E).
```
2019-07-23 14:44:33 +01:00
132ikl c0714e33d6 Update Stretch to Buster in README (#310) 2019-07-23 14:43:07 +01:00
Serge Schneider 9c0a25d852 stage2: add ntfs-3g 2019-07-18 12:07:45 +01:00
Serge Schneider 210b51c293 stage3: remove Epiphany 2019-07-10 15:47:17 +01:00
Serge Schneider 4c575d7e8f Add missing release notes 2019-07-10 15:42:40 +01:00
Serge Schneider 175dfb027f Update release notes 2019-07-08 19:52:00 +01:00
Serge Schneider 37482277d3 stage2: Add vl805fw 2019-07-08 19:42:09 +01:00
Serge Schneider 4118f8d524 stage5: Add Mathematica 2019-07-08 19:29:35 +01:00
gscscnd 1143530351 Revert "stage2: Add apt-transport-https" (#304)
This reverts commit 1806504983.

In Buster, APT has built‐in support for HTTPS repos (since version 1.5).
The ca-certificates package is already included in the modified file, so
this commit shouldn’t break anything.
2019-07-03 16:42:26 +01:00
gscscnd fd21eff626 README.md shouldn’t be executable (#303)
In 667318116a, README.md’s file mode bits
were changed from 0o644 to 0o755. This commit reverts them back to
0o644.
2019-07-03 13:02:48 +01:00
Samuele Maci ae4ec6445e build-docker.sh does not assume PWD=<repo root> (#302)
* bash variables in build-docker.sh are wrapped by curly brackets

* Ensure presence of config file while running build-docker.sh

* Do not assume that build-docker.sh is run from the repository root directory

* Mount config file in predictable location in docker container
2019-07-02 14:56:41 +01:00
Xerxes Rånby 57ef9b88e3 stage2/00-copies-and-fills/02-run.sh: Fix re-runs of script (#301)
Check that file exist before move.
2019-07-02 14:54:25 +01:00
Ondřej Caletka 3db1168fbd Remove obsolete IPv6 turnoff (#300)
Turning off IPv6 by aliasing `net-pf-10` to `off` does not work anymore.
Also, turning off IPv6 on system level breaks apps depending on IPv6
loopback or IPv6 link-local addresssing and creates issues in both
dual-stack and IPv6-only environments.

Signed-off-by: Ondřej Caletka <ondrej@caletka.cz>
2019-07-01 13:21:44 +01:00
Russ Kubes 4ca539cb45 Fix build.sh to run dependencies_check prior to using curl. (#299) 2019-07-01 13:17:12 +01:00
Ryan Walmsley 3c559aeb56 64 bit message (#298)
Added a handy message to make it clear that compiling on a 64 bit OS is causing issues. After spending over an hour of troubleshooting think it should be made clearer as it'll affect quite a few users.
2019-06-28 12:40:06 +01:00
Hugo Hromic cf50ff26d2 Update Dockerfile base image to Debian Buster (#295)
With the Qemu version shipped with Debian Stretch, the `man-db` package
being installed for Buster in the image triggers many of these errors:

    qemu: Unsupported syscall: 383

This is a manifestation of the following bug:
https://bugs.debian.org/cgi-bin/bugreport.cgi?bug=891109

This is resolved in the current Qemu version shipped with Debian Buster.
2019-06-27 11:47:23 +01:00
Serge Schneider a4be08e275 Update release notes 2019-06-21 15:54:50 +01:00
Serge Schneider 150e25c4f8 Revert "Java 8 workaround"
This reverts commit 3682a747bd.
2019-06-17 10:32:11 +01:00
Serge Schneider e5310d6824 stage5: install greenfoot-unbundled 2019-06-17 10:32:06 +01:00
Serge Schneider d942d5487c export-noobs: update OS.png 2019-06-07 19:51:08 +01:00
Serge Schneider 0cdec86026 Disable ld.so.preload while building 2019-06-07 19:51:08 +01:00