This involves having passwd bits available so that seteuid("_dhcp") work,
and creating /var/empty so that chroot(_VAR_EMPTY) works. My gut feeling
is that the better solution is to make privsep and chroot optional in
dhclient, but this works well for now and is low-risk.
Approved by: re
the INDEX file is taken from the package source tree as defined by the
PKG_TREE variable. This change allows using the (possibly incomplete)
packages on pointyhat.
MFC after: 2 days
bloat on disc1.
- Output a message letting the user know that we are generating MD5 sums
during the long pause after the last mkisofs invocation in the iso.1
target.
MFC after: 3 days
- When a separate livefs ISO is used, change the name of the tree directory
to be R/cdrom/livefs and change the name of the iso to livefs.iso rather
than using disc2. Instead, disc2 is now always going to be the packages
disc regardless of the presence of a separate livefs which makes things
simpler.
- Build a very simple disc2 tree under R/cdrom that just includes the
cdrom.inf file that the packages need. Also, build a disc2.iso image
in the iso.1 target. Disc 2's volume label is "FreeBSD_Packages".
- Retire CD_EXTRA_BITS and replace it with CD_PACKAGE_TREE. CD_EXTRA_BITS
was specific to disc1, but CD_PACKAGE_TREE instead should point to a
directory that has disc1 and disc2 subdirectories that contain the extra
bits for each image.
- Rename the bootonly volume label to "FreeBSD_bootonly" and just hardcode
the disc1 and livefs volume names and iso suffixes.
there are still a couple of places under src/release such like fixit.profile
assumed that system binary can be retrivied from /stand(ex: boot with the
live CD and run "disklabel -e" in the fixit CD shell).
Since /stand is still functional in fixit floppy, and there are more than
one places in src/release needs to be updated(document for example) if we
want to make use of something like /mnt2/rescue/vi. This commit try to
deal with aforementioned inconsistency with minimal effort by simply
create a symbolic link to /rescue.
Reviewed by: jhb, sam (mentor)
The new system tries to be more automated so that there is less work for
the re's to do. It also no longer uses a /usr/ports tree as its input,
but uses the generated package build including its INDEX file as its input.
It parses the INDEX file, determines which packages should go on which ISO
images, and then builds full-fledged trees of packages that can be added
as an argument to mkisofs along with the tree built by 'make release' to
build a full CD image. The INDEX files in the populated trees are
generated with volume media number to make use of sysinstall's multiple
volume support so that the user is kindly prompted to insert the
appropriate disc for a package if it is not on the current disc. There is
still some more tweaking to be done here, but this part needs to be
committed. This stuff will all be used to prep the 5.4 release as well.
Tested by: kensmith, others on re@
Reviewed by: re
release documentation snapshots, force URLS_ABSOLUTE. This
allows us to relocate HTML renderings anywhere without worrying
about breaking relative links to pages on the Web site. This can
happen when we use &base; as part of a URL.
I wonder if we should do this for all release documentation builds,
but I'm not quite sure what all the cases are.
is only the first stage and does not yet handle packages (which will move
to their own disc(s)).
- By default, include the live filesystem on disc1, so that disc1 is now
both an install disc and a fixit disc. The images can still be split
into two separate discs by specifying SEPARATE_LIVEFS=yes.
- Remove /usr/ports from the live filesystem disc as the merged images
do not fit on even i386 otherwise. Also, the merged image already
contains a tarball of the entire ports tree, so /usr/ports was a
duplicate anyway.
- Stop building port READMES for the release sinc they are no longer used
(/usr/ports in the live filesystem was the last remaining user).
- Split cdrom.1 into three targets. cdrom.1 builds the live filesystem
image. cdrom.2 builds the disc1 image. Note that if SEPARATE_LIVEFS
is not specified these both write to the same area. cdrom.3 builds
the bootonly image and the UFS miniroot.
- The various distribution directories are now in a <buildname>
subdirectory of disc1 rather than being in the root so that the merged
image's root is not quite as cluttered.
- The disc1 image is now called disc1.iso and we no longer build a
miniinst.iso image. If CD_EXTRA_BITS is defined, then the iso.1 target
will simply include those bits in the disc1 image that it builds.
MFC after: 1 day
Requested by: re (scottl, etc.)
chroot area. This fixes make release of HEAD on systems that have a
/usr/src with bsd.endian.mk and a matching /usr/obj but haven't installed
the world in /usr/obj and thus have no bsd.endian.mk /usr/share/mk.
ports/ tree on it. This makes putting the ports/ tree on disc2 at
all dependent on the NOPORTREADMES knob as well as the NOPORT knob
(at the moment NOPORT may be set while NOPORTREADMES isn't, that
should probably be revisited). And it hardcodes ia64 for NOPORTREADMES
by request the ia64 release builder.
While here really get rid of the temp file 'make index' leaves behind.
Tested by: i386 and ia64 'make release'
Reviewed by: marcel
MFC after: 1 day
release:
- run 'make index' as part of release build, INDEX* files in
CVS no longer updated and likely to be removed from CVS soon
- don't include README.html files in ports.tgz tarball because
they cause cvsup problems for users later
- put a copy of /usr/ports on disc2 that does include README.html
files so users have access to them
Reviewed by: ru (found one bug, several clean-up suggestions)
Tested by: 'make release'
MFC before: BETA6 build (this isn't likely to be tested in -current
anyway...)
but it's disconnected from the build. Remove it from the architecture
independent release documentation set, because it breaks "make release"
when NODOC is undefined.
Reviewed by: hrs
install media on i386 and amd64. While the current default is very
useful in server environments, the ability to use USB keyboards is
vastly more important then the ability to use PS/2 keyboards the admin
forgot to plug in during the install process.
PR: kern/71443
to subordinate make(1) invocations through MAKEFLAGS, we cannot add
CFLAGS onto the make(1) command line. This will conflict with the
individual makefiles wanting to append to it, which is not respected
when CFLAGS is given on the command line. Hence build breakage.
So, put CFLAGS in the environment instead.
of releases. The -DNOCRYPT build option still exists for anyone who
really wants to build non-cryptographic binaries, but the "crypto"
release distribution is now part of "base", and anyone installing from a
release will get cryptographic binaries.
Approved by: re (scottl), markm
Discussed on: freebsd-current, in late April 2004
some unknown reason, when LOCAL_PATCHES and LOCAL_SCRIPT were originally
added, they were silently ignored if the actual file did not exist. As a
result, if one mistyped the pathname to a patch or script, then the release
silently succeeded. However, it was not built with the desired changes and
no warning was given to inform the builder either. This commit explicitly
checks to see that all of the defined patches and scripts exist up front
and bails if any of them do not exist. I lost several hours of valuable
sleeping time this evening due to this "feature" so I've finally gone and
ripped out. I've tripped over this in the past several other times as
well.
Glanced at by: scottl
we only pay attention to it during release time and at the moment it's
a -stable release in progress.
This is a MFS I guess, moving rev 1.536.2.113 to HEAD. I ran out of
time to follow through with doing this last night. Sorry for the delay.
Adding a couple of variables to make tweaking the CD-ROM label and .iso
filenames easier.
- Use the regular forth scripts including the stock device.hints file on
the boot floppies.
- gzip all the various text files on the boot floppy that are larger than
one sector to save on space.
- Generate a loader.conf to load the kernel, acpi.ko, and mfsroot making
use of the 'module_before' variables to prompt for a floppy to be entered
when required and use a stock loader.rc. This enables the beastie menu
on i386.
- PC98 has been changed to split the mfsroot for now so that the floppies
fit. When the PC98 bootstrap is updated to work with ELF files (which
needs to happen anyway), then SPLIT_MFSROOT can be turned off for PC98.
