IF YOU WOULD LIKE TO GET AN ACCOUNT, please write an
email to Administrator. User accounts are meant only to access repo
and report issues and/or generate pull requests.
This is a purpose-specific Git hosting for
BaseALT
projects. Thank you for your understanding!
Только зарегистрированные пользователи имеют доступ к сервису!
Для получения аккаунта, обратитесь к администратору.
Weird but the last round of image builds on e2k started complaining:
Makefile:95: *** target file `debug' has both : and :: entries. Stop.
Looks like these should have been fixed indeed.
But why didn't this surface before then?
NB: PKGDIR might be overridden now so that these functions
can be used within both metaprofile and generated profile
contexts; this isn't clean but I couldn't come up with
anything better so far.
The issue at hand is that:
/etc/tcb/USER/shadow gets USER:auth ownership (OK);
/etc/tcb/USER/shadow- backup file is root:root (broken);
/etc/tcb/USER/shadow.lock file is also root:root (broken).
This is observed for all pseudousers created by package installation
process within working chroots as well as for users created by deflogin
feature; the problem is that e.g. echo USER:PASS | chpasswd will break.
Looks like the cuplrit might be fakeroot/faked.
Looks like there's some issue with fakeroot as pseudousers
created with useradd during package installation have their
/etc/tcb/*/shadow files with proper permissions ($user:auth)
but shadow- and shadow.lock belong to root:root which makes
passwd(1) fail.
Aimed at live images at first but should cover installers as well.
This has been brewing for quite some time and while the proper
implementation is considerably more complex (and hard to do)
looks like there's demand for the particular important use case,
namely LiveCDs for Russian users, so this code has been shared
with a few people before merge.
This function's got its argument order chosen for "aesthetical"
reason of $(2) following $(1) in the macros but the logical order
is exactly the opposite: we care for kernel flavour much more than
for module set (which is dependent upon it).
So while silent dropout of kernel-image if KFLAVOURS is set
but KMODULES is empty could be fixed by testing for $(2) only,
it looks like a good time to fix this discrepancy altogether.
This change is done to reduce ambiguity in some cases;
the previous intention has been to ease navigation when
staying in a particular directory, now it's been changed
in favour of convenient toplevel `git grep' in fact.
Both variants have their pros and cons, I just find myself
leaning to this one by now hence the commit. Feel free to
provide constructive criticism :)
Some path-related bitrot has also been fixed while at that.
It appears that manually specified IMAGEDIR, e.g. by adding
IMAGEDIR = ~/out/$(shell date +%Y%m%d)
to ~/.mkimage/profiles.mk, might be problematic due to
missing globbing. Let's make sure the paths are globbed
and directories are created -- since make's wildcard()
returns an empty string when there's nothing there [yet].
The initial approach required some quite involved postprocessing
as described in http://www.altlinux.org/UEFI#HOWTO; after having
ironed out the kinks so that initial EFI support could be merged
into mkimage proper we're better off just using it, eh?
The issue actually hit image.in/Makefile: "metadata" target
in features.in/metadata/lib/50-metadata.mk wasn't reached
even if features.in/build-distro/lib/90-build-distro.mk
would ACK that the "whatever" actions included "metadata";
thus Metadata/pkg-groups.tar wasn't created and the installer
silently failed to install the .base system.
Let's armour the rest of the cases where the order of inclusion
might be important as well.
hsh-initroot leaves the chroot's root directory permissions
as 1775 while these should really be 755 at most; let's fix it
(important for both VE and VM images, useful for rescue/livecd
ones as well -- especially those with an installer onboard).
Another feature suggested by Michael Radyuk (torabora):
some images are known alpha/beta quality, it's more handy
to just state this at the build time than to rename by hand.
As it happens, adding another architecture required almost no changes;
native 32-bit ppc build took only ARCH and a repo, qemu-ppc one still
has problems (/.host/entry hangs while unpacking setup for fakedata).
Proof of concept on a QS22:
$ make ve/bare.tar.gz
** ARCH: ppc
/bin/sh: rpmvercmp: command not found
21:41:01 cleaning up
21:41:03 initializing BUILDDIR: build/
21:41:03 preparing distro config
21:41:05 starting image build (coffee time)
21:42:48 done (1:42)
** image: $TMP/out/bare-20120716-ppc.tar.gz [21M]
mkimage and hasher can make use of qemu to run
non-native binaries while working on the chroots;
thanks kas@, manowar@ and sbolshakov@ for implementing
this functionality as well as providing nice examples
through mkimage-profiles-arm and mkimage-profile-armrootfs.
This required the architecture check to be added since baking
a tarball with "arm" as its specified arch and x86_64 inside
isn't particularly good thing to let slip through; however
the implementation is quite fragile, bugreports and patches
are seriously welcome.
NB: APTCONF evaluation order between lazy make and nimble shell
turned out to be quite a delicate issue in this particular case.
The only thing to be fixed was setarch(8) symlinks assumption
that is correct for x86 but not for ARM.
There's also some hasher(7) setup to be done:
mkdir -p ~/.hasher
echo >> ~/.hasher/config <<-EOF
def_target=arm
#cache_dir=$HOME/tmp # depends on RAM/storage configuration
EOF
...and of course apt(8) should be properly set up too.
An example PoC build on a CM-A510 board (tmpfs):
$ make BRANDING=altlinux-centaurus ve/bare.tar.gz
** ARCH: arm
18:10:45 initializing BUILDDIR: build/
18:10:45 preparing distro config: build/distcfg.mk
18:10:46 starting image build: tail -f build/build.log
18:14:49 done (4:02)
** image: $TMP/out/bare-20120706-arm.tar.gz [23M]
An initial draft of it was done half a year ago but several tricky
thingies had kept the code from showing up as it was rather brittle
and incomplete.
This implementation involves quite a few changes all over the place
but finally works good enough for live and installer images.
Please pay attention to the versions of these packages:
- installer-feature-setup-plymouth (0.3.2-alt1+)
- branding-altlinux-sisyphus (20110706-alt2+ if used)
- plymouth (0.8.3-alt20.git20110406+)
See also:
- http://www.altlinux.org/Branding
- http://www.altlinux.org/Plymouth
This further refines the modular build by making
metadata being a clearly separated feature rather
than having to rely on runtime tests, and also by
moving the code which cares for kernel bits of base
installation (.base list) in a feature of its own.
There's more to it but let's get the ball rolling first.
As duly noted by glebfm@, branding issues need more attention
by now since only stage1/install2 got some of it so far in this
regard. Hence the dedicated feature comes to the rescue
(well no, it doesn't actually mess with rescue!).
Following m-p-d, a more involved default output directory
structure is feasible now:
~/out/name/date/name-date-arch.type
instead of plain
~/out/name-date-arch.type
This particular behaviour can be achieved by passing
SORTDIR='$(IMAGE_NAME)/$(DATE)'; note the single quotes.
Reports are also saved in this resulting structure
albeit the place is still highly debatable.
From now on, non-empty SAVE_PROFILE variable will indicate
the need to carry the particular generated profile inside
the image built from it.
Thanks gns@ for this feature in liveflash.eeepc.
It was briefly mentioned in QUICKSTART but somehow managed
to evade the commandlines provided. And while at it, let's
make errors like this more explicit to avoid extra lookups.
Oh, and fix QUICKSTART so that readers miss the hassle. :)
Thanks Vladimir Karpinsky for pointing this problem out.
The former toplevel Makefile is now toplevel main.mk;
this change allows for multi-target, multi-arch processing
in the current toplevel Makefile.
As the "build" symlink semantics change quite considerably
when one is doing bulk builds (several pruned builddirs might
be useful for comparison), BUILDDIR is now much more likely
to be recreated: the cases when it will persist are when it's
either a single-image build or when the prefix hasn't changed.
There are some more or less subtle bugfixes and enhancements
all over the map as well.
Done within 20111230..20120102 timeframe, actually...
Here we go with postprocessing priorities along the way
as ISO hybridization has to occur before implanting
final MD5 sum (which must happen earlier than e.g. some
external MD5SUM file generation).
Unfortunately proper dependencies aren't applicable here
(though I'd like to be proved wrong on this one).
Please note that this needs propagator > 20101130-alt9
for automatic mode to work (has also been worked around
in gfxboot case with design-bootloader-source-6.0-alt1).
Thanks rom_as@ for asking about the hybrid image status
and helping out with testing.
We've got some parts of it in build-distro feature,
and some went to dev feature for no real reason.
But a bare installer might go without package base,
and LiveCDs other than live-builder might find local
repository useful given aufs2 root overlay.
Now the overall scheme is more straightforward:
- a distro:
+ asks that a package repo be included
+ cares to further add the packages to it
- a repo feature:
+ pulls in sub/main for it to happen
+ provides genbasedir script to create repo metadata
+ supplements live feature with repo configuration
This is a base for "media check" to become available:
using this feature will implant a checksum into the image
so that it can be verified during install.
Also added a test/demo distro/live-isomd5sum target.
For real distros an alterator module is probably due.
