A user reported symptoms where the losetup line used to detach the
loopback device was failing in tar mode. We don't need to detach a
device that does not exist.
Change-Id: I807996e16199288927b49b4f300ae9b461cb8fe7
Closes-Bug: #1378033
Finding the Debian boot kernel and initrd is collapsed into a
single statement so that the script doesn't bail out when
'set -o pipefail' is set and the *generic kernel and initrd are
absent.
Change-Id: Ifd616818f7387e6f102636c7cf186be6097f1d5c
Closes-Bug: 1335016
When running inside a Docker container, we cannot rely on devices in
/dev/mapper to be automagically created by udev, because we probably
don't have a udev at all. To work around this, run dmsetup mknodes
after every kpartx run.
Change-Id: If7e30579224ce54c5ed26d08974d8293c144719a
For some of the scripts, the unversioned name is not
good enough. so determine the link target of "vmlinuz" and
"initrd" to get the full versioned path.
Change-Id: I52f9b5435fafaf7925e2332058a9e41b26478f1c
Now that dib-run-parts has been moved to the dib-utils project, we
need to update diskimage-builder to use it instead of the version
directly in diskimage-builder.
This change removes the old copy of the dib-run-parts script in
the element, adds dib-utils as a dependency of diskimage-builder,
and updates the uses of dib-run-parts to correctly handle the fact
that it is now external to the project.
Requires I0be1f876d0e4a7d38e0d5c6010a552a8ebb158a4
Change-Id: Ia0a0df7784a14c49b5c47ac0b03e6c2602c84b3b
We need to be able to do install.d like things for ramdisks
themselves, but install.d runs outside the ramdisk context - and its
likely to break peoples brains if we mangle the two together - so this
adds a new hook point, ramdisk-install, specifically for installing
things into the ramdisk.
Change-Id: I37d1660309cda6e28bd0b316b08f61db4e080613
Optimized the creation of directories by using the Brace expansion mechanism.
And corrected the dirname resolution to follow correct convention.
Change-Id: I048c04385d2706c4e38f8548555a59b0437e1fcd
The two duplicated functions, save_image and finish_image will move
an existing image out of the way if it exists, but it isn't
configurable. Check an environment variable is 0 before doing so.
Switch save_image to just calling finish_image, rather than
duplicating its code exactly.
Change-Id: I26a5a8fa4b6e853c9440bffab195b0bc3728be40
Add a --options flag that allows disk-image-create to send custom
options to qemu-img calls. It is useful for forcing compat into
qemu-img created images.
Usage:
disk-image-create --qemu-img-options compat=0.10 base ubuntu
Change-Id: I73ee2c86abc115220d51ffebbbe9ce3c4d92f188
Closes-Bug: #1329746
On Debian, /lib/firmware is provided by optional packages like
firmware-iwlwifi or firmware-linux-free. That's why this directory
may not exist.
This change ensure the directory exist before trying to copy it
content.
Change-Id: Icf9eedf4148141c92e9171f16ae744b88a8d8519
The script to find best kernel from image (used by vm
and baremetal element) is duplicated and is not in sync
with each other. Moving the code to img-functions as a
function will reduce duplicated code and make it reusable
in future.
Since img-functions is not accessible in chroot env,
kernel selection is being moved from finalise.d to
cleanup.d in the vm element.
Change-Id: I8fbccc13a2c61a5191ef9ea5d2a8302a3e43b000
If we have the old image file with same name as new image file
we shouldn't rewrite it, just need to rename old.
Uses date number in old image name.
Fixes TODO in lib/common-functions
Change-Id: Ica58a0f4435217912bddf950322299d2ed2889f3
Ensure the return code of element-info is actually checked by moving the
export IMAGE_ELEMENT to the subsequent line.
A recent change (Ic0a061995e2ae708c95a535cba6237bff58f7d93) exported the
IMAGE_ELEMENT variable. This caused the return code of element-info to not
actually be checked, since the export command is now checked instead and
almost always will return 0.
This caused a misleading error message when using diskimage-builder with an
unknown element since the script didn't exit when it should have:
$ bin/disk-image-create -a amd64 foobar vm
Building elements: base foobar vm
ERROR: Element 'foobar' not found in 'bin/../elements'
Expanded element dependencies to:
Building in /var/tmp/image.DuIDY1lW
Please include at least one distribution root element.
Change-Id: I3d9c02bf0b1a4f320280185dd3fab0697d2c92f2
Adds a new pkg-map element which can be used to install
packages based on an in element 'pkg-map' JSON config
file format.
As part of this change we expose DIB's IMAGE_ELEMENT variable
so that we can have access to the element names which are being
installed in a clean manner.
This change is intended to decouple elements from DIB
and allow new elements to support multiple distributions
(with various package naming schemes) without having to
constantly maintain DIB's various bin/map-packages files.
Change-Id: Ic0a061995e2ae708c95a535cba6237bff58f7d93
If the tmpfs being used to hold the image filesystem is close to being
full then housing the raw disk image in the same place can be
problematic as it tends to grow faster then the filesystem shrinks when
data is being moved into it.
Putting them both into separate tmpfs's will allow there to be an
overlap where they jointly use more then the size limit for a singe tmpfs.
Change-Id: Ia17ca357d3b865d70a2d3e13e0479d008ca5f924
Closes-Bug: #1289582
In some instances, the content of the image may want to convey specific
resolv.conf settings. For instance, if unbound is installed, it's also
desirable to set resolv.conf to point to it.
Short circuit diskimage-builder's friendliness around resolv.conf files
by not dying in a fire if it finds the file to be immutable.
Change-Id: I88632fb79289681dcf95d32f39a0ad658ba39a60
Fixes problems found by set -eu and pipefail, including:
-Many unset variables
-Commands that can fail under normal circumstances, which breaks
with set -e. This change swallows those expected errors to allow
our existing error code to handle them.
-The dkms element was not finding Fedora kernel versions correctly.
This may be an issue for other distros too, but since Fedora was
working fine without this functionality I only changed it to print
a warning message rather than failing the build when it happens.
-The ramdisk init script will not be set -eu because if it fails
the result is a kernel panic, which can be tricky to debug.
However, in testing with set -e a few failing commands were found
and have been fixed in this patch.
Change-Id: I44cf98dfc80cfcaec54b88cc83be80a3dbf2cec3
save_image is used to copy kernel and ramdisks out of the image, which
we will sometimes want to keep the source, and sometimes not. However
for the main image itself, the temp copy is never kept, so use mv
rather than cp and avoid the excess IO.
Change-Id: I5a9f0d69ffee3e6b872a8927537ac17f02f5aa4d
This will help to ensure images that would cause bug #1289582 would
still fail even on systems that have more RAM. Also this will ensure we
only reserve enough tmpfs space for this build.
Change-Id: Icece64e691d126a5fc89262897630fdbef1ffdf1
On some systems ldd gives a slightly different output for VDSOs. It doesn't
contain a '=>'. E.g.:
linux-vdso.so.1 (0x00007fff2f4a6000)
instead of:
linux-vdso.so.1 => (0x00007fff2f4a6000)
This patch simply skips all entries in the ldd output which don't expand to a
path name to workaround that.
Change-Id: Ie37637890b775b36bb31af4e586e61131bd80fa8
cleanup for ramdisk failed to umount TMP_BUILD_DIR with device or
resource busy error.
The patch unmounts all the mountpoints under TMP_BUILD_DIR and
detaches loop devices associated with TMP_IMAGE_PATH.
The unmounts are applied with both force(-f) and lazy(-l)
options. Force option is only for NFS mounts, it's kept here since no
harm for lazy option.
Change-Id: I84035e6a003d8135186b2fda3facbd2c37967529
The sbin paths (/sbin, /usr/sbin, /usr/local/sbin) need to be in $PATH.
Since $PATH is set in the build environment from the host, if the host
does not have these paths set, they need to be added to $PATH. In
particular, they must be set for apt-get which calls out to ldconfig,
start-stop-daemon, etc.
The previous check was insufficient because if you had /usr/sbin or
/usr/local/sbin in your $PATH, the addition to $PATH was not done since
the "if" statement evaluated to true.
Change-Id: I52b41d2eb9794e7ee0ae600215cb3ceca191582f
disk-image-get-kernel is pretty noisy and you see e.g.:
$> load-image overcloud-compute.qcow2
XXX -d '/tmp/image.lWGCgPoj' -o 'tmp' -i '/home/stack/overcloud-compute.qcow2' --
Extracting kernel + ramdisk from /home/stack/overcloud-compute.qcow2 and writing them to /tmp/image.lWGCgPoj
nbd 17554 0
nbd 17554 0
basename: missing operand
Try 'basename --help' for more information.
/dev/nbd0 disconnected
tmp-vmlinuz,tmp-initrd
Clean all this up so we just get:
$> load-image overcloud-compute.qcow2
Extracting kernel + ramdisk from /home/stack/overcloud-compute.qcow2 to tmp-vmlinuz and tmp-initrd in /tmp/image.g6b0lG88
Change-Id: I8971ec0bbcd87157b07fc17254c56bb9f9f2a597
busybox on rhel has no `--list' option, which cause populate_busybox
can not create busybox symlinks. So the ramdisk created will not work
totally.
Fixes bug #1257669
Change-Id: I4f905673bb577867c99e0ccc0756c56a292474f5
Converts our existing default root element code to be just a check
which exits with a failure message if no root/distribution element
is found.
Change-Id: I954a6abfd7871d5807b1a171a03fa98932410cff
ramdisk-defaults file is sourced before chroot to an image being created
is executed. Busybox path should be set inside chroot instead.
Change-Id: Ic0e1d3093ca9d489dd96b14157e93b336cf0baee
We now forcibly remove all framebuffer video drivers from ramdisks, as
part of our ongoing effort to keep every kernel we boot, in text mode,
so it can be more easily intercepted by LOM hardware, nova console
logging, etc.
Fixes-Bug: #1178409
Change-Id: Ia72211c97c03c3f60eb9a1483297fe587de51f0b
Commit c7d80dd (Cleanup mount points automatically) removed the unmount
of $TMP_MOUNT_PATH/tmp/ccache in run_d_in_target() and moved the
"rm /tmp/ccache" to elements/base/finalise.d/02-remove-ccache. There
are two problems with this:
1) Not unmounting at the end of run_d_in_target() results in tmp/ccache
being bind mounted muliple times on top of itself (three times, if you
just run `disk-image-create base`). It is eventually unmounted, but
somehow the auto unmount code is confused, and tries to unmount it
one more time than it was mounted, which results in an error like
"umount: /tmp/image.THQkZxQa/mnt/tmp/ccache: not mounted".
This doesn't actually break anything, but it's a little messy.
2) "rm /tmp/ccache" in elements/base/finalise.d/02-remove-ccache never
succeeds in removing /tmp/ccache, because that hook is invoked by
run_d_in_target(), *while* /tmp/ccache is mounted.
This present commit solves the above by moving the ccache setup glue out
of img-functions and into the base element's root.d. This has the
following implications:
1) lib/img-functions is a little cleaner.
2) /tmp/ccache is available in the chroot during the root, extra-data,
pre-install, install and post-install stages. It is not available
during block-device, finalise and cleanup stages as it will have been
automatically unmounted by then.
3) /tmp/ccache won't be setup if you're building an image that doesn't
include the base element.
Change-Id: Ief4c0a6f4ec622db6c6f652776215684178d8943
Print a message and pause the build for 10 seconds to ensure interactive
users see the message.
Fixes bug #1212080
Change-Id: Ia388a54892c479e428b0ed7b8c70d64d65010e21
A problem with unmounting the dev filesystem in Ubuntu images caused
the umount of the /dev bind mount to fail, which left it there to be
removed during the mv -t step, causing the build host's /dev to be
wiped out.
The lazy umount will detach it from the filesystem hierarchy and then
clean up the mount reference later.
Change-Id: I8f8cea857c445fb0b4fd02bc063722fb1553c947
In some scenarios, the required space in the tmpfs partition can be
larger (or smaller) than the default one, producing errors due to
the lack of enough space (or performance penalties for not using
tmpfs).
Using --min-tmpfs <size>, we can hint the working set size we'll need
and let dib choose to avoid or use tmpfs.
Change-Id: I7d5fe498302a100c8555ae542268e14b21f3a0c5
When compressing an image, this is done in the same dir where the raw
image resides, doubling the amount of space needed (scarce when
using tmpfs), and then it's moved to the .cache folder in disk.
Combining these two functions, we reduce the amount of space needed
in the tmpfs partition (when in use), and the compressed image is
created directly on the .cache folder disk, so there is no need to
move the compressed image after the process into disk.
Change-Id: I451d24bdd6fa0983414244135dff5e96c0549833
here is a bug associated with this patch because there is a
workaround included. Calling sync to enable the unmounting...
https://bugs.launchpad.net/diskimage-builder/+bug/1214388
Change-Id: I9d5a1f092fd8f3ae1c14de03dd516203baab4da3
Manually listing the mount points that need cleanup is not
maintainable and makes it hard to write elements that use bind
mounting to inject resources into the build environment.
Change-Id: I7c9ade444f4ebe42552f8e321f257a7ec0a674ef
Most of our build process runs as root and the overhead of explicitly
tracking sudo commands in the shipped sudoers.d file, plus the
operational overhead of keeping the sudoers.d file updated on build
hosts over time, is now considered to be unnecessary.
Instead, we now document that the build process will expect sudo
commands to work and the operator must make this happen (either by
running as root, via sudo -E, or allowing passwordless sudo for the
build user).
Change-Id: If2628e7d0c7efde0fb99c84a217523ba29b1b38d
A user running di-b several times while developing an element may not
want to drop to a shell in all cases but may only want to do so if one
of their in target hooks failed.
This patch gives them the ability to do so, If break=after-error is set
then a user will be provided a in target shell taking over from where the
last failed command left off.
Change-Id: Ia2f7ac4c21b64b971f87f4ae9cb867981b13eb5e
Previously if mktemp failed in function mk_build_dir, die was not
called. The check for the exit code was actually checking that the
previous export statement completed successfully, not the call to
mktemp.
Change-Id: I477a2ce75c87c8167883ce7aa342e93c40770e29
(Based on review https://review.openstack.org/#/c/36009)
Scripts test for existence of ../share/diskimage-builder and
fall-back to ../ if not found. This allows scripts to run unmodified
from a packaged installation or a local archive/repository.
Change-Id: I0cf4c1fdb8e42ec284c56860cb15818632b93b9e
Symlink /usr/lib -> ../lib for Fedora deployment ramdisk. This
symlink is needed for systemd-udevd. Otherwise, the network device is
not loaded and initialized when the deployment image is booted. This
logic was previously contained in
elements/deploy/root.d/50-redhat-prepare-fs, but now that ramdisk
images are built in a chroot, the change needs to be moved so that it
is applied to the ramdisk build.
Change-Id: Icea43230126956ccf5fb8a6a96ca706b75d5c32f
During ramdisk cleanup if for some reason the filesystems mounted
in $TMP_BUILD_DIR were not removed this recursive delete ended up
deleting parts of the host filesystem that were mounted (in particular
most of the contents of /dev got removed.
This commit doesn't deal with the reason the filesystem didn't unmounted
but ensures the consequences are not as severe.
Fixes Bug 1202612
Change-Id: Id8eba0753c2fe76c79ae1d952ce690d26e33e3ed
- Ensures /sbin and friends are in $PATH when invoked (without this,
various sudo invocations fail in exciting ways).
- Use dib-run-parts in lib/common-functions instead of run-parts
(neither SLES nor openSUSE ship run-parts).
- Ensure dib-run-parts doesn't descend into subdirectories (same
behaviour as run-parts).
- Move dib-run-parts from root.d to bin (cleaner, consistent with
other elements with separate bin scripts).
- Tested by building Ubuntu image on openSUSE 12.3.
- Note: this doesn't add support for creating SUSE images, it just
lets you run disk-image-create on SUSE-based distros.
Change-Id: I906c6bc3cf51cdf2c4415adeae1ca250faac25e1
I missed the getopt parameter and forgot defaults are imported after
option processing. Untested code is broken code!
Change-Id: I133a691909d38e834c204950276a57f4884fc4ed
Complex image builds can download hundreds of MB of data from the
internet with many separate lookups. It would be nice to allow users
to ask for a fast build where those lookups are entirely avoided,
using locally cached resources (where possible). This new interface
allows users to signal to elements that they wish to operate without
updating cached resources, which will in turn allow us to avoid
checking for stale data at all.
As part of this I've also documented where we cache data, so that
things like the ccache cache dir and image cache files are not a
surprise to users.
Change-Id: I27f5de6ceaa4e9c6390721b7c434fe0908df84f5
Ramdisks are now built inside a chroot which is built by the normal
image build process. Doing so improves our independence of the
precise state of the build host.
This fixes bug 1194055.
Change-Id: Ibc254fbb9e7b404b5f38c1b35bcde8a4136e8e28
Many python extensions are built from source each time an image is
built. Repeating these is wasteful, so ccache is employed to eliminate
that waste. A cache dir outside the chroot is used to speed up
subsequent image builds.
Change-Id: Ib73563ddbe5f3be7454bfc54ab91cedb559a1304
systemd on Fedora 19 has made /lib/udev/rules.d/50-firmware.rules
optional:
http://lists.freedesktop.org/archives/systemd-devel/2013-March/009708.html
Firmware is now handled by the in-kernel loader.
Added a file exist check for Fedora 19 and any other distros that
may have dropped the rules file. Backwards compatible with existing
distros.
Fixes: bug #1196409
Change-Id: I1d2acf3f88543736d75eb2e8766e83a3da194638
RAMDISK_ELEMENT is a spurious difference, to merge the code we need
only semantic difference to exist.
Change-Id: I7d38b2457b6fc838d1a536406bb3c34c00f2b1f9
The finalise_base function was creating a broken link at /etc/resolv.conf
when building a non ubuntu image.
Change-Id: I974d43fd6db365dbff9c85195e3b72950f3409da
This is simply a prerequisite and we should not try to
apt-get install it inline.
Remove this code and exit with an error if qemu-img is
not found.
Change-Id: I52d47a95d24d4bfe801ba664200604bd385dc67f
Setting up a routine to run at first time the system boot may vary from
distribution to distribution. This patch will remove the logic from the
diskimage-builder code and put it into the elements.
The base element will now deploy a dib-first-boot script and each OS
element will ensure that this script runs at rc.local time. By doing
that we will put the OS specific stuff in the OS elements and remove
the embedded code out of the img-function file.
Change-Id: I24c5d1b1185de5693f145347fe912245f1ba7dfe
With systemd rc.local lives in /etc/rc.d and is only run
if the rc-local service is enabled.
This change creates a /etc/rc.d/rc.local symlink, and modifies img-functions
prepare_first_boot to tolerate an initially missing /etc/rc.local.
Fixes bug #1179674
Change-Id: I7e6102b62224950c1de5d48205dc4feff60f9389
The recent change to use loopdev instead of nbd stopped
honoring the DIB_IMAGE_SIZE setting.
This change adds it back, by resizing the image to
DIB_IMAGE_SIZE GB, if specified. If unspecified, it resizes
the image to (actual usage + 20%), as it did before this change.
Change-Id: I69afd9584e644ddacc948619100f153d3d8713a4
When partition tables are added to loopback devices, this can set off a
chain of udev hooks that may still be holding the loopback open. Failing
to detach loopback devices was the reason we were seeing leaked tmpfs
volumes.
Fixes bug #1178091
Change-Id: I836d6e2bbce824951dd4786e3ef28273ea18ee73
Sudo compiled with the --with-secure-path option enabled (default on
Ubuntu, Fedora and potentially other distros) will cause sudo to ignore
any changes made to the PATH variable in the user's environment (even when
the -E option is used). Also, some distros might not have /usr/local/bin
set in their sudoers secure_path causing the diskimage-builder to fail
when building an image using such distros as host. This patch will force
the inclusion of the /usr/local/bin in the PATH enviroment variable
inside the chroot.
Change-Id: Idb7156d8ff124bfe685cc721337d9bb21f49dc9e
Fixes: bug #1175980
Ramdisk-image-create sources both img-functions and ramdisk-functions
causing duplicate functions. To correct this situation this patch
removes the import of img-functions from ramdisk-image-create.
It moves cleanup_dirs and run_d functions from img-functions to
common-functions to allow this.
fixes bug 1175427
Authored-by: Chris Krelle <nobodycam@gmail.com>
Change-Id: I971d019b0d92b06d9de661663b49c433e823ea42
When trying to get the library dependencies from a static program
it fails and die.
This scapes the error and continue with the execution.
Change-Id: Id3463f2dd77a182ce4f9d9d28165d35f17806892
For machines with low RAM (or no access to /proc/meminfo) the builder
will still run, but will just build inside the filesystem that is
hosting /tmp. This will result in a slower build (especially if there
are a lot of .deb packages installed).
Fixes bug #1175453
Change-Id: I79f2672058c11e377548820df0ab4fad8f47ffdc
If we are using uname to collect the name of the architecture, we need
to filter its output to make it consistent with dpkg. This is now
happening in the right place to avoid a spurious warning on i386 builds.
Change-Id: Iae80dbfb757fc068155730686841c742acb7a6a2
Sudo command (even using -E) resets $PATH for security reasons, Fedora
and other distos might have to explicitly add /usr/local/bin to $PATH
in order to run the commands in the chroot environment.
Change-Id: I95068ecf1e7437152e11c6d6789f59bcfd6ae9cc
