| Commit message (Collapse) | Author | Age | Files | Lines |
... | |
|
|
|
|
|
|
| |
We could also set the '-regextype posix-basic' option to do this
in GNU find, but that would fail with busybox.
Change-Id: Ideaa9950efc23d8351e2796b7c6f06988c8c0176
|
|
|
|
|
|
|
|
| |
- common API to 3.1.2
- udiomanager to 7.0
- node-health-monitor to 1.3.5
Change-Id: Ided9707f30fe9f1f532af36430ee827403bf4195
|
|
|
|
|
|
|
|
| |
Add newer versions of existing codec sources ogg and vorbis.
Add new codecs: tremor (integer vorbis decoder), flac and
libmad (integer MP3 decoder).
Change-Id: I39ff6c10451ad51b722ae3ef39de4fe5572af86b
|
|
|
|
|
|
|
| |
It will now show a row when a build has started, with a link to
the current build log.
Change-Id: Iec01529a80b45b2a61a6cf454a0ac8300c73e139
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
Recent autoconf (and autoconf generated ./configure files) uses
the "file" program to work out what binary format is currently
being produced by the compiler when the architecture is MIPS.
If "file" doesn't exist, or it fails to identify the binary
format, autoconf/configure silently disables the building of
shared libraries, which causes later chunks to fail to build at
link stage due to missing .so files.
Change-Id: Ia98a22d121fc8cb95d36bcb1d3ae2ce44cc6ddf5
|
|
|
|
|
|
| |
This later version will be needed by rygel and its dependents.
Change-Id: I5fd1f14108d722b5cb76f8fb74954fbe248ba2a3
|
|
|
|
| |
Change-Id: I1e96327e30051079968e39db0032b98e0c81d352
|
|
|
|
|
|
|
|
| |
The conversion from cliapp to the direct use of subprocess left a call
which did not capture stderr, so `stderr=subprocess.STDOUT` is added to
maintain the expected behaviour.
Change-Id: I3ee0c064f9813d15004234d4daea81dc8219fc42
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
This includes the following changes:
Add support for Baserock definitions version 6
Remove support for Baserock definitions format versions 0, 1 and 2
Don't provide default strip-commands for 'manual' build system
Extract downloaded tarball as current user
Add bitbucket repo alias to default aliases
sourceresolver_tests: Remove bogus test
sourceresolver_tests: Fix invalid backup definition
distbuild: Fix cache status message
distbuild: Fix crash when worker disconnects
distbuild: Fix partial distbuilding
distbuild: Add __str__() and __repr__() to ArtifactReference
distbuild: Hide a log message
distbuild: Check cache status each time we enqueue new artifacts
distbuild: Add docstrings to BuildController state machine
Fix crash in `morph distbuild` command
Make `morph show-build-log` look in local repo cache for build logs
Make more commands work outside a system-branch/workspace
Determine whether to include_local_changes from app.settings
Raise exception if show-build-log doesn't find the build log
Fix `morph show-build-log` for chunks of a different architecture
localrepocache: Count freshly cloned repositories as updated
Fix anchor yarn
Add script to run a distbuild network on the local machine
Don't ignore the 'cwd' parameter
Expose reset_workdir in GitDirectory class
Swap repr() for str() in ref add errmsg
Improve error when local HEAD of definitions repo is not found in remote
Use DefinitionsRepo class in deploy plugin
Use DefinitionsRepo in build command
Add DefinitionsRepo class
Treat True and False as actual commands, rather than booleans.
buildsystem: Comment strip command
Change-Id: Ic55e12bfe5e98f4a1a4df875c23ad3d447f7038c
|
|
|
|
| |
This was an oversight in the previous commit.
|
|
|
|
| |
Change-Id: I80ce9eee253b25689f9a360047dc9b3e9b1cb12a
|
| |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
See README for more information on how the migrations are intended work.
These migrations are probably not widely useful, as our definitions have
already been migrated manually. However, I want to come up with a good
pattern for writing migration scripts, and actually doing it seems like
the best way.
There is a 'migrations/indent' tool, which reformats a set of
definitions according to how the ruamel.yaml program writes them out.
This tool is nice if you like everything to have consistent indent and
line wrapping, and you can run it before running the migrations to
ensure that the migrations don't do any reformatting when writing the
.moprh files back to disk.
The migration scripts require the ruamel.yaml Python library. I have
sent a separate change to add this to the 'build' and 'devel' reference
systems.
Change-Id: Ibd62ba140d3f7e8e638beed6d714f671405bdc79
|
|
|
|
|
|
|
|
| |
It was moved to strata/docutils.morph in the commit
490f87c9ccc4046caeb3ab41a067a9a5ad3ad95c and we missed this
build depenency.
Change-Id: Ic40ca21f2d066d5fe859e44c647fe5b42faaf9b3
|
|
|
|
| |
Change-Id: I309f2ee3d9ff0569907c51ee9984b365b31cdbf5
|
|
|
|
| |
Change-Id: I46ca2add3ec07be66053bdf8b2a9956c680855b8
|
|
|
|
| |
Change-Id: I40f9bfc413da1dc3ea5b8110fa7fc2ab7cb2275b
|
|
|
|
|
|
| |
And upgrade from 1.16 to 1.16.3 at the same time
Change-Id: I592f49b671ad0c4ea93f4be9a8231a7f87560a29
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
This was done using the 'indent' tool, which uses a fork of PyYAML named
'ruamel.yaml' to rewrite YAML files without losing comments, ordering,
or certain elements of formatting.
My aim with doing this is to open the door to automated editing of the
reference system definitions using the 'ruamel.yaml' library. This can
be used to implement automated migration when we want to make changes to
the YAML format that we use to represent Baserock system definitions.
Although this looks drastic, remember that it's actually only altered
65 out of 608 .morph files -- the vast majority already pass unchanged
through my version of ruamel.yaml.
Change-Id: I95ec978714b5bd1c02c90183336a9fbb846cb692
|
|
|
|
|
|
|
|
|
|
|
|
| |
This is currently done by adding it to the 'morph-utils' stratum,
because it is used as a tool for working with .morph files.
The ruamel.yaml library allows writing YAML files while preserving
comments, ordering and some formatting form the input files. This makes
it more practical for us to write automated migration scripts for
Baserock definitions.
Change-Id: I886232dad4d64bcf1803f52e94ae25b0fa2d9969
|
|
|
|
| |
Change-Id: I1faa4db9e3bb5c2e8e3ef80d88b3ef3e5cf2ae26
|
|
|
|
| |
Change-Id: Ib66f3f56b60cc5dc78d08e28e281d120d83a7b9d
|
|
|
|
| |
Change-Id: I2cd784c45a4e96f59b124ca9b9bb06993b9e3f12
|
|
|
|
|
|
|
| |
* "reproducible" tar lorries
* zip support
Change-Id: Id12e868c511996d7e44651bfb4b4c885cf9748c4
|
|
|
|
|
|
| |
We forgot about this in 4a81346961031aecc50c657423bd8d68250fc40d
Change-Id: I60c73295f622abff4bafeaebfa830db0d7efcd9a
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
The 'name' field is set in both a chunk's definition, and in the
chunk reference point in each stratum where that chunk is included.
Some of them didn't agree on the name of the chunk. In all cases I've
made 'name' match the filename of the chunk .morph file, either by
fixing the chunk .morph file or the stratum .morph file (and in one
case, the filename).
This should have no effect on behaviour, it's just a tidyup.
Change-Id: I90bcd3fa382e24c8aa2d370ea5f545491169d1db
|
|
|
|
|
|
|
|
|
| |
As well as checking whether it's defined,
we make sure there isn't already a system named $VERSION_LABEL
Note, this depends on the new json-list feature of system-version-manager.
Change-Id: I0dee5253ba95a002b493f4179c2acda734afc37a
|
|
|
|
|
|
|
|
| |
A patch to fix this has been sent upstream, but the upstream
repository seems inactive.
Pull request is here: https://github.com/mitsuhiko/babel/pull/164
Change-Id: I1170e8b66c3975459dfed8ffdf66f053c2c477a5
|
|
|
|
|
|
| |
But a simple Makefile instead
Change-Id: I0d73e2cf44fc97ba9f6452ea417c1b317b68efa6
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
systemd is moving towards having /usr to be vendor operating system only,
whilst /etc will be used for configuration only
(the contents of this directory could be erased to do a factory reset) [1].
/usr/lib is the recommended place to place this file [2]
[1] http://0pointer.net/blog/projects/stateless.html
[2] http://www.freedesktop.org/software/systemd/man/os-release.html
This has been reported by Tiago Gomes
Change-Id: I58dd711fac162751a71e8a7d9c65ce9583a23cc7
|
|
|
|
|
|
| |
Also attr doesnt depend on automake to build
Change-Id: Ied3ba4af7991c44ed080362a8dae5d00092e5514
|
|
|
|
|
|
|
|
|
| |
Some of the extensions had extraneous imports, and a couple didn't
import things they used. The same bug as in the kvm deployment
extension existed in the virtualbox deployment extension, and is
fixed in this commit.
Change-Id: I2bd8169818ce5444af522c135a9e0326c1212ce2
|
|
|
|
|
|
|
|
| |
Deployment to kvm got broken by the patch to remove dependencies
on morphlib and cliapp. This commit fixes the issues that were
introduced.
Change-Id: Iface1b38e638be4d84cf86b434ee30e1fffc34a0
|
|
|
|
|
|
| |
This script helps developers to share logs via paste.baserock.org
Change-Id: I22b255c90e0b42a4b65dea41ec196ae92577a7f8
|
|
|
|
| |
Change-Id: Ifaaced5256ce7aeca786118a2075c991b3fa103f
|
|
|
|
|
|
|
|
|
| |
"OpenStackClient (aka OSC) is a command-line client for OpenStack that
brings the command set for Compute, Identity, Image, Object Store and
Volume APIs together in a single shell with a uniform command
structure."
Change-Id: Icf3200e99100f46e5d8044844232025e7bada75d
|
|
|
|
| |
Change-Id: Iffce1a4c25965ca4d7e635534fab16d0e8ccb39b
|
|\
| |
| |
| |
| |
| |
| |
| | |
This stops the deployment extensions from depending on morphlib
and cliapp, as well as setting the definitions version to 5.
Reviewed-by: Sam Thursfield <sam.thursfield@codethink.co.uk>
Reviewed-by: Paul Sherwood <paul.sherwood@codethink.co.uk>
|
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| | |
This is done by either copying some utility functions from morph
into writeexts.py, and using the `subprocess` module rather than
cliapp's runcmd and ssh_runcmd.
Note that this means that these extensions will require
"$definitions_checkout/extensions" in PYTHONPATH when they are
run.
This commit also updates VERSION to 5, since the PYTHONPATH
requirement means that this change is incompatible with old
versions of morph.
Change-Id: Iec6fa7e3c7219619ce55e18493e5c37c36e97816
|
| |
| |
| |
| | |
Change-Id: I46ed5f3ec85f9662bebac592eff7a6eb6d628f28
|
| |\
| | |
| | |
| | | |
This merge commit adds the history of the two scripts as well.
|
| | |
| | |
| | |
| | | |
Change-Id: I992dc0c1d40f563ade56a833162d409b02be90a0
|
| | |
| | |
| | |
| | |
| | | |
os.read is limited to an int in size. copy_slice_from_file was trying
to os.read more than that causing an OverflowError.
|
| | | |
|
| | | |
|
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| | |
The KVM and VirtualBox deployments use sparse files for raw disk
images. This means they can store a large disk (say, tens or hundreds
of gigabytes) without using more disk space than is required for the
actual content (e.g., a gigabyte or so for the files in the root
filesystem). The kernel and filesystem make the unwritten parts of the
disk image look as if they are filled with zero bytes. This is good.
However, during deployment those sparse files get transferred as if
there really are a lot of zeroes. Those zeroes take a lot of time to
transfer. rsync, for example, does not handle large holes efficiently.
This change introduces a couple of helper tools (morphlib/xfer-hole
and morphlib/recv-hole), which transfer the holes more efficiently.
The xfer-hole program reads a file and outputs records like these:
DATA
123
binary data (exaclyt 123 bytes and no newline at the end)
HOLE
3245
xfer-hole can do this efficiently, without having to read through all
the zeroes in the holes, using the SEEK_DATA and SEEK_HOLE arguments
to lseek.
Using this, the holes take only take a few bytes each, making it
possible to transfer a disk image faster. In my benchmarks,
transferring a 100G byte disk image took about 100 seconds for KVM,
and 220 seconds for VirtualBox (which needs to more work at the
receiver to convert the raw disk to a VDI). Both benchmarks were from
a VM on my laptop to the laptop itself.
The interesting bit here is that the receiver (recv-hole) is simple
enough that it can be implemented in a bit of shell script, and the
text of the shell script can be run on the remote end by giving it to
ssh as a command line argument. This means there is no need to install
any special tools on the receiver, which makes using this improvement
much simpler.
|
| |
| |
| |
| | |
Change-Id: I7f3702e80678aeee89dd22116510a6d8d7e04841
|
| |
| |
| |
| | |
Change-Id: I6eb8fe69416bbf483ffa8c1c317c8f8cea56ef0e
|
| |\
| | |
| | |
| | |
| | |
| | | |
This is only used by write extensions, all of which are now in
definitions. This merge commit also adds all the history of the
writeexts.py file from morphlib.
|
| | |\ |
|