* test that renames from old manifests persist
* test that renamed-to deleted is handled properly with the renamed-from
becoming deleted.
Signed-off-by: Matthew Johnson <matthew.johnson@intel.com>
Perform a final link on all renames in the manifest in order to track
renames from older versions. Do not remove orphaned renames but instead
keep them around as deleted files as well.
Signed-off-by: Matthew Johnson <matthew.johnson@intel.com>
- Track ghosted files to enable renames on files that will be deleted by
third-party programs (boot files via clr-boot-manager).
- Prune orphaned renames from manifest (renamed-to deleted or renamed
again)
Signed-off-by: Matthew Johnson <matthew.johnson@intel.com>
Ghost files that will be deleted by third-party software on the
client-side instead of marking them as deleted in the manifest. This
allows the client to treat these as deleted when doing rename detection.
Remove those ghosted files from the manifest in the next update.
Signed-off-by: Matthew Johnson <matthew.johnson@intel.com>
When renamed-to files are deleted or are renamed again, the original
renamed-from file is orphaned as it has no renamed-to peer. Prune these
from the manifests by checking if the rename_peer field is NULL.
Signed-off-by: Matthew Johnson <matthew.johnson@intel.com>
Because renames are an opt-in feature (i.e. not enabled by default),
pass the required configure option for Travis builds, which enables the
rename tests.
Signed-off-by: Patrick McCarty <patrick.mccarty@intel.com>
Fixes#78
Instead of using the deprecated readdir_r convert to readdir(3). This
fixes a compiler warning in version.c.
Signed-off-by: Matthew Johnson <matthew.johnson@intel.com>
This release improves full-file creation to make it thread-safe,
allowing parallelized runs of many instances of swupd_make_fullfiles.
Signed-off-by: Matthew Johnson <matthew.johnson@intel.com>
When running many instances of swupd_make_fullfiles in parallel, the
read end of the pipe between the tar's for creating the full file of a
directory becomes reused. The observed behavior of this is
swupd_make_fullfiles hangs indefinitely with the expected tar reader
missing. The corresponding tar writer is not killed with SIGPIPE because
there is at least one reader still for the pipe, swupd_make_fullfiles.
First forking from swupd_make_fullfiles, creating the pipe, and then
fork-and-exec'ing for each tar ensures that pipe and file descriptor
management is contained for the directory rename in question and cannot
be reused by other directory renames.
Signed-off-by: George T Kramer <george.t.kramer@intel.com>
Recent outages to sourceforge break Travis CI runs. Use the github
release instead and update to the latest version of libcheck (0.11.0).
Signed-off-by: Matthew Johnson <matthew.johnson@intel.com>
Improve the test for banning debuginfo via server.ini by first creating
an update without the ban followed by an update with the ban to make
sure the debuginfo is being properly pruned from the manifest.
Signed-off-by: Matthew Johnson <matthew.johnson@intel.com>
The previous release (v3.6.1) was tagged at the wrong commit. This
release includes the changes described in that release note.
This release fixes a bug where swupd-server was creating the staged and
delta directories within the packs with 0750 permissions instead of the
expected 0700 permissions. It also adds a configuration option to
server.ini to ban debuginfo from the manifests and configure where
debuginfo libs and src are stored. Also prunes mistakenly added
debuginfo from the manifests when added accidentally and configured to
do so via server.ini.
Signed-off-by: Matthew Johnson <matthew.johnson@intel.com>
Fixes#71
Prune debuginfo from manifests when the [Debuginfo][pruned]
configuration is set to "true" in server.ini.
Signed-off-by: Matthew Johnson <matthew.johnson@intel.com>
This release fixes a bug where swupd-server was creating the staged and
delta directories within the packs with 0750 permissions instead of the
expected 0700 permissions. It also adds a configuration option to
server.ini to ban debuginfo from the manifests and configure where
debuginfo libs and src are stored.
Signed-off-by: Matthew Johnson <matthew.johnson@intel.com>
These directories were previously created with 750 permissions, which
caused a new check in swupd-client to remove them in order to correct
the permissions to 700. Create them the right way in the first place.
Signed-off-by: Matthew Johnson <matthew.johnson@intel.com>
Instead of only calculating the contentsize for files that were updated
in the current version (an inaccurate number for download size since
this is not the compressed size), calculate for all files in the
manifest. Additionally, do not add submanifest contentsizes to the
current manifest contentsize, as this will result in overcount on client
systems when multiple bundles include the same bundle.
With this change the contentsize field of the manifests will only report
the size of the files unique to that bundle. It is then the client's
responsibility to calculate total bundle size including included
bundles. This is reasonably easy to accomplish with the upcoming
swupd-client bundle-list --deps feature.
Although the above use case is a bit more work for the client, it
additionally allows a user to calculate the installation size of
multiple bundles much more easily, since it only has to count bundle
dependencies once to ensure files are not over counted.
If two bundles not in the same include chain have overlapping content,
summing the include chain of each bundle in the client will result in an
over-estimation of the total size on the system. The more content is
shared, the higher the over-estimation. In reality this overlap will not
be large, but it is currently impossible to calculate the exact
installed size using just the contentsize.
Signed-off-by: Matthew Johnson <matthew.johnson@intel.com>
This release removes a stale, unused script, adds a check to ensure format
number validity to ensure format numbers do not decrease, and enables adding
an actions: field in the Manifest.MoM to tell swupd-client to perform certain
actions post update.
Signed-off-by: Tudor Marcu <tudor.marcu@intel.com>
Adds a functional test to check for the
actions: update
line in the Manifest.MoM for the new format.
Signed-off-by: Matthew Johnson <matthew.johnson@intel.com>
When a format bump occurs and the new format is greater than the old
format, an actions field is written to the Manifest.MoM containing the
string "update". This "update" action tells the client that it is
necessary to re-execute swupd update to bring the client to the latest
version within the new format.
Signed-off-by: Matthew Johnson <matthew.johnson@intel.com>
Because the intention is for swupd format numbers to either remain the
same between LAST_VER and current, or to increase as part of a format
bump, this leaves the remaining undesirable case.
Add a basic check to make sure the format never decreases, and add a
functional test.
Signed-off-by: Patrick McCarty <patrick.mccarty@intel.com>
This release adds rename detection support which can be enabled via the
--enable-rename-detection option during configure time.
Signed-off-by: Tudor Marcu <tudor.marcu@intel.com>
If the environment variable RUN_JUST_ONE is set then skip any test
which is not numerically equal to it.
Signed-off-by: Icarus Sparry <icarus.w.sparry@intel.com>
There is no need for duplicate includes to exist in manifest headers, so
search the includes lists first before adding a new entry.
Signed-off-by: Patrick McCarty <patrick.mccarty@intel.com>
This release fixes subtracting files from manifests when both versions are marked deleted. Server subtracting of files when both are deleted
proves problematic for client updates, because the version at which the
files were deleted will differ, and the client uses the versions to
determine when it should delete files for an update.
Signed-off-by: Tudor Marcu <tudor.marcu@intel.com>
Subtracting files from manifests when both versions are marked deleted
proves problematic for client updates, because the version at which the
files were deleted will differ, and the client uses the versions to
determine when it should delete files for an update.
As an example, consider a distro with 2 bundles, "A" and "B". Bundle B
includes bundle A, which results in files in A being subtracted from B.
In this situation the following sequence of four changes result in a
subtraction that prevents a client update from deleting a file.
1) In version 10, file /usr/foo is added to bundles A and B.
2) In version 20, /usr/foo is deleted from bundle A.
3) In version 30, /usr/foo is deleted from bundle B.
4) In version 40, bundle B is modified.
Due to arbitrary modifications to bundle B in step 4, /usr/foo is
subtracted from bundle B, because it's also deleted in bundle A.
However, the file versions mismatch. So, an update from version 20 to 40
will result in /usr/foo not being deleted because the deleted entry from
bundle A was not deleted in a version newer than 20.
This patch fixes the issue by ensuring that the deleted entry in bundle
B remains intact for version 30. And the client update then works
correctly: an update from 20 to 40 will properly delete /usr/foo,
because 20 < 30 <= 40.
Signed-off-by: Patrick McCarty <patrick.mccarty@intel.com>
This release fixes server to only create deltas between two regular files,
removing errors from attempting to make deltas to/from/between symlinks.
Signed-off-by: Tudor Marcu <tudor.marcu@intel.com>
In the event that a symlink target changes between two versions, and the
files the symlink points to also change and would create a delta, do not
create a delta between the symlinks. The real delta is between the
underlying files the symlinks point to, which is created separately.
Signed-off-by: Patrick McCarty <patrick.mccarty@intel.com>
For the F->L and L->F type changes, make sure that there are no delta
files created between the dereferenced symlink on one end and the
regular file on the other.
Signed-off-by: Patrick McCarty <patrick.mccarty@intel.com>
For three different Clear Linux OS builds in the last few months, deltas
were created between files with type change L->F (symlink to file).
This was allowed to occur because there is no check if
file->peer->is_link in __create_delta().
Instead, remove the file->is_link check and simply ensure that the
from/to file types are both F (i.e. "regular file").
Signed-off-by: Patrick McCarty <patrick.mccarty@intel.com>
To simplify the discovery of C code style issues and enforce the rules
specified in .clang-format, I've added a new 'compliant' target for
running the appropriate clang-format command.
In case code style issues are found, source files are modified in place,
and the resulting diff can be viewed. The exit code in this case will be
1, so make will exit with an error. This helps to automate testing for
code style issues.
Signed-off-by: Patrick McCarty <patrick.mccarty@intel.com>
This release includes changes across various areas of server, specifically:
- Fixing fallthrough detection logic to detect files and directories correctly
when checking if they are "state" files.
- Removing stale and unused signature creation code
- Adding functional tests and enabling travis-ci integration
- Create alternative input layout to save IO for some cases(backward compatible)
- Update parallelism code to make it more versatile and editable
- Fix log call and add logging to stdout instead of just logfiles
- Honor proxy and cert checking settings
- Fix extracting files with bsdtar
- Enable locales in all programs
Signed-off-by: Tudor Marcu <tudor.marcu@intel.com>
In Ostro OS, we already have a "full" directory with all files.
Splitting it up into bundles just so that swupd-create-update can
reconstruct the "full" directory is a waste of IO, and noticably slow
when run under pseudo.
To streamline the required work, a new layout for the "image" input
directory gets introduced:
- The "full" directory gets created by the caller before invoking
swupd-create-update.
- For each bundle, instead of a <bundle> directory, there is a
<bundle>.content.txt file, listing all entries (including directories)
of the bundle.
The traditional mode of operation still works as before because each operation
which normally works with a bundle directory checks whether there is such a
directory and if not, switches to the new mode.
That way it is even possible to mix the two modes, i.e. replacing only
some bundles with a content list, although that's probably not all
that useful.
This revised commit fixes the use of an uninitialized newversiondircontent
pointer in populate_dirs().
Fixes: swupd-server/#54
Signed-off-by: Patrick Ohly <patrick.ohly@intel.com>
Matthew Johnson