Age | Commit message (Collapse) | Author |
|
This should not happen via FUSE as the kernel caps the size,
but with fsck we have the first user that calls Read directly.
For symmetry, check it for Write as well.
|
|
This is where the results are used, so call it
there.
This simplifies doMount a bit because we can get
rid of the block protecting masterkey.
|
|
The new helper will also be used by fsck.
|
|
These can cause EBUSY errors when unmounting.
|
|
The fsck operation is not yet implemented, this commits
just adds the flag and improves cli flag handling.
|
|
To avoid confusion with fsck, rename to isDir*()
and move the functions into init_dir.go.
|
|
The replaces the "does not return" comments with
an explicit os.Exit, which is unambigous.
|
|
Now that https://github.com/pkg/xattr/pull/24
has been merged there is no reason to keep
our private copy.
Switch to the upstream version.
|
|
Also, clear out Gopkg.toml. Everything is unconstrained
anyway, so make it an empty file. Seems to work fine.
|
|
This function has been deprecated by the pkg/xattr
upstream, so write our own.
|
|
We handle empty files by storing an actual empty file
on disk. Handle xattrs similarily and encrypt the
empty value to the empty value.
|
|
At the moment, only for reverse mode.
https://github.com/rfjakob/gocryptfs/issues/217
|
|
SwitchTestParentDir changes testParentDir. This is used when you want
to perform tests on a special filesystem. For example, the xattr tests
cannot run on tmpfs and use /var/tmp instead of /tmp.
|
|
These will be reused by the upcoming xattr support.
|
|
https://github.com/rfjakob/gocryptfs/issues/218
|
|
We need the bugfix in go-fuse
|
|
|
|
macos does not have /proc, so don't try to read it.
|
|
MacOS does not have it installed by default.
|
|
Extracting the symlink fails with
linux-3.0/arch/microblaze/boot/dts/system.dts: Can't set permissions to 0755
so just exclude it.
The ln error Looks scary but is harmless, so get rid of it.
The symlink is only created to make it more convenient to view the
csv log.
|
|
macos' bash and ln lack a few features we used.
|
|
Makes the scripts work when wget is not available (macos)
|
|
|
|
|
|
macos does not have "rm --one-file-system", so try to unmount
all remaining test filesystems to protect the backing
stores.
https://github.com/rfjakob/gocryptfs/issues/213
|
|
macos does not have lazy unmount, so let's not use it
on linux either.
If the unmount fails, run "lsof" to find the open file.
Also fix the first bug we found this way.
|
|
macos rm does not understand --one-file-system,
and it cannot handle unreadable directories.
|
|
Causes "Resource busy" unmount failures on macos
|
|
Fixes test-without-openssl.bash.
|
|
|
|
test-without-openssl.bash now fails, as it should:
gocryptfs has been compiled without openssl support but you are still trying to use openssl
mount failed: exit status 18
FAIL github.com/rfjakob/gocryptfs/tests/matrix 1.943s
|
|
Not supported on macos.
Beef up the first test case a little by using different second
values.
|
|
|
|
We relied on the finalizer to close a few fds.
For some reason, this did not cause problems on Linux,
but on MacOS, it causes unmount failures:
umount(/private/tmp/gocryptfs-test-parent/194654785/default-plain): Resource busy -- try 'diskutil unmount'
|
|
Accept extra args, like test.bash does.
|
|
Gets rid of the touch error message upon running the tests.
|
|
Limit is much lower than on linux
|
|
|
|
MacOS does not have it installed by default.
|
|
These cannot work on MacOS.
|
|
The limit is much lower than on Linux.
https://github.com/rfjakob/gocryptfs/issues/213
|
|
On MacOS, symlinks don't have their own permissions,
so don't check for them.
|
|
To Go test logic waits for stderr and stdout to close, so
when we share it with a subprocess, it will wait for it to
exit as well.
We don't want the tests to hang when the unmount fails.
Seen on MacOS as reported at
https://github.com/rfjakob/gocryptfs/issues/213
|
|
A few places have called tlog.Warn.Print, which directly
calls into log.Logger due to embedding, losing all features
of tlog.
Stop embedding log.Logger to make sure the internal functions
cannot be called accidentially and fix (several!) instances
that did.
|
|
"1.10" does not work as expected because it
is parsed as a floating point number:
https://github.com/travis-ci/travis-ci/issues/9247
Added benefit is that we always get the latest point
release.
|
|
|
|
This Warn() is causing panics in the test suite
on MacOS: https://github.com/rfjakob/gocryptfs/issues/213
|
|
Go 1.10 has introduced test result caching and
enabled it by default.
This does not work properly for our integration
tests because they test the compiled binary and
do not have a source level dependency on the
gocryptfs code.
Disable caching.
|
|
|
|
We now print the number in a debug message, so define
the numeric values explicitely instead of using iota.
This way you don't have to understand how iota works
to find out what the number means. Lack of understanding
of how iota works is also the reason why the numbers
start at 3 (to keep the current behavoir).
|