Age | Commit message (Collapse) | Author |
|
This is not supported (yet? does it make sense?).
|
|
TrezorPayload stores 32 random bytes used for unlocking
the master key using a Trezor security module. The randomness makes sure
that a unique unlock value is used for each gocryptfs filesystem.
|
|
configfile.LoadConfFile() -> configfile.Load()
configfile.CreateConfFile() -> configfile.Create()
|
|
Using an external program for the Trezor PIN is not supported
at the moment.
|
|
Verify that the Trezor feature flag is set after "-init -trezor".
|
|
readpassword.Trezor() is not implemented yet and returns
a hardcoded dummy key.
|
|
10 second animation that shows what gocryptfs does.
|
|
|
|
We are clean again.
Warnings were:
internal/fusefrontend/fs.go:443:14: should omit type string from declaration
of var cTarget; it will be inferred from the right-hand side
internal/fusefrontend/xattr.go:26:1: comment on exported method FS.GetXAttr
should be of the form "GetXAttr ..."
internal/syscallcompat/sys_common.go:9:7: exported const PATH_MAX should have
comment or be unexported
|
|
Also update the performance numbers. I see some slowdown, reason
is not yet clear, but nothing to block the release.
|
|
Gopkg.toml was emptied out by commit c3e12b5e68637 which
seemed to work fine at the time. It turns out that, in
absence of a
branch = "master"
constraint, dep will use the last tag. We want latest
master, as this is what "go get" fetches, and hence
what Travis uses for testing.
|
|
Reading system.posix_acl_access and system.posix_acl_default
should return EOPNOTSUPP to inform user-space that we do not
support ACLs.
xftestest essientially does
chacl -l | grep "Operation not supported"
to determine if the filesystem supports ACLs, and used to
wrongly believe that gocryptfs does.
|
|
Define our own, with the value from Linux.
|
|
Don't dump the help text on the unsuspecting user, but
give a short error message:
$ gocryptfs -foobar
flag provided but not defined: -foobar
Invalid command line: gocryptfs -foobar. Try 'gocryptfs -help'.
For comparison: This is what cp does:
$ cp --foo
cp: unrecognized option '--foo'
Try 'cp --help' for more information.
And this what we used to do:
$ gocryptfs -foobar
flag provided but not defined: -foobar
gocryptfs v1.4.4-45-gfb772da; go-fuse v20170619-35-gb16719c; 2018-06-08 go1.10.2
Usage: gocryptfs -init|-passwd|-info [OPTIONS] CIPHERDIR
or gocryptfs [OPTIONS] CIPHERDIR MOUNTPOINT
Common Options (use -hh to show all):
-aessiv Use AES-SIV encryption (with -init)
-allow_other Allow other users to access the mount
-config Custom path to config file
-ctlsock Create control socket at location
-extpass Call external program to prompt for the password
-fg Stay in the foreground
-fusedebug Debug FUSE calls
-h, -help This short help text
-hh Long help text with all options
-init Initialize encrypted directory
-info Display information about encrypted directory
-masterkey Mount with explicit master key instead of password
-nonempty Allow mounting over non-empty directory
-nosyslog Do not redirect log messages to syslog
-passfile Read password from file
-passwd Change password
-plaintextnames Do not encrypt file names (with -init)
-q, -quiet Silence informational messages
-reverse Enable reverse mode
-ro Mount read-only
-speed Run crypto speed test
-version Print version information
-- Stop option parsing
You passed: "-foobar"
flag provided but not defined: -foobar
|
|
On Linux, where /proc exists, this makes sure that we are
executing ourselves again, and not some other copy of the
gocryptfs executable.
This usually does not matter, but mount(1) unsets $PATH
and sets argv[0] to just "gocryptfs".
|
|
mount(1) unsets PATH before calling mount.fuse. Therefore it's not set
in gocrpytfs either and daemonization fails if gocryptfs was not
executed via an absolute path.
mount.fuse handles this by leaving the execution of the helper to
/bin/sh. /bin/sh handles an empty PATH by searching a few default
locations.
This patch sets the PATH to a sane default if it's empty or unset.
|
|
When mounted via /etc/fstab like this,
/a /b fuse.gocryptfs default 0 0
we always get extra options passed. As reported by @mahkoh
at https://github.com/rfjakob/gocryptfs/pull/233 :
mount passes `-o noexec` if `-o user` is set and `-o exec` is not set.
If both `-o user` and `-o exec` are set, it passes `-o exec`.
Make these options work, and in addtion, also make -suid and -rw
work the same way.
Reported-by: @mahkoh
|
|
By returning an error instead of calling os.Exit,
error cases can be tested easily. Error cases
were not tested until now.
|
|
We need LGet support from the newest release.
|
|
Support has been merged into the xattr package
( https://github.com/pkg/xattr/pull/29 ), use it.
|
|
This currently fails because we do not use llistxattr
yet.
|
|
Fedora 28 executes A LOT of stuff from /usr/share/Modules/
on bash startup. Having -x in the shebang means we see
it all.
|
|
mv is unhappy when we return EPERM when it tries to set
system.posix_acl_access:
mv: preserving permissions for ‘b/x’: Operation not permitted
Now we return EOPNOTSUPP like tmpfs does and mv seems happy.
|
|
|
|
Values a binary-safe, there is no need to base64-encode them.
Old, base64-encoded values are supported transparently
on reading. Writing xattr values now always writes them binary.
|
|
This is what Go GCM does as well.
|
|
Pass the "flags" parameter to the lower layer syscall.
This makes Apple applications being able to successfully save data.
|
|
|
|
The tests write to the example_filesystems folder, which
1) May leave your source tree in a modified state
2) Triggers test failures when the fsck tests run concurrently,
which happens on Travis CI every now and then.
Fix both problem by copying the example_filesystems folder
to a private location in /tmp.
|
|
We previously returned EPERM to prevent the kernel from
blacklisting our xattr support once we get an unsupported
flag, but this causes lots of trouble on MacOS:
Cannot save files from GUI apps, see
https://github.com/rfjakob/gocryptfs/issues/229
Returning ENOSYS triggers the dotfiles fallback on MacOS
and fixes the issue.
|
|
These were using stale err values.
|
|
* Fixed xattr filtering for MacOS. "system." and "user." prefixes are only relevant for Linux.
* Small cleanup and additional tests.
|
|
SUSE has /bin/logger, everybody else has /usr/bin/logger,
so try both.
Fixes https://github.com/rfjakob/gocryptfs/issues/225
|
|
Reported by https://goreportcard.com/report/github.com/rfjakob/gocryptfs
|
|
Gnome may still have files open causing spurious test
failures.
|
|
One fd leak found in TestMountBackground.
|
|
And fix two in test_helpers.Mount().
Leftover fds can cause an unmount failure like this later:
fusermount: failed to unmount /tmp/gocryptfs-test-parent/873632270/default-plain: Device or resource busy
so try to catch them early.
|
|
OpenDir and ListXAttr skip over corrupt entries,
readFileID treats files the are too small as empty.
This improves usability in the face of corruption,
but hides the problem in a log message instead of
putting it in the return code.
Create a channel to report these corruptions to fsck
so it can report them to the user.
Also update the manpage and the changelog with the -fsck option.
Closes https://github.com/rfjakob/gocryptfs/issues/191
|
|
With testcases.
|
|
"ls -l" queries security.selinux, system.posix_acl_access, system.posix_acl_default
and throws error messages if it gets something else than ENODATA.
|
|
|
|
Make sure we get only 1 warning output per
problem.
Also, add new corruption types to broken_fs_v1.4.
|
|
This makes fsck runs deterministic.
|
|
Most corruption cases except xattr should be covered.
With test filesystem.
The output is still pretty ugly. xattr support will
be added in the next commits.
|
|
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.
|