Age | Commit message (Collapse) | Author |
|
Generates unique inode numbers for files on different
devices.
https://github.com/rfjakob/gocryptfs/issues/435
|
|
Check Stat() vs Fstat() result. Not very useful
at the moment as the kernel never seems to call Fstat().
|
|
Cli tool to compare the result of Stat() and Fstat().
|
|
Broken mounts may accumulate when the fs crashes.
|
|
No longer supported by go-fuse loopback.
|
|
The comment is outdated, at this point, we should
really not get any errors from ReadDirIVAt.
The change is best seen when running the fsck tests. Before:
fsck: error opening dir "missing_diriv": 2=no such file or directory
After:
OpenDir "K2m0E6qzIfoLkVZJanoUiQ": could not read gocryptfs.diriv: no such file or directory
fsck: error opening dir "missing_diriv": 5=input/output error
See https://github.com/rfjakob/gocryptfs/issues/403 , where
the extra info would have been helpful.
|
|
Suggested at https://github.com/rfjakob/gocryptfs/issues/403
|
|
This was meant as a way to inform the user that
something is very wrong, however, users are hitting
the condition on MacOS due to ".DS_Store" files, and
also on NFS due to ".nfsXXX" files.
Drop the whole thing as it seems to cause more pain
than gain.
Fixes https://github.com/rfjakob/gocryptfs/issues/431
|
|
loopback needs to receive SIGUSR1 to write a memory profile.
|
|
The "-l" flag is no longer supported in go-fuse loopback, drop
it.
|
|
|
|
|
|
|
|
There is a good chance that /tmp is tmpfs, and we want to
run our tests on a real filesystem.
|
|
We have to export TMPDIR, otherwise the Go tests to
not pick it up.
|
|
|
|
On Fedora, /tmp is tmpfs, which behaves differently than ext4
(inode numbers are never reused, for example).
Use /var/tmp, which is ext4 on Fedora, to get a more realistic
test environment.
This also allows us to drop the xattr workaround.
|
|
Closing the fd means the inode number may be reused immediately
by a new file, so we have to get the old fileID out of the table
beforehand!
Hopefully fixes https://github.com/rfjakob/gocryptfs/issues/363
|
|
This should help debugging https://github.com/rfjakob/gocryptfs/issues/363 ,
but does no harm in normal operation as it only prints ciphertext to the log.
|
|
https://github.com/rfjakob/gocryptfs/issues/363
|
|
When running
$ go test ./tests/matrix/
in isolation, it failed like this:
fd leak? before, after:
[0r=/dev/null 3w=/dev/null 5r=/proc/8078/fd (hidden:4)]
[0r=/dev/null 3w=/dev/null 5w=/tmp/go-build366655199/b001/testlog.txt 7r=/proc/8078/fd (hidden:4)]
Filter by prefix to get rid of this spurious test failure.
|
|
The tests check if they leak fds themselves, but we also
check if gocryptfs leaks fds. Clarify what is what in the
error message.
|
|
|
|
Just added a note to preinstall fuse, so the standalone
binary will work without problem.
|
|
Found while debugging https://github.com/rfjakob/gocryptfs/pull/413
|
|
|
|
Fixes https://github.com/rfjakob/gocryptfs/issues/421
|
|
Mount with idle timeout of 100ms read something every 10ms. The fs should
NOT get unmounted. Regression test for https://github.com/rfjakob/gocryptfs/issues/421
|
|
This used to dump a backtrace to the console which obscured what
is going on.
|
|
|
|
|
|
When generating man pages, pandoc marks indented code blocks with the
roff macro '.nf'. That avoids a warning from man related to the long
line about the master key.
|
|
|
|
|
|
Test if https://github.com/rfjakob/gocryptfs/pull/413 works
as intended.
|
|
Cast to uint32 to fix the int32 overflow, and define BTRFS_SUPER_MAGIC
locally to fix the next failure on darwin.
|
|
The function actually answers the question:
"is this an empty dir"?
|
|
Preallocation on Btrfs is broken ( https://github.com/rfjakob/gocryptfs/issues/395 ,
https://lore.kernel.org/linux-btrfs/CAPv9Zmk46As_P9Gyf_icET53xRda63h7iC1meES9xbdDEt9qow@mail.gmail.com/ )
and slow ( https://github.com/rfjakob/gocryptfs/issues/63 ).
|
|
The local user ID (1026 jakob) appears in the source tarballs gocryptfs_v1.7_src.tar.gz and gocryptfs_v1.7_src-deps.tar.gz as the owner of VERSION, Documentation, and vendor. This issue is already fixed for the binary releases by commit 07f57314afb260d6b14227b932d66345c55ffab3, and the solution here is the same: use "tar --owner=root --group=root".
|
|
Error was
+GOOS=darwin
+GOARCH=amd64
+go build -tags without_openssl
# github.com/rfjakob/gocryptfs/internal/syscallcompat
internal/syscallcompat/unix2syscall_darwin.go:22:32: u.Atimespec undefined (type unix.Stat_t has no field or method Atimespec)
internal/syscallcompat/unix2syscall_darwin.go:23:32: u.Mtimespec undefined (type unix.Stat_t has no field or method Mtimespec)
internal/syscallcompat/unix2syscall_darwin.go:24:32: u.Ctimespec undefined (type unix.Stat_t has no field or method Ctimespec)
caused by https://github.com/golang/sys/commit/87c872767d25fb96dfe96c794fd028b38a08440b#diff-4913a9178621eadcdf191db17915fbcb
|
|
|
|
1500 lines out output makes it hard to see where the
failure happends, especially on mobile. Drop the verbose
flag again.
This reverts commit 8cad0e2f4f288482d3528c80fb64fc1e55f26e34.
|
|
https://github.com/rfjakob/gocryptfs/issues/400
|
|
Show the GOOS/GOARCH tuple as displayed by "go version".
|
|
Fails with
get "golang.org/x/crypto/scrypt": verifying non-authoritative meta tag
package math/bits: unrecognized import path "math/bits" (import path does not begin with hostname)
Other projects have dropped the old Go version for the same
reason, example: https://github.com/nmrshll/gphotos-uploader-cli/issues/7
|
|
|
|
|
|
The parent directories need execute all permissions.
|
|
Handled the same way in GlusterFS, disorderfs, libfuse.
Fixes https://github.com/rfjakob/gocryptfs/issues/394
|
|
First test is a reproducer for
https://github.com/rfjakob/gocryptfs/issues/394 "Group permissions: No write access"
|