gocryptfs/internal/fusefrontend, branch LockSharedStorage_rebase2Mirror of gocryptfs source code on Githubhttp://nuetzlich.net/cgit/gocryptfs/atom?h=LockSharedStorage_rebase22026-07-16T19:12:20+00:00fusefrontend: sharedstorage: truncate: error out on failed file lock2026-07-16T19:12:20+00:00Jakob Unterwurzacher2026-07-16T19:10:14+00:00urn:sha1:9efd5ec6c5e290a24a413dbcdbabc14e34d3177dtests/cluster: fix possible fd exhaustion2026-07-16T19:01:01+00:00Jakob Unterwurzacher2026-07-16T19:01:01+00:00urn:sha1:6a098702f11f1c2c9f5a3d102c56c3b9232719eb
...and one typo.
fusefrontend: LockSharedStorage: loop on EINTR, add debug output2026-07-16T19:00:08+00:00Jakob Unterwurzacher2026-07-16T18:47:57+00:00urn:sha1:c3f386d511094362ad5bb28c3a5c6c4f52a77236fusefrontend: sharedstorage: add two missing unlocks2026-07-16T18:34:50+00:00Jakob Unterwurzacher2026-07-16T18:34:11+00:00urn:sha1:833103864f18a4e2f514d50d5816bdc2c03ab1e0fusefrontend: sharedstorage: introduce UnlockSharedStorage helper2026-07-16T18:34:45+00:00Jakob Unterwurzacher2026-07-16T18:32:51+00:00urn:sha1:a244f1ae9e5a67c1bc2fedb80e82bb646993d936fusefrontend: sharedstorage: lock truncate agains concurrent access2026-07-16T18:10:19+00:00Jakob Unterwurzacher2024-12-30T11:08:41+00:00urn:sha1:0acbbc0913a3a7069984c8e52f64e302d7f52df4
Prevent reads and writes concurrent with the truncate operation. It's
racy on tmpfs and ext4 ( https://lore.kernel.org/all/18e9fa0f-ec31-9107-459c-ae1694503f87@gmail.com/t/ )
as evident by TestOpenTruncate test failures:
=== RUN TestOpenTruncate
cluster_test.go:209: POSIX compliance issue: non-exlusive create failed with err=file exists
doRead 16384215: corrupt block #0: cipher: message authentication failed
ino16384215 fh8: RMW read failed: errno=5
cluster_test.go:214: iteration 1: WriteAt: write /tmp/gocryptfs-test-parent-1026/1358464214/TestOpenTruncate.1788296708.mnt2/foo: input/output error
--- FAIL: TestOpenTruncate (0.06s)
FAIL
exit status 1
FAIL github.com/rfjakob/gocryptfs/v2/tests/cluster 7.880s
Relates-to: https://github.com/rfjakob/gocryptfs/issues/56
fusefrontend: sharedstorage: add warnings for lock failure2026-07-16T18:10:19+00:00Jakob Unterwurzacher2023-06-19T11:35:29+00:00urn:sha1:c12811349ec04a10e4090dc1becbf92380693f62
Complain loudly when the underlying storage does not support
byte-range locks.
https://github.com/rfjakob/gocryptfs/issues/754
fusefrontend: sharedstorage: retry read-path on EIO error2026-07-16T18:10:19+00:00Jakob Unterwurzacher2023-06-19T11:26:44+00:00urn:sha1:781f1b8be16c1ce4d5f9648641c28a13b258b2e9
With -sharedstorage, when we get a decryption error, we lock the
byte range and try again.
This makes concurrent R/W safe agains torn writes.
https://github.com/rfjakob/gocryptfs/issues/754
fusefrontend: sharedstorage: add file content byte-range locks2026-07-16T18:10:19+00:00Jakob Unterwurzacher2023-06-09T12:34:20+00:00urn:sha1:28fe256b2f48466f2d47417494b436527ca875a3
As we must write complete ciphertext blocks (except at EOF), non-overlapping
plaintext writes can overlap in the ciphertext.
And because overlapping writes can turn the data into data soup (see
TestPoCTornWrite) we serialize them using fcntl locking.
fusefrontend: sharedstorage: keep file header on truncate2026-07-16T18:10:19+00:00Jakob Unterwurzacher2023-06-05T12:22:21+00:00urn:sha1:2bac7de2b34278a9a438cb7e948d37b55959011a
With -sharedstorage, we keep the on-disk file header.
Other mounts may have the file ID cached so we cannot mess with it.
This makes TestOpenTruncate pass.