Daemon-mode tests default to the stdio-pipe transport (RSYNC_CONNECT_PROG),
which opens no listening socket -- so `make check` never exposes a network
service. Real TCP is opt-in via `runtests.py --use-tcp`, with the daemon
bound to loopback (127.0.0.1) on a claim_ports()-reserved port; CI runs the
suite both ways.
start_test_daemon() is the single seam every daemon test uses: the secure
pipe by default, a real rsyncd on a claimed loopback port under --use-tcp.
Tests with no pipe equivalent (the fake-proxy listener and the reverse-DNS
hostname-ACL daemon test) are gated behind require_tcp().
`make check` also now runs the suite in parallel by default (CHECK_J=8);
the claim_ports() byte-range locks make that safe across concurrent runs.
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
rsyncfns.claim_ports(*ports) takes exclusive POSIX byte-range locks on
/tmp/rsync_test.lck (offset = port number) so any number of test
processes can run concurrently without colliding on a TCP port: a test
asking for a port already held blocks until the holder exits. The
kernel drops the locks automatically when the holding process dies, so
a crashed test releases its ports with no manual cleanup.
Ports are claimed in sorted order so two callers requesting the same
set in different orders can't deadlock. The lock file is forced to
mode 0o666 after creation (the umask would otherwise trim it and lock
out a second user on a shared CI runner; EPERM when we're not the
owner is fine).
proxy-response-line-too-long is the first user: it switches from an
ephemeral port to a claimed fixed port (12873).
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
Replace the entire shell-based testsuite with Python. runtests.py
already drove the suite (it had replaced runtests.sh earlier); this
converts all 60 test scripts from *.test shell to *_test.py and adds
testsuite/rsyncfns.py as the shared helper module -- the Python
counterpart of the now-removed rsync.fns.
runtests.py:
* Discovers and runs both *.test and *_test.py; dispatches the
Python tests via the same python3 that runs the harness.
* Extends PYTHONPATH so tests can `import rsyncfns`.
testsuite/rsyncfns.py provides everything the ports need:
* environment wiring (scratchdir / srcdir / TOOLDIR / RSYNC /
TLS_ARGS, and HOME pointed at the per-test scratch dir);
* result reporting -- test_fail / test_skipped / test_xfail mapping
to the 0 / 1 / 77 / 78 exit-code convention;
* the transfer-and-verify helpers checkit, checkdiff, verify_dirs,
rsync_ls_lR, check_perms and the v_filt output filter;
* fixture builders hands_setup, build_symlinks, build_rsyncd_conf,
make_data_file, cp_p / cp_touch, makepath / rmtree.
All 60 tests are converted, including the four split-variant tests
that share one source via a Makefile-built symlink (chown/chown-fake,
devices/devices-fake, xattrs/xattrs-hlink, exclude/exclude-lsh);
Makefile.in's CHECK_SYMLINKS now points at the *_test.py names.
The dead rsync.fns shell library is removed.
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
Cross-compiles statically-linked rsync binaries with the Android NDK for
arm64-v8a (all modern phones) and armeabi-v7a (older 32-bit devices), and
uploads them as workflow artifacts for adb push / Termux use.
The build is self-contained (optional external libraries disabled; keeps
md5/md4 and the bundled zlib) and forces a few configure cache values
that can't be probed when cross-compiling: lchmod()/lutimes() off (Bionic
doesn't declare them until API 36 though the symbols link), and
socketpair / mknod-FIFO / mknod-socket on (Android runs a Linux kernel,
so these match the native result). IPv6 is enabled explicitly.
Since the binaries are cross-compiled the test suite can't run; the job
instead asserts each binary is static and the correct architecture, and
smoke-tests `--version` under qemu-user.
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
symlink-dirlink-basis.test and chdir-symlink-race.test both
require a multi-kilobyte non-trivial-content source file for the
rsync delta algorithm to exercise. Both used dd / head against
/dev/urandom directly, which fails on platforms that don't ship
/dev/urandom (e.g. HPE NonStop). The dd error gets swallowed by
'2>/dev/null' and the test then fails with a misleading 'failed
to create test file' that hides the real cause.
Add make_data_file <path> <size> to testsuite/rsync.fns. Prefers
/dev/urandom when readable (kernel-provided randomness, fast),
falling back to a deterministic awk LCG seeded from PID and a
POSIX cksum of the destination path. Output is constrained to
printable ASCII (33..126) so the helper survives two awk-portability
quirks:
- printf '%c', 0 terminates the string in some awks, emitting
fewer than sz bytes;
- gawk in UTF-8 locales encodes printf '%c', N for N > 127 as
a 2-byte UTF-8 sequence, emitting more than sz bytes.
The tests don't need 8-bit binary entropy -- they just need
non-trivial bytes for rsync's block-matching algorithm.
Update both call sites to use the helper. Linux/FreeBSD/macOS
still take the /dev/urandom fast path; NonStop and any other
platform missing the device get the awk fallback transparently.
Both paths verified locally with the symlink-dirlink-basis test.
The chmod-symlink-race test was previously a no-op on Solaris,
OpenBSD, NetBSD, and Cygwin via a case 'uname -s' skip. The skip
was too broad: of the four scenarios the helper exercises, only
the 'legitimate within-tree dir-symlink' one actually needs
RESOLVE_BENEATH-equivalent kernel support. The other three
(attack rejection, plain relative path, top-level file) behave
identically on the per-component O_NOFOLLOW fallback and would
have caught the t_stub.c max_alloc=0 bug fixed in the previous
commit if the test had been allowed to run.
Make the helper probe the running kernel for either
openat2(RESOLVE_BENEATH) on Linux 5.6+ or openat(O_RESOLVE_BENEATH)
on FreeBSD 13+ / macOS 15+ by opening '.' under the requested
confinement. Honour the result:
- If RESOLVE_BENEATH-equivalent confinement is available, the
within-tree symlink scenario must succeed (status quo).
- If not, the per-component O_NOFOLLOW fallback rejects every
symlink including legitimate ones; expect the within-tree
symlink scenario to be rejected (rc != 0) and the file mode
to remain unchanged.
The attack-rejection, plain-path and top-level scenarios are
unchanged: they expect the same outcome on both code paths.
Drop the case-based skip from chmod-symlink-race.test so the test
runs everywhere and the per-component fallback gets the CI
coverage that the SunOS/OpenBSD/NetBSD/Cygwin runners can
provide. HPE NonStop -- which lacks RESOLVE_BENEATH but isn't in
the existing skip list -- is also covered by this change.
The t_stub.c shim defined max_alloc = 0 as a placeholder to satisfy
the link against util2.o. This was harmless when the test helpers
made no allocations, but the secure_relative_open() implementation
in 3.4.0+ calls my_strdup() in its per-component O_NOFOLLOW
fallback (syscall.c around line 1857), and the 3.4.3 do_*_at()
hardening series added more such calls. With max_alloc=0, every
allocation in that path trips the 'exceeded --max-alloc=0' check in
util2.c's my_alloc(), and t_chmod_secure (which exercises
do_chmod_at via secure_relative_open) fails on the very first
my_strdup.
The failure is invisible on Linux 5.6+ / FreeBSD 13+ / macOS 15+ /
recent Cygwin because those platforms take the kernel-enforced
openat2(RESOLVE_BENEATH) or openat(O_RESOLVE_BENEATH) branch and
never reach the per-component fallback. It also goes unobserved
on the SunOS/OpenBSD/NetBSD/CYGWIN* CI runners because the
chmod-symlink-race.test script case-skips on those platforms (the
legitimate dir-symlink scenario the test exercises can't pass on
the per-component fallback). HPE NonStop is the first platform
that lacks RESOLVE_BENEATH support AND isn't in the skip list AND
has someone actually running the test suite, so it surfaced the
latent bug.
Raise max_alloc to SIZE_MAX so the helpers can allocate freely.
A follow-up patch makes t_chmod_secure adapt at runtime so the
skip list can be removed and the per-component fallback gets real
CI coverage.
The .filt file in /home/ftp/pub/rsync on samba.org controls which
subtrees release.py's FTP mirror excludes (currently /binaries/
and /generated-files/). Without it, step-10-push-ftp's
'rsync --del' would propagate local deletions to the server even
for those archive subtrees.
Until now the only copy of this two-line file lived on the server.
Bundle it in source at packaging/ftp.filt so it survives a disaster
on samba.org, and have step_1_fetch seed FTP_DIR/.filt from the
bundled copy on every run (with --exclude=/.filt on the rsync pull,
so the server's copy can't silently drift the bundled one).
step-10-push-ftp then propagates any in-source updates to the
filter back to the server.
Both scripts were pre-release.py legacy helpers:
* samba-rsync rsync'd ~/samba-rsync-{ftp,html}/ to the samba.org
server. release.py step-10-push-ftp and step-11-push-html now
do exactly this, using ../release/rsync-{ftp,html}/ as the
local mirrors.
* send-news copied README/INSTALL/NEWS .md + .html files into
~/samba-rsync-ftp/ and rsync'd them to samba.org.
release.py step-8-update-ftp already does this
(./md-convert --dest=FTP_DIR README.md NEWS.md INSTALL.md and
the surrounding rsync of html files into FTP_DIR), and
step-10-push-ftp pushes the result.
Update the trailing instructions printed at the end of
step-12-push-git to drop the now-obsolete 'run packaging/send-news'
suggestion, and tighten the comment in step_1_fetch that referred
to samba-rsync as a current sibling tool.
Track the move of rsync-web from sibling git checkout to a regular
subdirectory of the rsync source tree:
* HTML_SRC: '../rsync-web' -> 'rsync-web'.
* step_1_fetch: drop the .git-presence probe and the 'make sure
it's up to date' reminder. Both made sense when rsync-web was
a separate repo the maintainer had to clone and pull, but the
directory is now part of the same checkout as this script.
* rsync invocation no longer needs --exclude=/.git: there is no
.git inside rsync-web/ (it is just a subdir of the parent
rsync-git checkout).
* Header comment block and step-1 help text rewritten to describe
the new layout.
Fold the standalone rsync-web repo into the rsync source tree as
rsync-web/, eliminating the sibling-checkout convention and the
drift it causes between the release-time HTML snapshot in
../release/rsync-html and the source of truth in ../rsync-web.
Flat-copy import (no git history merge). The standalone repo at
github.com/RsyncProject/rsync-web is retained for historical
reference and will be archived once the in-tree copy proves itself.
Add /rsync-web/ to .gitattributes with export-ignore so the
website content does not bloat the release source tarball
produced by 'git archive' in packaging/release.py step_7_tarball.
A follow-up commit repoints HTML_SRC in packaging/release.py at
the new in-tree location.
Most Ubuntu users landing on INSTALL.md want to install rsync, not
build it. Add a short section near the top that offers the
Launchpad PPA as the one-line path for the four currently supported
series (jammy 22.04 LTS, noble 24.04 LTS, questing 25.10,
resolute 26.04 LTS), and clarify that the rest of the file is about
building from source.
Drops the "dev" suffix on RSYNC_VERSION ahead of the
2026-05-20 00:00 UTC public release.
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
Set the date to 20 May 2026, add a SECURITY FIXES section listing
all six May 2026 CVEs (CVE-2026-29518, -43617, -43618, -43619,
-43620, -45232) with reach, root cause, fix and reporter for each,
plus a note on the defence-in-depth hardening that goes with them.
Also list the new symlink-race regression tests under DEVELOPER
RELATED.
fixes a one byte stack overflow when using RSYNC_PROXY with a
malicious proxy.
Reach: only when RSYNC_PROXY is set and a malicious or MITM'd
proxy returns the pathological response. The byte written is
always '\0' and the attacker doesn't choose the offset, so impact
is corruption of one adjacent stack byte and possible later
misbehaviour or crash -- no information disclosure beyond the
existing rprintf of buffer contents.
Reported by Aisle Research via Michal Ruprich
read_del_stats() in main.c accumulates 5 wire-supplied counts into
the int32 stats.deleted_files field:
stats.deleted_files = read_varint_bounded(..., MAX_WIRE_DEL_STAT, ...);
stats.deleted_files += stats.deleted_dirs = ...;
stats.deleted_files += stats.deleted_symlinks = ...;
stats.deleted_files += stats.deleted_devices = ...;
stats.deleted_files += stats.deleted_specials = ...;
With the previous MAX_WIRE_DEL_STAT = 2^30 (1.07 GB) the worst-case
sum is 5 * 2^30 = 5.37 GB; three maximal values already exceed
INT32_MAX = 2.15 GB on the third "+=", triggering signed integer
overflow (C99 6.5/5 -- undefined behaviour, the compiler may assume
it cannot happen and elide subsequent checks).
The bound was introduced in f0155902 ("defence-in-depth: bound
wire-supplied counts and lengths") with a commit message claiming
"per-summand cap so the total can't overflow", but 2^30 * 5 does
overflow. Lower the per-summand cap to 2^28 (= 268M) so the worst
case is 5 * 2^28 = 1.34 GB < INT32_MAX with margin. 2^28 deletions
per category is still vastly above any plausible real transfer.
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
Two assorted audit findings:
- receive_data() never bounds-checked the block index returned
by recv_token() against sum.count before computing offset2
and feeding it to map_ptr(). An out-of-bounds index from a
hostile sender produces invalid memory access. Add a
sum.count bounds check.
- read_delay_line()'s strchr() call could return NULL when no
space was found, but the code unconditionally added 1 to the
result before dereferencing. Low impact (just a disconnect on
exit of the client-specific forked process) but the NULL
deref is real. Guard the NULL.
Both reported by Joshua Rogers.
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
Two cumulative-snprintf patterns in log.c (rsyserr) and main.c
(output_itemized_counts) had the shape
len = snprintf(buf, sizeof buf, ...);
len += snprintf(buf+len, sizeof buf - len, ...);
with no guard between calls. snprintf returns the would-have-been
length on truncation, so a truncated first call leaves
"sizeof buf - len" as a negative-then-promoted-to-size_t value,
underflowing into a huge size_t and writing past buf.
Realistic exposure is small in both cases (log header well under
buffer, only ~5 itemized iterations writing ~25 chars each into a
1024-byte buffer) but the defect class matches bb0a8118 and the
fix is cheap. Guard before each subsequent call.
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
Multiple receiver-side fields read from the wire were trusted
without upper-bound checks. A hostile peer could either request
extreme allocations (DoS via --max-alloc) or, on platforms where
read_varint returned a negative value, push ~SIZE_MAX through the
size_t conversion to wrap downstream length checks.
Introduce read_int_bounded(), read_varint_bounded() and
read_varint_size() in io.c so wire-derived integer ranges are
checked at the read site rather than scattered across each
caller, with RERR_PROTOCOL on out-of-range input.
Apply the bounded primitives to:
- sum->count (checksum count -- previously could overflow
(size_t)count * xfer_sum_len on 32-bit with raised max-alloc)
- xattrs: count, name_len, datum_len, plus rel_pos overflow
detect to stop chain wrapping the num accumulator
- acls: ida-entry count
- flist: file mode S_IFMT validation, modtime_nsec range check
- delete-stat counters in main: per-summand cap so the total
can't overflow a signed 32-bit accumulator
Reporters include Joshua Rogers (checksum-count overflow finding).
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
On an rsync daemon configured with "daemon chroot", the reverse-DNS
lookup of the connecting client was performed *after* the chroot
had been entered. If the chroot did not contain the files glibc
needs for resolution (/etc/resolv.conf, /etc/nsswitch.conf,
/etc/hosts, NSS service modules), the lookup failed and
client_name() returned "UNKNOWN". Hostname-based deny rules
("hosts deny = *.evil.example") therefore could not match, and
an attacker controlling their PTR record could connect from a
hostname the administrator had intended to deny. IP-based ACLs
were unaffected.
Do the reverse DNS lookup before chroot/setuid; client_name()
caches its result, so the post-chroot call uses the cached value
and hostname-based ACLs work even when DNS is unavailable
post-chroot.
Adds testsuite/daemon-chroot-acl.test as end-to-end regression
coverage. The test sets up an empty chroot directory, configures
"hosts deny = <localhost-resolved-name>" with daemon chroot, and
asserts the connection is refused with @ERROR access denied.
Uses unshare --user --map-root-user for non-root CAP_SYS_CHROOT;
skips cleanly on non-Linux or when user namespaces aren't
available.
Reporter: Joshua Rogers (MegaManSec).
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
Commit 797e17f ("fixed an invalid access to files array") added a
parent_ndx < 0 guard to send_files() in sender.c, but the visually-
identical block in recv_files() in receiver.c was not updated. A
malicious rsync:// server can therefore drive any connecting client
into the same out-of-bounds dir_flist->files[-1] read followed by a
file_struct dereference in f_name() one line later.
Reach: protocol-30+ default (inc_recurse) makes flist.c:2745 set
parent_ndx = -1 on the first received flist when the sender omits a
leading "." entry; rsync.c flist_for_ndx() does not reject ndx == 0
in that state because the range check evaluates 0 < 0 = false; and
read_ndx_and_attrs() only validates ndx with the ITEM_TRANSFER bit
set, so iflags=ITEM_IS_NEW (or any other non-transfer iflag word)
bypasses the check.
Apply the same guard receiver-side. Confirmed: the same PoC (a
minimal Python rsyncd that handshakes with CF_INC_RECURSE, sends a
no-leading-"." flist, and emits ndx=0 with ITEM_IS_NEW) crashes
unpatched 3.4.2 with SEGV_MAPERR si_addr=0x4101a-class in the
receiver child; with this guard it exits cleanly with code 2
(RERR_PROTOCOL).
The attack surface delta over the sender variant is large:
the original was malicious-client -> daemon, this is
malicious-server -> any rsync client doing a normal rsync://
or remote-shell pull.
Reported by Pratham Gupta (alchemy1729).
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
Add a daemon-refuse-compress test that builds a module configured with
'refuse options = compress' and asserts that:
1. an attempted -z transfer to that module fails with an error
mentioning --compress, and
2. the same transfer without -z still succeeds.
This pins down the documented way to disable all compression on a
daemon, which previously had no automated coverage.
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
The receiver's three compressed-token decoders --
recv_deflated_token (zlib), recv_zstd_token, and
recv_compressed_token (lz4) -- accumulated rx_token (a 32-bit
signed counter) without overflow checking. A malicious sender
could craft a compressed-token stream that walked rx_token past
INT32_MAX, with careful manipulation leaking process memory
contents to the wire (environment variables, passwords, heap
pointers, library pointers -- significantly weakening ASLR
and facilitating further exploitation).
Cap rx_token at MAX_TOKEN_INDEX = 0x7ffffffe. Fold the
bookkeeping into recv_compressed_token_num() and
recv_compressed_token_run() shared by all three decoders. Reject
negative or out-of-range token values explicitly. Also cap the
simple_recv_token literal-block length at the source: any
wire-supplied length > CHUNK_SIZE is ill-formed (the matching
simple_send_token never writes a chunk larger than CHUNK_SIZE),
so reject before looping on attacker-controlled bytes.
Reach: an authenticated daemon connection with compression
enabled (the default for protocols >= 30 when both peers
advertise it). Disabling compression on the daemon
("refuse options = compress" in rsyncd.conf) is the available
workaround.
Reporter: Omar Elsayed (seks99x).
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
Cygwin lacks RESOLVE_BENEATH-equivalent kernel support and the
per-component O_NOFOLLOW fallback also can't be exercised meaningfully
under the cygwin runner's filesystem semantics, so every test that
asserts the secure_relative_open / do_*_at machinery actually blocks
the attack would skip. Make those skips expected in the workflow's
RSYNC_EXPECT_SKIPPED list:
- chdir-symlink-race
- chmod-symlink-race
- bare-do-open-symlink-race
- sender-flist-symlink-leak
- daemon-chroot-acl
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
testsuite/chdir-symlink-race.test runs an actual rsync daemon
(via RSYNC_CONNECT_PROG to avoid the network) configured with
"use chroot = no", plants a symlink at module/subdir -> ../outside,
and runs four flavours of attacker-shaped transfer (single-file
poc_chmod, -r push into the symlinked subdir with --size-only and
without, -r push into the module root). All four must leave the
outside-the-module sentinel file's mode AND content unchanged.
Portability:
- file_mode() helper falls back to BSD stat -f %Lp when GNU
stat -c %a is unavailable (macOS, FreeBSD).
- Pre-saved pristine copy + cmp(1) replaces sha1sum, which
differs across platforms (sha1sum / shasum / sha1).
Tests are kept running as root in the user-namespace re-exec
wrapper used by symlink-race tests so the daemon's setuid path
doesn't drop into the test user's identity (which on Linux
would mean the chmod-escape code path can't trigger because
the test user doesn't have CAP_FOWNER over the outside file).
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
Three related codex audit findings:
Finding 3a: copy_file()'s source open in util1.c used
do_open_nofollow(), which only rejects a final-component
symlink. A parent-component symlink (e.g. --copy-dest=cd where
cd -> /outside) follows freely and reads outside the module.
Route through secure_relative_open() with O_NOFOLLOW.
Finding 3b: generator.c's in-place backup-file create still
used a bare do_open with O_CREAT, leaving a tiny but reachable
parent-symlink window between the secure unlink (already
through do_unlink_at) and the create. Add do_open_at() that
goes through a secure parent dirfd, and route the call site
through it.
Finding 3c: copy_file()'s destination open in
unlink_and_reopen() had the same bare-do_open pattern; route
through do_open_at as well.
Adds testsuite/copy-dest-source-symlink.test and
testsuite/bare-do-open-symlink-race.test as regression coverage
for both attack shapes.
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
Add the rest of the path-based syscall wrappers and migrate every
receiver-side caller:
- do_lchown_at, do_rename_at, do_mkdir_at, do_symlink_at,
do_mknod_at, do_link_at, do_unlink_at, do_rmdir_at,
do_utimensat_at, do_stat_at, do_lstat_at
Same shape as do_chmod_at: open each parent under
secure_relative_open(), call the *at() variant against the dirfd,
fall through to the bare path-based syscall in non-daemon /
chrooted / absolute-path / no-parent cases. macOS's
setattrlist-based set_times tier is also routed through the
utimensat_at path on daemon-no-chroot.
Hardenings to secure_relative_open() itself:
- confine basedir resolution under the same kernel mechanism
used for relpath (basedirs from --copy-dest / --link-dest are
sender-controllable in daemon mode)
- reject any '..' component (bare '..', 'foo/..', 'subdir/..')
so the per-component O_NOFOLLOW fallback can't escape
- return the dirfd we built up from the per-component fallback
when the caller passed O_DIRECTORY (otherwise every do_*_at
failed with EINVAL on platforms without RESOLVE_BENEATH)
Adds testsuite/alt-dest-symlink-race.test and
testsuite/secure-relpath-validation.test (with t_secure_relpath
helper) as regression coverage for the new hardenings.
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
The receiver's chdir(2) into a destination subdirectory followed
attacker-planted symlinks at every path component. Once CWD
escaped the module, every subsequent path-relative syscall (open,
chmod, lchown, ...) inherited the escape -- defeating
secure_relative_open's RESOLVE_BENEATH anchor against AT_FDCWD,
since the anchor itself was now outside the module.
Route change_dir's relative target through secure_relative_open()
and fchdir() to the resulting dirfd in am_daemon && !am_chrooted
mode, so the chdir step itself can no longer follow a parent-
symlink. Same treatment applied to the CD_SKIP_CHDIR /
set_path_only path so it also can't follow attacker symlinks
during path tracking.
Adds testsuite/sender-flist-symlink-leak.test covering the
sender-side flist resolution variant of the same primitive.
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
CVE-2026-29518's fix routed the receiver's open() through
secure_relative_open(), but every other path-based syscall the
receiver runs on sender-controllable paths is vulnerable to the
same TOCTOU primitive. This commit closes the chmod variant.
Add do_chmod_at() that opens the parent of fname under
secure_relative_open() and uses fchmodat() against the resulting
dirfd. Gate the secure path on am_daemon && !am_chrooted (the same
gate use_secure_symlinks already uses for the receiver basis-file
open), so non-daemon callers and chrooted daemons keep the original
do_chmod() fast path.
Migrate the receiver-side do_chmod() call sites in delete.c,
generator.c, rsync.c, and xattrs.c.
Adds testsuite/chmod-symlink-race.test (with t_chmod_secure helper)
as regression coverage.
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
The sender's file open was vulnerable to the same TOCTOU symlink
race as the receiver-side basis-file open. change_pathname() calls
chdir() into subdirectories, which follows symlinks; an attacker
could race to swap a directory for a symlink between the chdir and
the file open, allowing reads of privileged files through the
daemon.
Reconstruct the full relative path (F_PATHNAME + fname) and open
via secure_relative_open() from the trusted module_dir, which
walks each path component without following symlinks. This is
independent of CWD, so the chdir race is neutralised.
CVE-2026-29518.
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
CVE-2026-29518: an rsync daemon configured with "use chroot = no"
is exposed to a TOCTOU race on parent path components. A local
attacker with write access to a module can replace a parent
directory component with a symlink between the receiver's check
and its open(), redirecting reads (basis-file disclosure) and
writes (file overwrite) outside the module. Under elevated daemon
privilege this allows privilege escalation. Default
"use chroot = yes" is not exposed.
Add secure_relative_open() in syscall.c. It walks the parent
components under RESOLVE_BENEATH (Linux 5.6+) /
O_RESOLVE_BENEATH (FreeBSD 13+, macOS 15+) / per-component
O_NOFOLLOW elsewhere, anchored at a trusted dirfd, so a parent-
symlink swap is rejected by the kernel. Route the receiver's
basis-file open in receiver.c through it when use_secure_symlinks
is set in clientserver.c rsync_module().
Reporters: Nullx3D (Batuhan SANCAK); Damien Neil; Michael Stapelberg.
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
The default python3 on AlmaLinux 8 is 3.6, but runtests.py uses
subprocess.run(capture_output=...) and check_output(text=...) which
were introduced in 3.7. Install the python39 module stream and point
/usr/bin/python3 at it via alternatives so the existing shebang
resolves correctly.
Reproduced as: TypeError: __init__() got an unexpected keyword
argument 'capture_output' at runtests.py line 75.
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
The intent is to validate that future security fixes still build and
test cleanly on the oldest still-supported LTS releases of the two
mainstream Linux families, so backports can be developed against the
same CI surface as the trunk:
- ubuntu-22.04: oldest GitHub Actions runner image still available
(20.04 was retired in April 2025). Mirrors the existing
ubuntu-build.yml step list.
- almalinux-8: RHEL 8 rebuild, full support until 2029. Runs in an
almalinux:8 container on ubuntu-latest because GHA has no native
runner for the Fedora/RHEL family. Pulls libzstd/xxhash/lz4 dev
headers from PowerTools + EPEL; commonmark via pip for the man
page generator.
Both jobs follow the same paths-ignore convention as the other
workflows so a workflow-only change to one file won't fan out across
the whole CI matrix.
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
Use unshare with user namespace UID mapping to run the
protected-regular test without real root privileges. Falls back
to skipping if unshare or uidmap is not available.
Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
The test correctly skips on Cygwin (which lacks RESOLVE_BENEATH), but
the workflow's RSYNC_EXPECT_SKIPPED list still treats any change in
the skipped set as a CI failure. Add the new test name so the
skipped/got comparison matches.
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
secure_relative_open() has a kernel-enforced "stay below dirfd" path
on Linux 5.6+ (openat2 RESOLVE_BENEATH) and FreeBSD 13+ (openat
O_RESOLVE_BENEATH). On Solaris, OpenBSD, NetBSD, and Cygwin the code
falls back to the per-component O_NOFOLLOW walk, which by design
rejects every directory symlink in the path -- the very case this
test exercises. Mark the test skipped there rather than have it
fail with a known regression that's tracked separately.
macOS is intentionally not in the skip list: although it does not
have O_RESOLVE_BENEATH either, the test passes there in practice;
investigation of the underlying reason is left as follow-up.
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
FreeBSD and MacOS have O_RESOLVE_BENEATH as an openat() flag with the same
"must not escape dirfd" semantics as Linux's RESOLVE_BENEATH. The
kernel rejects ".." escapes, absolute symlinks, and symlinks whose
target lies outside dirfd, while still following symlinks that
resolve within it -- the same trade-off that fixes issue #715 on
Linux.
Add a parallel BSD path in secure_relative_open(), gated on
declared. Unlike Linux, BSD doesn't have the header/runtime split
where the symbol can exist without kernel support, so no runtime
fallback is needed: if the flag compiles in, the kernel honours it.
OpenBSD and NetBSD have no equivalent kernel primitive and continue
to use the existing per-component O_NOFOLLOW walk; issue #715
remains visible on those platforms (a userland resolver or
unveil(2)-based fence would be follow-up work).
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
The CVE fix in commit c35e283 made secure_relative_open() walk every
component of relpath with O_NOFOLLOW. That blocks every symlink in the
path, which is stricter than the threat model required: legitimate
directory symlinks within the destination tree (e.g. when using -K /
--copy-dirlinks) are also rejected, breaking delta transfers with
"failed verification -- update discarded". See issue #715.
On Linux 5.6+, openat2(RESOLVE_BENEATH | RESOLVE_NO_MAGICLINKS) gives
us exactly what we want: the kernel rejects any resolution that would
escape the starting directory (via "..", absolute paths, or symlinks
pointing outside dirfd) while still following symlinks that resolve
within it. /proc magic-links are blocked too.
Use openat2 first; fall back to the existing per-component O_NOFOLLOW
walk on ENOSYS (kernel < 5.6). The lexical "../" checks at the head
of the function are kept as defense in depth. The Linux gate is
plain #ifdef __linux__: the runtime ENOSYS fallback covers the only
case that actually matters (header present + old kernel), and any
Linux build environment without linux/openat2.h will fail with a
clear "no such file" error rather than silently disabling the
protection.
Verified manually that openat2(RESOLVE_BENEATH) blocks all four
escape patterns (absolute symlink, ../ symlink, lexical .., absolute
path) while allowing direct and within-tree symlinks. The new
testsuite/symlink-dirlink-basis.test (taken from PR #864 by Samuel
Henrique) exercises the issue #715 regression and passes; full
make check passes 47/47.
Test: testsuite/symlink-dirlink-basis.test (8 scenarios)
Fixes: https://github.com/RsyncProject/rsync/issues/715
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
The Solaris xls() function listed every entry in the file's xattr
directory, which on Solaris includes OS-managed SUNWattr_ro and
SUNWattr_rw pseudo-attributes. SUNWattr_rw embeds the file creation
time, so its bytes naturally differ between the source and destination
files, making the xattrs and xattrs-hlink tests fail with diffs that
have nothing to do with rsync.
Rsync's own listxattr wrapper already filters these out
(lib/sysxattrs.c), so the right fix is to filter them in the test
display too. Other platforms are unaffected because each has its own
xls() branch in the case statement.
With the test now actually passing on Solaris, drop the CI hack that
overwrote testsuite/xattrs.test with a skip stub.
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
Mirror the existing FreeBSD workflow for OpenBSD and NetBSD using
vmactions/openbsd-vm and vmactions/netbsd-vm so we get cross-BSD
coverage on push, PR, and the nightly schedule.
Also extend the FreeBSD and Solaris workflows to actually exercise the
test suite by running 'make check' after the build. The Linux, macOS,
and Cygwin jobs already did this.
The Solaris xattrs and xattrs-hlink tests are removed before 'make
check' because the Solaris SUNWattr_ro / SUNWattr_rw system attributes
leak into the test diff; that's a real rsync-on-Solaris issue to follow
up on, but skip the tests for now so the suite goes green.
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
When invoked directly (rather than via 'make check'), runtests.py
previously left the user with a wall of confusing "not found" errors
from inside individual test scripts if the CHECK_PROGS helpers had not
been built. Detect this up front and point the user at the make
target that builds them.
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
