Similar as the previous commit: the field was renamed by accident, and
we should revert to the old name to not break compat needlessly, and
because the old name actually made sense, given the underlying .transfer
file knob is AppStream=
The "isEnabled" name was changed from "enabled" post v261 by accident
(because a6f7ded859 replaced json
formatting via table formatter to a native json generator which was
introduced for varlink, which changed the names). THis was later made
official in #43603, but it should rather have changed things back to the
original name, in particular as the underling .feature file field
actually is called "Enabled=" and not "IsEnabled=".
We really should take care to generate the same names for Varlink JSON
data as for the data returned by the systemd-sysupdate cli.
a6f7ded859 ensured that, so that's great.
But at the same time I think the varlink names should be adjusted to the
old names, since varlink support for this was just added in this cycle.
Given that v262 is not tagged yet, let's now revert to the old name.
This adds a fallback to `blockdev_get_root`, to handle the case where
`/usr` is a sysext overlay. Additinal changes are:
- move `extension_overlay_block` to `extension-util.h`, and make it
generic to work in both confext and sysext cases (though only sysext is
actually used currently)
- add a usr-only integration test to exercise the failure
Closes#43565.
The `systemd-nspawn` manpage says:
The `--private-users-ownership=auto` option is implied if
`--private-users=pick` is used.
This is true for the CLI, but not when set in a `.nspawn` file.
Assuming these defaults should be the same across both.
This default was also not updated when `managed` was added by #35685.
As a result, this is a change in behavior for users who didn't set an
explicit `PrivateUsersOwnership=` value in their `.nspawn` file.
When an automount goes away, the tokens we answer are only those we have
read from the pipe. Requests the kernel queued after our last read stayed
blocked: closing the pipe does not wake them, the kernel only notices the
closed pipe and turns the autofs catatonic when yet another request fails
to reach us, and after the lazy unmount no request reaches it anymore. In
the normal stop path that window is one event loop iteration; for an
automount torn down at coldplug it spans the whole reload.
Set the autofs to catatonic state before letting go of it, which fails
everything still queued with ENOENT. Answer the known tokens first:
closing our end of the pipe before that made a racing request flip the
kernel into catatonic mode on its own, which discards the queue and
turns our answers into "token not found" errors.
Signed-off-by: Martin Pitt <martin@amutable.com>
When an automount unit ceases to exist across daemon-reload (observed:
because a generator no longer emits it), its deserialized state still
makes `automount_coldplug()` pick up the autofs pipe and resume the unit
as waiting/running. But nothing can serve that mount point anymore. The
kernel keeps blocking every process that touches the mount point until
PID 1 answers (which it won't), forever.
To fix this, fail pending requests and unmount the autofs when
coldplugging an automount unit that is not loaded anymore.
Fixes#43700
Signed-off-by: Martin Pitt <martin@amutable.com>
When an automount goes away, answering the requests the kernel already
queued is the only thing that unblocks the processes waiting on them, and
nothing retries it later. If the `EHOSTDOWN` calls fail, it leaves the
requesting processes hung with no trace in the journal.
Log the failures to make this easier to debug. Also move that teardown into
a helper function, the next commit needs it as well.
Signed-off-by: Martin Pitt <martin@amutable.com>
Releasing the requests the kernel queued goes through `open_ioctl_fd()`, which
asserts on the manager fd for /dev/autofs. That fd stays unset when
`open_dev_autofs()` fails while coldplugging an automount, while the
deserialized pipe and its tokens outlive it.
Report the failure to the caller instead of asserting/crashing PID 1.
Signed-off-by: Martin Pitt <martin@amutable.com>
For re-use in systemd-gpt-auto-generator. This also makes it generic
across sysext/confext - no planned use for this, but to avoid confusion
later when someone tries to use it on a confext.
This updates a couple of links talking about Varlink:
1. anything that refers to specific Varlink concepts now point to the
UAPI.20 spec
2. More general references to Varlink as a whole point to both the
varlink.org domain and the spec.
RFC 4861 section 6.3.7 states that a host should send at least one Router
Solicitation even if it receives an unsolicited Router Advertisement
before sending its first solicitation, and should stop soliciting only
after receiving a valid Router Advertisement with a non-zero Router Lifetime.
Keep the RS timer active until both conditions are met, and add a
regression test for a zero-lifetime RA arriving before the initial RS.
Fixes#43205.
The conversion to the OPTION macros in c3f4c1b8d6 ("journalctl:
convert to OPTION macros") replaced strv_copy_unless_empty() with
strv_copy().
This is problematic because the positional arguments are now an empty
strv rather than NULL when none are given, but `--{list,dump}-catalog`
rely on that.
Instead let's hand NULL back to them when there is nothing on the
command line.
Encrypted credentials produced by systemd-creds are stored as Base64.
Decode them before decrypting in read_credential_with_decryption(), while
leaving plaintext credentials unchanged and retaining their precedence.
Bound the decoded encrypted input and report oversized credentials
explicitly. Match the plaintext reader's string semantics by rejecting
embedded NUL bytes when no size output is requested, while preserving
binary payloads for callers that request their size.
Add regression coverage for missing and malformed credentials, plaintext
precedence, decoded size boundaries, single-line and wrapped Base64,
empty strings, and binary data. Keep test outputs and host secrets scoped
to private fixtures.
Fixes#43621
Otherwise, UBSAN complains when src is NULL even though the shift is zero:
../src/basic/compress.c:2074:29: runtime error: applying zero offset to null pointer
When a multi-byte delimiter straddles two reads, json_stream_parse()
only scans the bytes from the latest read. As a result, a QMP reply
whose "\r\n" terminator is split between reads is not recognized.
Include up to delimiter_size - 1 trailing bytes from the existing buffer
in the next scan, so delimiters spanning reads can be matched.
Add regression coverage for a split CRLF delimiter followed by another
message in the same read. Check that both objects are parsed correctly
and that no buffered input remains afterwards.
Component definitions may contain multiple Documentation= URLs, but the
components command previously displayed only the first URL in text output.
Use TABLE_STRV_WRAPPED to display all component documentation URLs in the
table, making the full documentation list visible to users.
Add integration test coverage for multiple Documentation= assignments.
The feature JSON payload switched to `id`, `isEnabled`, and `appstream`,
and made `documentation` an array of URLs.
`updatectl` still parsed legacy table-derived names and a single URL.
This broke feature JSON parsing and made output inconsistent.
Features without documentation could also cause a null dereference.
Align `updatectl` with the current schema and preserve all
documentation URLs across sysupdate, D-Bus, and Varlink payloads.
Propagate condition evaluation errors through `condition_test_list()`
and `feature_is_suggested()`, with consistent JSON error handling and
tests.
The `suggested` field is now optional and always a boolean when
present; it is omitted rather than set to an error string when
SuggestOn*= conditions cannot be evaluated.
Add `suggested` metadata and document the field transition.
Follow-up for 3e34ccf6df.
Follow-up for 75a247d4de.
Follow-up for 40e727c93b.
Fixes#43603
Make condition_test_list() return negative errno on condition
evaluation errors instead of coercing them to false. This allows
callers to distinguish between a condition that evaluated to false
and a condition that failed to evaluate.
Callers that only need a boolean result explicitly check for a
positive return value, preserving their current behavior.
Add comprehensive unit tests covering error propagation with trigger
and non-trigger conditions. Tests use uname_architecture() to get the
runtime architecture instead of native_architecture(), ensuring they
work correctly in environments where binaries run under architecture
emulation.
Standard Linux POSIX ACL tag values are 0x00, 0x01, 0x02, 0x04, 0x08,
0x10 and 0x20. ELEMENTSOF(tag_map) evaluates to 33, which is one past
the last valid index 32.
The json mode was returning the string `none` when unmerged and
an array for merged extensions. Return an empty array instead.
Instead of branching on the output format, always add an empty strv
for an unmerged hierarchy: the table code renders an empty strv as
the ersatz string ("-") in text mode and as an empty array in json
mode. This also changes the human-readable output from "none" to "-".
It also updated the TEST-50-DISSECT.sysext.sh test. Technically
this is a compatbility break but its also arguably a bugfix.
The PolicyPCR digest is not always the complete input to an offline
signature. PolicyAuthorize also includes the hashed policy reference.
This is now required for NVPCRs to work in the initrd.
Expose the complete data as "tbs" (to-be-signed) and use it when ukify
matches offline signatures. Keep "pol" as the compatibility fallback.
Fixes: https://github.com/systemd/systemd/issues/43635
Follow-up for f0f3687700
OCI pulls create .mstack machine images, which vmspawn can
discover via -M. Since vmspawn only supports directory and disk
images, this currently reaches assert_not_reached() and aborts.
Return EOPNOTSUPP with a readable error instead.
Fixes#41541
Split out of https://github.com/systemd/systemd/pull/42983
## Problem
`dns_packet_append_name()` fails with `-EEXIST` once a name that was
first appended at an offset beyond RFC 1035's 14-bit compression-pointer
range (0x4000) occurs again in the same packet: the fall-through that
re-emits the labels trips `hashmap_ensure_put()` on the existing map
entry, and the whole packet build fails — reported by the caller as, for
example, "Failed to build reply packet: File exists".
The map is meant to hold only offsets a pointer can actually reference.
Names past that range were being inserted anyway, so they were dead
weight that could never be pointed at, and re-encountering one turned
into an error.
## Fix
Keep names at offsets ≥ 0x4000 out of the compression map entirely.
"Mapped implies referenceable" becomes an invariant of the map rather
than something callers have to reason about, and the `-EEXIST` collision
cannot arise. A re-occurrence of such a name is emitted as labels, still
compressed from the first suffix that *is* mapped in range — which is
the only encoding RFC 1035 section 4.1.4 permits in that case, since a
pointer must address ≤ 0x3FFF.
## Impact
Any packet large enough to push a name past 16KiB hits this — it is not
mDNS-specific, though that is where it was found: a DNS-SD announcement
covering a few hundred published services crosses the threshold easily,
and every record after it fails to encode.
## Test
`test-dns-packet-append` gains coverage for the boundary: a name first
written past the pointer range and then repeated, the same for a full
resource record, the rollback path when an append fails mid-name, and
the map invariant itself (asserted over the whole map, so a future
change that inserts an unreferenceable offset is caught rather than
silently reintroducing the bug).
## Testing status
Builds clean with `-Dwerror=true`; `meson test --suite resolve` passes
(19 ok, 1 skipped) at both commits.
dns_packet_append_name() appends a name's labels and then the root byte
that terminates it. A failure on that last byte returned directly,
leaving both the labels on the packet and the compression entries this
call had added -- entries naming offsets the caller then truncates away.
Take the same rollback path as every other failure in the function.
Independent of the map invariant fixed before it: this is reachable
whether or not any offset exceeded the pointer range.
The existing case only covered a fully-uncompressed re-occurrence. Every
DNS-SD name shares an in-range suffix, so also pin that a name first
seen beyond the pointer range still compresses from its mapped suffix --
one label plus a two-byte pointer per occurrence, map unchanged, and the
pointer addressing the suffix's prior occurrence rather than any
in-range octet.
A padding helper puts the next append at a chosen offset, for the two
sides of the boundary: at 0x3FFF the name is still mapped and re-occurs
as a pointer whose wire bytes are checked independently of the parser;
one past it, it is not mapped.
dns_packet_append_name() added every name it appends to the packet's
compression map, including names written at offsets >= 0x4000 that an
RFC 1035 compression pointer (14 bits) can never reference. Such
entries are not just useless -- when the same name occurs again, the
lookup finds it, the pointer cannot be emitted, the code falls through
to appending the labels again, and the re-insertion into the map then
fails the whole packet build with hashmap_ensure_put() returning
-EEXIST.
This is easily triggered with mDNS: an unsolicited DNS-SD announcement
covering a zone with a few hundred services exceeds 16KiB, and
dns_scope_announce() then fails with "Failed to build reply packet:
File exists".
This lifts the failure only: dns_scope_announce() still builds the whole
zone into one packet and hands it to the socket as a single datagram,
fragmented as the link requires. Splitting announcements across
MTU-sized packets, as RFC 6762 section 17 asks, is done separately.
Only insert a name into the map if its offset is pointer-expressible,
making "mapped implies referenceable" a map invariant. Re-occurrences of
out-of-range names are emitted as labels again -- the only encoding RFC
1035 section 4.1.4 permits in that case -- still compressed from the
first suffix that is mapped in range.
The lookup restates that range as control flow rather than asserting it.
Under -Db_ndebug=true, which CI builds, assert() is
__builtin_unreachable(), so an out-of-range entry would put 0xC000 | n
truncated to 14 bits on the wire -- a pointer to an arbitrary earlier
offset -- where falling through emits the labels instead. This layer is
shared with unicast DNS, DoT and LLMNR.
The pointer's tag bits and its offset bound are now named once, in
dns-def.h next to the other wire-format constants rather than in
dns-packet.h, since the format is RFC 1035's and not this packet
abstraction's -- dhcp-message.c open-codes the same two bits to parse
compressed names out of DHCP options. Compile-time checks pin that the
two partition the 16-bit field and that the tag is its top two bits,
which is what lets the reader take it from the first octet alone.
The offset's other bound becomes control flow along with it: it was an
assert(), which under -Db_ndebug=true is no check at all, for the bound
whose violation puts a pointer to the wrong bytes on the wire. It is
also tightened to the start of the name being appended rather than the
end of the packet, which is the "prior occurrence" RFC 1035 section
4.1.4 allows a pointer to address and what dns_packet_read_name()
enforces on the way in.
Add regression tests, including one through dns_packet_append_rr() --
the path the reported failure came from, where the name append is nested
inside the rdlength backpatch and two further rollbacks -- and one for
the reader's label-type dispatch, which the renamed constants touch.
Fixes#43645.
`FailureAction=exit` propagates the exit status of the unit's main
process. When there is none, `emergency_action()` left `m->return_value`
at its initial 0, so the manager exited 0 for a unit that failed —
indistinguishable from an orderly shutdown to whoever supervises it.
machinectl clone --read-only ignored the requested read-only mode when
cloning a foreign-UID-owned directory because the unprivileged clone path
did not receive the read_only argument.
Pass the flag through and apply the immutable bit via the already-open
destination directory fd after copying. Keep this best-effort, matching
the ordinary directory clone fallback, so failure to set the attribute
does not turn a completed copy into a reported clone failure.
Add a regression test using foreign-0 rather than a fixed UID. Verify
the cloned image's ReadOnly property, copied contents and ownership, and
that the immutable directory rejects new entries. Run this subtest only
when file attributes and the existing rootless nspawn prerequisites are
available, leaving the other machined tests unchanged.
unprivileged_clone() returns success when mountfsd_make_directory() fails.
Consequently, callers can report a successful clone even though the
destination directory was never created.
Return the original error instead of zero so the failed clone is
reported to the caller.
Add a regression test that places a regular file without an image suffix
at the destination path. Such a file is ignored by image discovery but
prevents directory creation. Check that cloning reports failure without
changing the file, then succeeds after the conflicting file is removed.
Vacuuming unlinks and deallocates archived journal files. A reader
may have already selected an entry from such a file, so final
validation can fail after candidate selection succeeded.
If the selected file is now unlinked, remove it and retry candidate
selection. Preserve errors from linked files, including the EIO
reported after SIGBUS recovery from a truncated mapping.
Lennart Poettering
Luca Boccassi
Daan De Meyer
Cathal Mullan
Armaan Sandhu
Kai Wohlfahrt
Martin Pitt
quanyeyang
Chris Down
Siteshwar Vashisht
Totoro
Yu Watanabe
Umut Tezduyar Lindskog
Zbigniew Jędrzejewski-Szmek
dongshengyuan
Michael Vogt
zouge666
Christian Glombek
Michael Renner