Martin Pitt [Fri, 30 Sep 2016 07:30:08 +0000 (09:30 +0200)]
resolved: don't query domain-limited DNS servers for other domains (#3621)
DNS servers which have route-only domains should only be used for
the specified domains. Routing queries about other domains there is a privacy
violation, prone to fail (as that DNS server was not meant to be used for other
domains), and puts unnecessary load onto that server.
Introduce a new helper function dns_server_limited_domains() that checks if the
DNS server should only be used for some selected domains, i. e. has some
route-only domains without "~.". Use that when determining whether to query it
in the scope, and when writing resolv.conf.
Extend the test_route_only_dns() case to ensure that the DNS server limited to
~company does not appear in resolv.conf. Add test_route_only_dns_all_domains()
to ensure that a server that also has ~. does appear in resolv.conf as global
name server. These reproduce #3420.
Add a new test_resolved_domain_restricted_dns() test case that verifies that
domain-limited DNS servers are only being used for those domains. This
reproduces #3421.
Clarify what a "routing domain" is in the manpage.
pid1: process zero-length notification messages again
This undoes 531ac2b234. I acked that patch without looking at the code
carefully enough. There are two problems:
- we want to process the fds anyway
- in principle empty notification messages are valid, and we should
process them as usual, including logging using log_unit_debug().
pid1: don't return any error in manager_dispatch_notify_fd() (#4240)
If manager_dispatch_notify_fd() fails and returns an error then the handling of
service notifications will be disabled entirely leading to a compromised system.
For example pid1 won't be able to receive the WATCHDOG messages anymore and
will kill all services supposed to send such messages.
coredump,catalog: give better notice when a core file is truncated
coredump had code to check if copy_bytes() hit the max_bytes limit,
and refuse further processing in that case.
But in 84ee0960443, the return convention for copy_bytes() was changed
from -EFBIG to 1 for the case when the limit is hit, so the condition
check in coredump couldn't ever trigger.
But it seems that *do* want to process such truncated cores [1].
So change the code to detect truncation properly, but instead of
returning an error, give a nice log entry.
coredumpctl: delay the "on tty" refusal until as late as possible
For the user, if the core file is missing or inaccessible, it is
more interesting that the fact that they forgot to pipe to a file.
So delay the failure from the check until after we have verified
that the file or the COREDUMP field are present.
Partially fixes #4161.
Also, error reporting on failure was duplicated. save_core() now
always prints an error message (because it knows the paths involved,
so can the most useful message), and the callers don't have to.
Propagate errors properly, so that if we hit oom or an error in the
journal, the whole command will fail. This is important when using
the output in scripts.
Support the output of multiple values for the same field with -F.
The journal supports that, and our official commands should too, as
far as it makes sense. -F can be used to print user-defined fields
(e.g. somebody could use a TAG field with multiple occurences), so
we should support that too. That seems better than silently printing
the last value found as was done before.
We would iterate trying to match the same field with all possible
field names. Once we find something, cut the loop short, since we
know that nothing else can match.
The column for "present" was easy to miss, especially if somebody had no
coredumps present at all, in which case the column of spaces of width one
wasn't visually distinguished from the neighbouring columns. Replace this
with an explicit text, one of: "missing", "journal", "present", "error".
$ coredumpctl
TIME PID UID GID SIG COREFILE EXE
Mon 2016-09-26 22:46:31 CEST 8623 0 0 11 missing /usr/bin/bash
Mon 2016-09-26 22:46:35 CEST 8639 1001 1001 11 missing /usr/bin/bash
Tue 2016-09-27 01:10:46 CEST 16110 1001 1001 11 journal /usr/bin/bash
Tue 2016-09-27 01:13:20 CEST 16290 1001 1001 11 journal /usr/bin/bash
Tue 2016-09-27 01:33:48 CEST 17867 1001 1001 11 present /usr/bin/bash
Tue 2016-09-27 01:37:55 CEST 18549 0 0 11 error /usr/bin/bash
Also, use access(…, R_OK), so that we can report a present but inaccessible
file different than a missing one.
In 'list', show present also for coredumps stored in the journal.
In 'status', replace "File" with "Storage" line that is always present.
Possible values:
Storage: none
Storage: journal
Storage: /path/to/file (inacessible)
Storage: /path/to/file
Previously the File field be only present if the file was accessible, so users
had to manually extract the file name precisely in the cases where it was
needed, i.e. when coredumpctl couldn't access the file. It's much more friendly
to always show something. This output is designed for human consumption, so
it's better to be a bit verbose.
The call to sd_j_set_data_threshold is moved, so that status is always printed
with the default of 64k, list uses 4k, and coredump retrieval is done with the
limit unset. This should make checking for the presence of the COREDUMP field
not too costly.
Added in 9fe13294a9 (by me :[```), and later obfuscated in d0c8806d4ab, if an
uncompressed external file or an internally stored coredump was supposed to be
written to a file descriptor, nothing would be written.
Back when external storage was initially added in 34c10968cb, this mode of
storage was added. This could have made some sense back when XZ compression was
used, and an uncompressed core on disk could be used as short-lived cache file
which does require costly decompression. But now fast LZ4 compression is used
(by default) both internally and externally, so we have duplicated storage,
using the same compression and same default maximum core size in both cases,
but with different expiration lifetimes. Even the uncompressed-external,
compressed-internal mode is not very useful: for small files, decompression
with LZ4 is fast enough not to matter, and for large files, decompression is
still relatively fast, but the disk-usage penalty is very big.
An additional problem with the two modes of storage is that it complicates
the code and makes it much harder to return a useful error message to the user
if we cannot find the core file, since if we cannot find the file we have to
check the internal storage first.
This patch drops "both" storage mode. Effectively this means that if somebody
configured coredump this way, they will get a warning about an unsupported
value for Storage, and the default of "external" will be used.
I'm pretty sure that this mode is very rarely used anyway.
Paweł Szewczyk [Mon, 26 Sep 2016 16:45:47 +0000 (18:45 +0200)]
core: Fix USB functionfs activation and clarify its documentation (#4188)
There was no certainty about how the path in service file should look
like for usb functionfs activation. Because of this it was treated
differently in different places, which made this feature unusable.
This patch fixes the path to be the *mount directory* of functionfs, not
ep0 file path and clarifies in the documentation that ListenUSBFunction should be
the location of functionfs mount point, not ep0 file itself.
journald,ratelimit: fix wrong calculation of burst_modulate() (#4218)
This patch fixes wrong calculation of burst_modulate(), which now calculates
the values smaller than really expected ones if available disk space is
strictly more than 1MB.
In particular, if available disk space is strictly more than 1MB and strictly
less than 16MB, the resulted value becomes smaller than its original one.
>>> (math.log2(1*1024**2)-16) / 4
1.0
>>> (math.log2(16*1024**2)-16) / 4
2.0
>>> (math.log2(256*1024**2)-16) / 4
3.0
→ This matches the comment in the function.
Matej Habrnal [Mon, 26 Sep 2016 15:28:58 +0000 (17:28 +0200)]
coredump: initialize coredump_size in submit_coredump() (#4219)
If ulimit is smaller than page_size(), function save_external_coredump()
returns -EBADSLT and this causes skipping whole core dumping part in
submit_coredump(). Initializing coredump_size to UINT64_MAX prevents
evaluating a condition with uninitialized varialbe which leads to
calling allocate_journal_field() with coredump_fd = -1 which causes
aborting.
Signed-off-by: Matej Habrnal <mhabrnal@redhat.com>
ProtectSystem= with all its different modes and other options like
PrivateDevices= + ProtectKernelTunables= + ProtectHome= are orthogonal,
however currently it's a bit hard to parse that from the implementation
view. Simplify it by giving each mode its own table with all paths and
references to other Protect options.
With this change some entries are duplicated, but we do not care since
duplicate mounts are first sorted by the most restrictive mode then
cleaned.
core:sandbox: add more /proc/* entries to ProtectKernelTunables=
Make ALSA entries, latency interface, mtrr, apm/acpi, suspend interface,
filesystems configuration and IRQ tuning readonly.
Most of these interfaces now days should be in /sys but they are still
available through /proc, so just protect them. This patch does not touch
/proc/net/...
Move out mount calculation on its own function. Actually the logic is
smart enough to later drop nop and duplicates mounts, this change
improves code readability.
---
src/core/namespace.c | 47 ++++++++++++++++++++++++++++++++++++-----------
1 file changed, 36 insertions(+), 11 deletions(-)
core:namespace: put paths protected by ProtectKernelTunables= in
Instead of having all these paths everywhere, put the ones that are
protected by ProtectKernelTunables= into their own table. This way it
is easy to add paths and track which ones are protected.
execute: filter low-level I/O syscalls if PrivateDevices= is set
If device access is restricted via PrivateDevices=, let's also block the
various low-level I/O syscalls at the same time, so that we know that the
minimal set of devices in our virtualized /dev are really everything the unit
can access.
units: further lock down our long-running services
Let's make this an excercise in dogfooding: let's turn on more security
features for all our long-running services.
Specifically:
- Turn on RestrictRealtime=yes for all of them
- Turn on ProtectKernelTunables=yes and ProtectControlGroups=yes for most of
them
- Turn on RestrictAddressFamilies= for all of them, but different sets of
address families for each
Also, always order settings in the unit files, that the various sandboxing
features are close together.
Add a couple of missing, older settings for a numbre of unit files.
Note that this change turns off AF_INET/AF_INET6 from udevd, thus effectively
turning of networking from udev rule commands. Since this might break stuff
(that is already broken I'd argue) this is documented in NEWS.
Let's merge a couple of columns, to make the table a bit shorter. This
effectively just drops whitespace, not contents, but makes the currently
humungous table much much more compact.
man: rework documentation for ReadOnlyPaths= and related settings
This reworks the documentation for ReadOnlyPaths=, ReadWritePaths=,
InaccessiblePaths=. It no longer claims that we'd follow symlinks relative to
the host file system. (Which wasn't true actually, as we didn't follow symlinks
at all in the most recent releases, and we know do follow them, but relative to
RootDirectory=).
This also replaces all references to the fact that all fs namespacing options
can be undone with enough privileges and disable propagation by a single one in
the documentation of ReadOnlyPaths= and friends, and then directs the read to
this in all other places.
Moreover a hint is added to the documentation of SystemCallFilter=, suggesting
usage of ~@mount in case any of the fs namespacing related options are used.
man: in user-facing documentaiton don't reference C function names
Let's drop the reference to the cap_from_name() function in the documentation
for the capabilities setting, as it is hardly helpful. Our readers are not
necessarily C hackers knowing the semantics of cap_from_name(). Moreover, the
strings we accept are just the plain capability names as listed in
capabilities(7) hence there's really no point in confusing the user with
anything else.
namespace: chase symlinks for mounts to set up in userspace
This adds logic to chase symlinks for all mount points that shall be created in
a namespace environment in userspace, instead of leaving this to the kernel.
This has the advantage that we can correctly handle absolute symlinks that
shall be taken relative to a specific root directory. Moreover, we can properly
handle mounts created on symlinked files or directories as we can merge their
mounts as necessary.
(This also drops the "done" flag in the namespace logic, which was never
actually working, but was supposed to permit a partial rollback of the
namespace logic, which however is only mildly useful as it wasn't clear in
which case it would or would not be able to roll back.)
execute: drop group priviliges only after setting up namespace
If PrivateDevices=yes is set, the namespace code creates device nodes in /dev
that should be owned by the host's root, hence let's make sure we set up the
namespace before dropping group privileges.
core: imply ProtectHome=read-only and ProtectSystem=strict if DynamicUser=1
Let's make sure that services that use DynamicUser=1 cannot leave files in the
file system should the system accidentally have a world-writable directory
somewhere.
This effectively ensures that directories need to be whitelisted rather than
blacklisted for access when DynamicUser=1 is set.
Let's tighten our sandbox a bit more: with this change ProtectSystem= gains a
new setting "strict". If set, the entire directory tree of the system is
mounted read-only, but the API file systems /proc, /dev, /sys are excluded
(they may be managed with PrivateDevices= and ProtectKernelTunables=). Also,
/home and /root are excluded as those are left for ProtectHome= to manage.
In this mode, all "real" file systems (i.e. non-API file systems) are mounted
read-only, and specific directories may only be excluded via
ReadWriteDirectories=, thus implementing an effective whitelist instead of
blacklist of writable directories.
While we are at, also add /efi to the list of paths always affected by
ProtectSystem=. This is a follow-up for b52a109ad38cd37b660ccd5394ff5c171a5e5355 which added /efi as alternative for
/boot. Our namespacing logic should respect that too.
Previously, if ReadWritePaths= was nested inside a ReadOnlyPaths=
specification, then we'd first recursively apply the ReadOnlyPaths= paths, and
make everything below read-only, only in order to then flip the read-only bit
again for the subdirs listed in ReadWritePaths= below it.
This is not only ugly (as for the dirs in question we first turn on the RO bit,
only to turn it off again immediately after), but also problematic in
containers, where a container manager might have marked a set of dirs read-only
and this code will undo this is ReadWritePaths= is set for any.
With this patch behaviour in this regard is altered: ReadOnlyPaths= will not be
applied to the children listed in ReadWritePaths= in the first place, so that
we do not need to turn off the RO bit for those after all.
This means that ReadWritePaths=/ReadOnlyPaths= may only be used to turn on the
RO bit, but never to turn it off again. Or to say this differently: if some
dirs are marked read-only via some external tool, then ReadWritePaths= will not
undo it.
This is not only the safer option, but also more in-line with what the man page
currently claims:
"Entries (files or directories) listed in ReadWritePaths= are
accessible from within the namespace with the same access rights as
from outside."
To implement this change bind_remount_recursive() gained a new "blacklist"
string list parameter, which when passed may contain subdirs that shall be
excluded from the read-only mounting.
A number of functions are updated to add more debug logging to make this more
digestable.
execute: move suppression of HOME=/ and SHELL=/bin/nologin into user-util.c
This adds a new call get_user_creds_clean(), which is just like
get_user_creds() but returns NULL in the home/shell parameters if they contain
no useful information. This code previously lived in execute.c, but by
generalizing this we can reuse it in run.c.
systemctl: suppress errors with "show" for nonexistent units and properties
Show is documented to be program-parseable, and printing the warning about
about a non-existent unit, while useful for humans, broke a lot of scripts.
Restore previous behaviour of returning success and printing empty or useless
stuff for units which do not exist, and printing empty values for properties
which do not exists.
With SYSTEMD_LOG_LEVEL=debug, hints are printed, but the return value is
still 0.
We might consider adding an explicit switch to fail on missing units/properties
(e.g. --ensure-exists or similar), and make -P foobar equivalent to
--ensure-exists --property=foobar.
sysctl: configure kernel parameters in the order they occur in each sysctl configuration files (#4205)
Currently, systemd-sysctl command configures kernel parameters in each sysctl
configuration files in random order due to characteristics of iterator of
Hashmap.
However, kernel parameters need to be configured in the order they occur in
each sysctl configuration files.
- For example, consider fs.suid_coredump and kernel.core_pattern. If
fs.suid_coredump=2 is configured before kernel.core_pattern= whose default
value is "core", then kernel outputs the following message:
Unsafe core_pattern used with suid_dumpable=2. Pipe handler or fully qualified core dump path required.
Note that the security issue mentioned in this message has already been fixed
on recent kernels, so this is just a warning message on such kernels. But
it's still confusing to users that this message is output on some boot and
not output on another boot.
- I don't know but there could be other kernel parameters that are significant
in the order they are configured.
- The legacy sysctl command configures kernel parameters in the order they
occur in each sysctl configuration files. Although I didn't find any official
specification explaining this behavior of sysctl command, I don't think there
is any meaningful reason to change this behavior, in particular, to the
random one.
This commit does the change by simply using OrderedHashmap instead of Hashmap.
Luca Bruno [Sat, 24 Sep 2016 12:30:42 +0000 (12:30 +0000)]
nspawn: decouple --boot from CLONE_NEWIPC (#4180)
This commit is a minor tweak after the split of `--share-system`, decoupling the `--boot`
option from IPC namespacing.
Historically there has been a single `--share-system` option for sharing IPC/PID/UTS with the
host, which was incompatible with boot/pid1 mode. After the split, it is now possible to express
the requirements with better granularity.
For reference, this is a followup to #4023 which contains references to previous discussions.
I realized too late that CLONE_NEWIPC is not strictly needed for boot mode.
journal: fix HMAC calculation when appending a data object
Since commit 5996c7c295e073ce21d41305169132c8aa993ad0 (v190 !), the
calculation of the HMAC is broken because the hash for a data object
including a field is done in the wrong order: the field object is
hashed before the data object is.
However during verification, the hash is done in the opposite order as
objects are scanned sequentially.
NSS modules (libnss_*.so.*) need to be installed into
${rootlibdir} (typically /lib) in order to be used. Previously, the
modules were installed into ${libdir}, thus usually ending up in
/usr/lib, even on systems where split usr is enabled, or ${libdir} is
passed explicitly.
If Kind is not specied, the message about "Invalid Kind" was misleading.
If Kind was specified in an invalid way, we get a message in the parsing
phase anyway. Reword the message to cover both cases better.
man: mention that netdev,network files support dropins
Also update the description of drop-ins in systemd.unit(5) to say that .d
directories, not .conf files, are in /etc/system/system, /run/systemd/system,
etc.