Lennart Poettering [Tue, 19 Jan 2016 16:19:14 +0000 (17:19 +0100)]
resolved: rename a few props to closer match ther counterparts in the various configuration files
Lennart Poettering [Tue, 19 Jan 2016 16:16:12 +0000 (17:16 +0100)]
resolved: add bus API for configuring per-link DNS settings
This is useful for alternative network management solutions (such as NetworkManager) to push DNS configuration data
into resolved.
The calls will fail should networkd already have taken possesion of a link, so that the bus API is only available if
we don't get the data from networkd.
Lennart Poettering [Tue, 19 Jan 2016 16:15:02 +0000 (17:15 +0100)]
resolved: add a couple of errors to the error mapping tables
These were previously forgotten, add them now.
Lennart Poettering [Tue, 19 Jan 2016 16:13:27 +0000 (17:13 +0100)]
resolved: allocate DNS scope for links only if the interface is up
For mDNS and LLMNR we already created the scopes only if the specific interfaces where actually up and suitable for
Multicasting. Add a similar (but weaker) logic for unicast DNS as well.
Lennart Poettering [Tue, 19 Jan 2016 16:12:12 +0000 (17:12 +0100)]
resolve-host: support --interface= as long form for -i
Lennart Poettering [Tue, 19 Jan 2016 16:11:28 +0000 (17:11 +0100)]
resolve-host: show whether DNSSEC is supported or not in --statistics output
This should be generally useful information, hence show it.
Lennart Poettering [Tue, 19 Jan 2016 16:10:34 +0000 (17:10 +0100)]
resolve-host: also show mDNS as source of resolving
Lennart Poettering [Tue, 19 Jan 2016 14:11:10 +0000 (15:11 +0100)]
Merge pull request #2369 from zonque/resolved
resolved: hide public mDNS configuration knobs for now
Lennart Poettering [Tue, 19 Jan 2016 14:09:53 +0000 (15:09 +0100)]
Merge pull request #2357 from keszybz/warnings-2
Remove gcc warnings v2
Daniel Mack [Tue, 19 Jan 2016 13:05:01 +0000 (14:05 +0100)]
resolved: remove configuration knobs for mDNS until it's ready
These bits were intenionally left out while mDNS is under development.
Remove the exposed knobs and man page entries again until this is settled.
Daniel Mack [Tue, 19 Jan 2016 12:58:55 +0000 (13:58 +0100)]
resolved: fix mDNS IPv6 multicast address
Fixes #2366
Tom Gundersen [Tue, 19 Jan 2016 00:44:20 +0000 (01:44 +0100)]
Merge pull request #2358 from poettering/dnssec16
sixteenth DNSSEC patch set
Lennart Poettering [Mon, 18 Jan 2016 23:51:26 +0000 (00:51 +0100)]
resolved: don't forget about lost OPT and RRSIG when downgrading a feature level
Certain Belkin routers appear to implement a broken DNS cache for A RRs and some others, but implement a pass-thru for
AAAA RRs. This has the effect that we quickly recognize the broken logic of the router when we do an A lookup, but for
AAAA everything works fine until we actually try to validate the request. Given that the validation will necessarily
fail ultimately let's make sure we remember even when downgrading a feature level that OPT or RRSIG was missing.
Lennart Poettering [Mon, 18 Jan 2016 21:45:18 +0000 (22:45 +0100)]
update DNSSEC TODO
Lennart Poettering [Mon, 18 Jan 2016 22:29:04 +0000 (23:29 +0100)]
resolved: don't try to print error strings, where errno isn't set
Lennart Poettering [Mon, 18 Jan 2016 22:27:16 +0000 (23:27 +0100)]
resolved: when restarting a transaction pick a new ID
When we restart a transaction because of an incompatible server, pick a new transaction ID.
This should increase compatibility with DNS servers that don't like if they get different requests with the same
transaction ID.
Lennart Poettering [Mon, 18 Jan 2016 22:15:35 +0000 (23:15 +0100)]
resolved: enforce maximum limit on DNS transactions
given that DNSSEC lookups may result in quite a number of auxiliary transactions, let's better be safe than sorry and
also enforce a limit on the number of total transactions, not just on the number of queries.
Lennart Poettering [Mon, 18 Jan 2016 21:43:21 +0000 (22:43 +0100)]
resolved: add DNAME test case to the complex DNSSEC test
Lennart Poettering [Mon, 18 Jan 2016 21:36:58 +0000 (22:36 +0100)]
resolved: fix how we detect whether auxiliary DNSSEC transactions are ready
Previously, when getting notified about a completed auxiliary DNSSEC transaction we'd immediately act on it, and
possibly abort the main transaction. This is problematic, as DNS transactions that already completed at the time we
started using them will never get the notification event, and hence never be acted on in the same way.
Hence, introduce a new call dns_transaction_dnssec_ready() that checks the state of auxiliary DNSSEC transactions, and
returns 1 when we are ready for the actual DNSSEC validation step. Then, make sure this is invoked when the auxiliary
transactions are first acquired (and thus possibly reused) as well when the notifications explained above take place.
This fixes problems particularly when doing combined A and AAAA lookups where the auxiliary DNSSEC transactions get
reused between them, and where we got confused if we reused an auxiliary DNSSEC transaction from one when it already
got completed from the other.
Lennart Poettering [Mon, 18 Jan 2016 21:34:41 +0000 (22:34 +0100)]
resolved: end log messages in a full stop
Lennart Poettering [Mon, 18 Jan 2016 21:33:23 +0000 (22:33 +0100)]
resolved: never consider following a CNAME/DNAME chain for a CNAME/DNAME lookup
Let's avoid thinking that a CNAME/DNAME chain traversal could be a good idea if QTYPE is already CNAME/DNAME.
(Also, let's bail out early when trying to see if some RR is a suitable CNAME/DNAME for some other RR).
Lennart Poettering [Mon, 18 Jan 2016 20:31:16 +0000 (21:31 +0100)]
resolved: when following a CNAME, turn off search domains
If the first step was done via a search domain, make sure the subsequent steps are not.
Lennart Poettering [Mon, 18 Jan 2016 20:30:45 +0000 (21:30 +0100)]
resolved: properly reset old collected data when following a CNAME redirect
Lennart Poettering [Mon, 18 Jan 2016 20:02:00 +0000 (21:02 +0100)]
resolved: beef up complex dnssec test to also use ResolveAddress() and do IDNA checks
Lennart Poettering [Mon, 18 Jan 2016 19:31:39 +0000 (20:31 +0100)]
resolved: rework IDNA logic
Move IDNA logic out of the normal domain name processing, and into the bus frontend calls. Previously whenever
comparing two domain names we'd implicitly do IDNA conversion so that "pöttering.de" and "xn--pttering-n4a.de" would be
considered equal. This is problematic not only for DNSSEC, but actually also against he IDNA specs.
Moreover it creates problems when encoding DNS-SD services in classic DNS. There, the specification suggests using
UTF8 encoding for the actual service name, but apply IDNA encoding to the domain suffix.
With this change IDNA conversion is done only:
- When the user passes a non-ASCII hostname when resolving a host name using ResolveHostname()
- When the user passes a non-ASCII domain suffix when resolving a service using ResolveService()
No IDNA encoding is done anymore:
- When the user does raw ResolveRecord() RR resolving
- On the service part of a DNS-SD service name
Previously, IDNA encoding was done when serializing names into packets, at a point where information whether something
is a label that needs IDNA encoding or not was not available, but at a point whether it was known whether to generate a
classic DNS packet (where IDNA applies), or an mDNS/LLMNR packet (where IDNA does not apply, and UTF8 is used instead
for all host names). With this change each DnsQuery object will now maintain two copies of the DnsQuestion to ask: one
encoded in IDNA for use with classic DNS, and one encoded in UTF8 for use with LLMNR and MulticastDNS.
Lennart Poettering [Mon, 18 Jan 2016 19:22:45 +0000 (20:22 +0100)]
resolved: minor optimization for dns_question_is_equal()
If the poinetrs are equal, we don't have to do a deep comparison.
This is similar to a similar optimization we already have in place for RRs and keys.
Lennart Poettering [Mon, 18 Jan 2016 19:21:55 +0000 (20:21 +0100)]
resolved: be slightly stricter when validating DnsQuestion
Also verify whether the DNS RR types are actually suitable for a question.
Lennart Poettering [Mon, 18 Jan 2016 19:21:30 +0000 (20:21 +0100)]
resolved: make key argument of dns_question_contains() const
Lennart Poettering [Mon, 18 Jan 2016 19:18:28 +0000 (20:18 +0100)]
resolved add dns_name_apply_idna() to convert a domain name into its IDNA equivalent
Tom Gundersen [Mon, 18 Jan 2016 22:10:53 +0000 (23:10 +0100)]
Merge pull request #2326 from poettering/dnssec15
Fifteenth batch of DNSSEC patches
Zbigniew Jędrzejewski-Szmek [Fri, 15 Jan 2016 19:58:50 +0000 (14:58 -0500)]
journal-remote: update to 64bit µhttp api, provide fallback
64 bit offset is now accepted, which is nice. The old function is
deprecated, and generates a compile time warning when used. We only
use an offset of 0, so we really don't care. Adapt to use the new
function, but fall back to the old one on older versions.
Zbigniew Jędrzejewski-Szmek [Fri, 15 Jan 2016 19:13:12 +0000 (14:13 -0500)]
journal-remote: fix warning about deprecated µhttpd macro
src/journal-remote/journal-remote.c:590:13: warning: Value MHD_HTTP_METHOD_NOT_ACCEPTABLE is deprecated, use MHD_HTTP_NOT_ACCEPTABLE
return mhd_respond(connection, MHD_HTTP_METHOD_NOT_ACCEPTABLE,
^
The new define was added in 0.9.38. Instead of requiring the new
libmicrohttpd version, provide the fallback, it is trivial.
Zbigniew Jędrzejewski-Szmek [Wed, 13 Jan 2016 20:08:48 +0000 (15:08 -0500)]
resolve: fix return value from dns_packet_append_rr()
Zbigniew Jędrzejewski-Szmek [Wed, 13 Jan 2016 02:02:56 +0000 (21:02 -0500)]
udevd: initialize fds to -1 and close them before exiting
Little change in practice, because the program will exit soon
afterwards, but the standard style of closing all fds is now followed.
Also gets rid of gcc warning about fd_ctrl and fd_uevent being
unitialized.
Zbigniew Jędrzejewski-Szmek [Tue, 12 Jan 2016 22:42:52 +0000 (17:42 -0500)]
journald: do not free uninitialized pointer in error path
Zbigniew Jędrzejewski-Szmek [Wed, 13 Jan 2016 03:02:18 +0000 (22:02 -0500)]
logind: simplify job variable handling
manager_{start,stop}_{slice,scope,unit} functions had an optional job
output parameter. But all callers specified job, so make the parameter
mandatory, add asserts. Also extract common job variable handling to
a helper function to avoid duplication.
Avoids gcc warning about job being unitialized.
Zbigniew Jędrzejewski-Szmek [Tue, 12 Jan 2016 22:27:33 +0000 (17:27 -0500)]
logind: do not use an uninitialized variable
We requested various fields using SD_BUS_CREDS_AUGMENT but at least
sd_bus_creds_get_tty can fail with ENXIO, not setting the output variable.
Zbigniew Jędrzejewski-Szmek [Fri, 25 Dec 2015 21:59:44 +0000 (16:59 -0500)]
sd-event: fix theoretical leak on error in event_make_signal_data
Zbigniew Jędrzejewski-Szmek [Wed, 13 Jan 2016 01:22:30 +0000 (20:22 -0500)]
sd-ndisc: simplify if stmt and reindent log messages
Merge separate two error handling statements into two nested ifs.
This looks cleaner, and avoids a gcc warning about *prefix being
uninitialized.
While at it, fix identation of logging statements elsewhere in the
file.
Zbigniew Jędrzejewski-Szmek [Mon, 11 Jan 2016 22:26:57 +0000 (17:26 -0500)]
sd-ndisc: simplify ndisc_prefix_new
If the initial allocation succeeded, there is no way to
fail, so cleanup function is not necessary.
Zbigniew Jędrzejewski-Szmek [Tue, 12 Jan 2016 22:42:06 +0000 (17:42 -0500)]
sd-device: initialize variables to avoid warning
The code is correct, assuming that the kernel does not feed
us garbled data. Let's initialize those variables to avoid the
warning anyway.
Zbigniew Jędrzejewski-Szmek [Mon, 11 Jan 2016 17:42:42 +0000 (12:42 -0500)]
bus-kernel: reword assignment of dst_id in bus_message_setup_kmsg
Setting of dst_id was based on interplay of two booleans,
making the logic hard to follow (for humans and compilers alike).
gcc was confused and emmitted a warning about an uninitialized
variable. Rework the code to make it obvious that dst_id is
set properly.
Zbigniew Jędrzejewski-Szmek [Mon, 11 Jan 2016 17:16:02 +0000 (12:16 -0500)]
sd-netlink: remove unused table
Zbigniew Jędrzejewski-Szmek [Fri, 25 Dec 2015 21:29:09 +0000 (16:29 -0500)]
basic: add missing word in comment
Zbigniew Jędrzejewski-Szmek [Tue, 12 Jan 2016 20:36:57 +0000 (15:36 -0500)]
basic/escape: merge utf8 and non-utf8 paths in cunescape_one
Not every byte sequence is valid utf8. We allow escaping of non-utf8
sequences in strings by using octal and hexadecimal escape sequences
(\123 and \0xAB) for bytes at or above 128. Users of cunescape_one
could infer whether such use occured when they received an answer
between 128 and 256 in *ret (a non-ascii one byte character). But this
is subtle and misleading: the comments were wrong, because ascii is a
subset of unicode, so c != 0 did not mean non-unicode, but rather
ascii-subset-of-unicode-or-raw-byte. This was all rather confusing, so
make the "single byte" condition explicit.
I'm not convinced that allowing non-utf8 sequences to be produced is
useful in all cases where we allow it (e.g. in config files), but that
behaviour is unchanged, just made more explicit.
This also fixes an (invalid) gcc warning about unitialized variable
(*ret_unicode) in callers of cunescape_one.
Zbigniew Jędrzejewski-Szmek [Mon, 11 Jan 2016 19:37:30 +0000 (14:37 -0500)]
core/dbus-execute: do not needlessly compare strings again
gcc complains that dirs might be unitialized. It cannot, but
we just checked that name has one of three values above, so
no need to check again.
Zbigniew Jędrzejewski-Szmek [Wed, 13 Jan 2016 03:44:44 +0000 (22:44 -0500)]
test-bus-error: add tests for setting error == 0
Zbigniew Jędrzejewski-Szmek [Mon, 11 Jan 2016 22:19:25 +0000 (17:19 -0500)]
sd-event: check clock argument to sd_event_now()
sd_event_now() is a public function, so we must check all
arguments for validity. Update man page and add tests.
Sample debug message:
Assertion 'IN_SET(clock, CLOCK_REALTIME, CLOCK_REALTIME_ALARM, CLOCK_MONOTONIC, CLOCK_BOOTTIME, CLOCK_BOOTTIME_ALARM)' failed at src/libsystemd/sd-event/sd-event.c:2719, function sd_event_now(). Ignoring.
Zbigniew Jędrzejewski-Szmek [Wed, 13 Jan 2016 19:23:54 +0000 (14:23 -0500)]
bus-error: verify additional error maps during installation
Go over the entries in the map and check that they make sense.
Tests are added. In the future we might want to do additional
checks, e.g. verifying that the error names are in the expected
format.
Zbigniew Jędrzejewski-Szmek [Tue, 12 Jan 2016 22:19:51 +0000 (17:19 -0500)]
basic,bus-error: return negative error from errno_from_name
errno_from_name used an unusual return convention where 0 meant
"not found". This tripped up config_parse_syscall_errno(),
which would treat that as success. Return -EINVAL instead,
and adjust bus_error_name_to_errno() for the new convention.
Also remove a goto which was used as a simple if and clean
up surroudning code a bit.
Zbigniew Jędrzejewski-Szmek [Tue, 12 Jan 2016 22:25:23 +0000 (17:25 -0500)]
Use negative_errno() to assert errno is positive after a few system calls
This is not particularly intrusive because it happens in simple
utility functions. It helps gcc understand that error codes
are negative.
This gets a rid of most of the remaining warnings.
Daniel Mack [Mon, 18 Jan 2016 10:49:32 +0000 (11:49 +0100)]
Merge pull request #2352 from martinpitt/master
keymap: Add HP ProBook 440 G3
Martin Pitt [Mon, 18 Jan 2016 10:01:32 +0000 (11:01 +0100)]
keymap: Add HP ProBook 440 G3
Fixes #2343
Daniel Mack [Mon, 18 Jan 2016 09:53:49 +0000 (10:53 +0100)]
Merge pull request #2347 from aroig/gh/fix-udev-user-wants
Fix broken SYSTEMD_USER_WANTS in udev rules.
Daniel Mack [Mon, 18 Jan 2016 09:52:15 +0000 (10:52 +0100)]
Merge pull request #2350 from evverx/fix-memory-leak-on-failed-preset-all
core: fix memory leak on failed preset-all
Daniel Mack [Mon, 18 Jan 2016 09:51:19 +0000 (10:51 +0100)]
Merge pull request #2349 from evverx/test-functions-cleanup
tests: various fixes
Evgeny Vereshchagin [Mon, 18 Jan 2016 06:45:20 +0000 (06:45 +0000)]
tests: add STRIP_BINARIES
We need a beautiful stacktraces sometimes
For example https://github.com/systemd/systemd/pull/2328
Evgeny Vereshchagin [Mon, 18 Jan 2016 06:10:33 +0000 (06:10 +0000)]
core: fix memory leak on failed preset-all
How to reproduce
$ systemctl set-default multi-user # https://github.com/systemd/systemd/issues/2298
$ systemctl preset-all
Failed to execute operation: Too many levels of symbolic links
$ systemctl poweroff
Fixes:
==1==
==1== HEAP SUMMARY:
==1== in use at exit: 65,645 bytes in 7 blocks
==1== total heap usage: 40,539 allocs, 40,532 frees, 30,147,547 bytes allocated
==1==
==1== 109 (24 direct, 85 indirect) bytes in 1 blocks are definitely lost in loss record 2 of 7
==1== at 0x4C2BBCF: malloc (in /usr/lib/valgrind/vgpreload_memcheck-amd64-linux.so)
==1== by 0x4C2DE2F: realloc (in /usr/lib/valgrind/vgpreload_memcheck-amd64-linux.so)
==1== by 0x23DA71: unit_file_changes_add (install.c:233)
==1== by 0x23E45D: remove_marked_symlinks_fd (install.c:453)
==1== by 0x23E267: remove_marked_symlinks_fd (install.c:405)
==1== by 0x23E641: remove_marked_symlinks (install.c:494)
==1== by 0x243A91: execute_preset (install.c:2190)
==1== by 0x244343: unit_file_preset_all (install.c:2351)
==1== by 0x18AAA2: method_preset_all_unit_files (dbus-manager.c:1846)
==1== by 0x1D8157: method_callbacks_run (bus-objects.c:420)
==1== by 0x1DA9E9: object_find_and_run (bus-objects.c:1257)
==1== by 0x1DB02B: bus_process_object (bus-objects.c:1373)
==1==
==1== LEAK SUMMARY:
==1== definitely lost: 24 bytes in 1 blocks
==1== indirectly lost: 85 bytes in 1 blocks
==1== possibly lost: 0 bytes in 0 blocks
==1== still reachable: 65,536 bytes in 5 blocks
==1== suppressed: 0 bytes in 0 blocks
==1== Reachable blocks (those to which a pointer was found) are not shown.
==1== To see them, rerun with: --leak-check=full --show-leak-kinds=all
==1==
==1== For counts of detected and suppressed errors, rerun with: -v
==1== ERROR SUMMARY: 1 errors from 1 contexts (suppressed: 0 from 0)
Evgeny Vereshchagin [Sun, 17 Jan 2016 03:21:52 +0000 (03:21 +0000)]
tests: fix TEST-03-JOBS
We have
JOB UNIT TYPE STATE
1 testsuite.target start waiting
81 end.service start waiting
187 sleep.service start waiting
136 hello.service start waiting
82 testsuite.service start running
135 hello-after-sleep.target start waiting
sometimes
Evgeny Vereshchagin [Sun, 17 Jan 2016 03:33:20 +0000 (03:33 +0000)]
tests: remove unnecessary setup_nspawn_root
we don't run nspawn in TEST-02-CRYPTSETUP
Evgeny Vereshchagin [Sun, 17 Jan 2016 03:32:37 +0000 (03:32 +0000)]
tests: fix TEST-02-CRYPTSETUP on Debian/Ubuntu
Evgeny Vereshchagin [Sat, 16 Jan 2016 08:06:59 +0000 (08:06 +0000)]
tests: install fsck*
systemd-fsck depends on /sbin/fsck*
Evgeny Vereshchagin [Fri, 15 Jan 2016 23:58:45 +0000 (23:58 +0000)]
tests: remove unnecessary --boot
* Use $ROOTLIBDIR/systemd always
* Don't pass $ROOTLIBDIR/systemd as the first argument:
$ cat /proc/1/cmdline
/lib/systemd/systemd/lib/systemd/systemd...
Lennart Poettering [Sun, 17 Jan 2016 20:53:16 +0000 (21:53 +0100)]
resolved: fix logging about DNAME redirection
Lennart Poettering [Sun, 17 Jan 2016 20:50:10 +0000 (21:50 +0100)]
resolved: when we find a DNAME RR, don't insist in a signed CNAME RR
If we have a signed DNAME RR response, there's no need to insist on a signature for a CNAME RR response, after all it
is unlikely to be signed, given the implicit synthethis of CNAME through DNAME RRs.
Abdo Roig-Maranges [Sun, 17 Jan 2016 20:28:06 +0000 (21:28 +0100)]
Fix broken SYSTEMD_USER_WANTS in udev rules.
The functionality of SYSTEMD_USER_WANTS that attaches dependencies to device
units from udev rules was broken since commit
b2c23da8. I guess it was due to
a mass replace s/SYSTEMD_USER/MANAGER_USER/.
Lennart Poettering [Sun, 17 Jan 2016 16:25:42 +0000 (17:25 +0100)]
units: don't fail if /root doesn't exist for shell units
As discussed on the ML:
http://lists.freedesktop.org/archives/systemd-devel/2016-January/035594.html
Lennart Poettering [Fri, 15 Jan 2016 20:40:20 +0000 (21:40 +0100)]
resolved: update DNSSEC TODO
Lennart Poettering [Fri, 15 Jan 2016 20:38:27 +0000 (21:38 +0100)]
resolved: try to reduce number or DnsResourceKeys we keep around by merging them
Quite often we read the same RR key multiple times from the same message. Try to replace them by a single object when
we notice this. Do so again when we add things to the cache.
This should reduce memory consumption a tiny bit.
Lennart Poettering [Fri, 15 Jan 2016 20:07:21 +0000 (21:07 +0100)]
resolved: when switching between DNSSEC modes, possibly flush caches
If the networkd configuration changes during runtime, make sure to flush all caches when we switch from a less trusted
to a more trusted mode.
Lennart Poettering [Fri, 15 Jan 2016 19:45:17 +0000 (20:45 +0100)]
resolved: when the server feature level changes between query and response restart transaction
In some cases we learn something about a server's feature level through its responses. If we notice that after doing
basic checking of a response, and after collecting all auxiliary DNSSEC info the feature level of the server is lower
than where we started, restart the whole transaction.
This is useful to deal with servers that response rubbish when talked to with too high feature levels.
Lennart Poettering [Fri, 15 Jan 2016 19:36:40 +0000 (20:36 +0100)]
resolved: check OPT RR before accepting a reply for verification of server feature level
Let's make sure we first check if the OPT was lost in the reply, before we accept a reply as successful and use it for
verifying the current feature level.
Lennart Poettering [Fri, 15 Jan 2016 19:34:09 +0000 (20:34 +0100)]
resolved: when restarting a DNS transaction, remove all auxiliary DNSSEC transactions
When we restart a DNS transaction, remove all connections to any auxiliary DNSSEC transactions, after all we might
acquire completely different data this time, requiring different auxiliary DNSSEC transactions.
Lennart Poettering [Fri, 15 Jan 2016 19:29:56 +0000 (20:29 +0100)]
resolved: when we receive an reply which is OPT-less or RRSIG-less, downgrade what we verified
If we receive a reply that lacks the OPT RR, then this is reason to downgrade what was verified before, as it's
apparently no longer true, and the previous OPT RR we saw was only superficially OK.
Similar, if we realize that RRSIGs are not augmented, then also downgrade the feature level that was verified, as
DNSSEC is after all not supported. This check is in particular necessary, as we might notice the fact that RRSIG is not
augmented only very late, when verifying the root domain.
Also, when verifying a successful response, actually take in consideration that it might have been reported already
that RRSIG or OPT are missing in the response.
Lennart Poettering [Fri, 15 Jan 2016 18:23:51 +0000 (19:23 +0100)]
resolved: downgrade server feature level more aggressively when we have reason to
This adds logic to downgrade the feature level more aggressively when we have reason to. Specifically:
- When we get a response packet that lacks an OPT RR for a query that had it. If so, downgrade immediately to UDP mode,
i.e. don't generate EDNS0 packets anymore.
- When we get a response which we are sure should be signed, but lacks RRSIG RRs, we downgrade to EDNS0 mode, i.e.
below DO mode, since DO is apparently not really supported.
This should increase compatibility with servers that generate non-sensical responses if they messages with OPT RRs and
suchlike, for example the situation described here:
https://open.nlnetlabs.nl/pipermail/dnssec-trigger/2014-November/000376.html
This also changes the downgrade code to explain in a debug log message why a specific downgrade happened.
Lennart Poettering [Fri, 15 Jan 2016 17:18:54 +0000 (18:18 +0100)]
resolved: ignore invalid OPT RRs in incoming packets
This validates OPT RRs more rigorously, before honouring them: if we any of the following condition holds, we'll ignore
them:
a) Multiple OPT RRs in the same message
b) OPT RR not owned by the root domain
c) OPT RR in the wrong section (Belkin routers do this)
d) OPT RR contain rfc6975 algorithm data (Belkin routers do this)
e) OPT version is not 0
f) OPT payload doesn't add up with the lengths
Note that d) may be an indication that the server just blindly copied OPT data from the response into the reply.
RFC6975 data is only supposed to be included in queries, and we do so. It's not supposed to be included in responses
(and the RFC is very clear on that). Hence if we get it back in a reply, then the server probably just copied the OPT
RR.
Lennart Poettering [Fri, 15 Jan 2016 01:48:56 +0000 (02:48 +0100)]
resolved: update RFCs list and TODO list
Lennart Poettering [Fri, 15 Jan 2016 01:24:39 +0000 (02:24 +0100)]
resolved: add complex test case
This new test case tries to resolve a couple of known domains, to verify the validation results. It talks to resolved
via the bus, thus comprehensively testing the whole shebang.
Of course, it requires network connectivity and a DNSSEC capable DNS server, hence this is a manual test.
Lennart Poettering [Fri, 15 Jan 2016 01:21:22 +0000 (02:21 +0100)]
resolved: complete NSEC non-existance proofs
This fills in the last few gaps:
- When checking if a domain is non-existing, also check that no wildcard for it exists
- Ensure we don't base "covering" tests on NSEC RRs from a parent zone
- Refuse to accept expanded wildcard NSEC RRs for absence proofs.
Lennart Poettering [Thu, 14 Jan 2016 20:05:57 +0000 (21:05 +0100)]
resolved: make sure the NSEC proof-of-non-existance check also looks for wildcard domains
Lennart Poettering [Thu, 14 Jan 2016 19:12:29 +0000 (20:12 +0100)]
resolved: on negative NODATA replies, properly deal with empty non-terminals
empty non-terminals generally lack NSEC RRs, which means we can deduce their existance only from the fact that there
are other RRs that contain them in their suffix. Specifically, the NSEC proof for NODATA on ENTs works by sending the
NSEC whose next name is a suffix of the queried name to the client. Use this information properly.
Lennart Poettering [Thu, 14 Jan 2016 19:11:11 +0000 (20:11 +0100)]
resolved: rename dnssec_verify_dnskey() → dnssec_verify_dnskey_by_ds()
This should clarify that this is not regular signature-based validation, but validation through DS RR fingerprints.
Lennart Poettering [Thu, 14 Jan 2016 17:14:43 +0000 (18:14 +0100)]
resolved: be stricter when using NSEC3
We can user signer and synthesizing source information to check that the NSEC3 RRs we want to use are
actually reasonable and properly signed.
Lennart Poettering [Thu, 14 Jan 2016 17:03:03 +0000 (18:03 +0100)]
resolved: when validating an RRset, store information about the synthesizing source and zone in each RR
Having this information available is useful when we need to check whether various RRs are suitable for proofs. This
information is stored in the RRs as number of labels to skip from the beginning of the owner name to reach the
synthesizing source/signer. Simple accessor calls are then added to retrieve the signer/source from the RR using this
information.
This also moves validation of a a number of RRSIG parameters into a new call dnssec_rrsig_prepare() that as side-effect
initializes the two numeric values.
Lennart Poettering [Thu, 14 Jan 2016 16:28:58 +0000 (17:28 +0100)]
resolved: do not use NSEC RRs from the wrong zone for proofs
When proving NODATA DS lookups we need to insist on looking at the parent zone's NSEC RR, not the child zone's.
When proving any other NODATA lookups we need to insist on looking at the child zone's NSEC RR, not the parent's.
Lennart Poettering [Thu, 14 Jan 2016 16:27:28 +0000 (17:27 +0100)]
resolved: ignore DS RRs without generating an error if they use an unsupported digest algorithm
Lennart Poettering [Thu, 14 Jan 2016 16:25:06 +0000 (17:25 +0100)]
resolved: some RR types may appear only or not at all in a zone apex
Add extra checks when validating with RRSIGs. This follows recommendations from:
http://www.george-barwood.pwp.blueyonder.co.uk/DnsServer/NotesOnDNSSSEC.htm
Lennart Poettering [Thu, 14 Jan 2016 17:08:29 +0000 (18:08 +0100)]
Update TODO
Daniel Mack [Sun, 17 Jan 2016 12:47:18 +0000 (13:47 +0100)]
Merge pull request #2340 from evverx/fix-memory-leak-on-enable-disable-etc
core: fix memory leak on set-default, enable, disable etc
Evgeny Vereshchagin [Sun, 17 Jan 2016 09:36:03 +0000 (09:36 +0000)]
core: fix memory leak on set-default, enable, disable etc
Fixes:
==1== by 0x23E44C: remove_marked_symlinks_fd (install.c:453)
==1== by 0x23E256: remove_marked_symlinks_fd (install.c:405)
==1== by 0x23E630: remove_marked_symlinks (install.c:494)
==1== by 0x2427A0: unit_file_disable (install.c:1876)
==1== by 0x18A633: method_disable_unit_files_generic (dbus-manager.c:1760)
==1== by 0x18A6CA: method_disable_unit_files (dbus-manager.c:1768)
==1== by 0x1D8146: method_callbacks_run (bus-objects.c:420)
==1== by 0x1DA9D8: object_find_and_run (bus-objects.c:1257)
==1== by 0x1DB01A: bus_process_object (bus-objects.c:1373)
==1==
==1== 228 (48 direct, 180 indirect) bytes in 2 blocks are definitely lost in loss record 8 of 14
==1== at 0x4C2BBCF: malloc (in /usr/lib/valgrind/vgpreload_memcheck-amd64-linux.so)
==1== by 0x4C2DE2F: realloc (in /usr/lib/valgrind/vgpreload_memcheck-amd64-linux.so)
==1== by 0x23DA60: unit_file_changes_add (install.c:233)
==1== by 0x23DDB2: create_symlink (install.c:298)
==1== by 0x240C5C: install_info_symlink_wants (install.c:1328)
==1== by 0x240FC8: install_info_apply (install.c:1384)
==1== by 0x241211: install_context_apply (install.c:1439)
==1== by 0x242563: unit_file_enable (install.c:1830)
==1== by 0x18A06E: method_enable_unit_files_generic (dbus-manager.c:1650)
==1== by 0x18A141: method_enable_unit_files (dbus-manager.c:1660)
==1== by 0x1D8146: method_callbacks_run (bus-objects.c:420)
==1== by 0x1DA9D8: object_find_and_run (bus-objects.c:1257)
==1==
==1== 467 (144 direct, 323 indirect) bytes in 3 blocks are definitely lost in loss record 9 of 14
==1== at 0x4C2DD9F: realloc (in /usr/lib/valgrind/vgpreload_memcheck-amd64-linux.so)
==1== by 0x23DA60: unit_file_changes_add (install.c:233)
==1== by 0x23DE97: create_symlink (install.c:320)
==1== by 0x242CFC: unit_file_set_default (install.c:1951)
==1== by 0x18A881: method_set_default_target (dbus-manager.c:1802)
==1== by 0x1D8146: method_callbacks_run (bus-objects.c:420)
==1== by 0x1DA9D8: object_find_and_run (bus-objects.c:1257)
==1== by 0x1DB01A: bus_process_object (bus-objects.c:1373)
==1== by 0x259143: process_message (sd-bus.c:2567)
==1== by 0x259326: process_running (sd-bus.c:2609)
==1== by 0x259BDC: bus_process_internal (sd-bus.c:2798)
==1== by 0x259CAD: sd_bus_process (sd-bus.c:2817)
==1==
==1== LEAK SUMMARY:
==1== definitely lost: 216 bytes in 6 blocks
==1== indirectly lost: 560 bytes in 14 blocks
==1== possibly lost: 0 bytes in 0 blocks
==1== still reachable: 65,536 bytes in 5 blocks
==1== suppressed: 0 bytes in 0 blocks
==1== Reachable blocks (those to which a pointer was found) are not shown.
==1== To see them, rerun with: --leak-check=full --show-leak-kinds=all
==1==
Daniel Mack [Fri, 15 Jan 2016 12:11:16 +0000 (13:11 +0100)]
Merge pull request #2334 from jwilk/spelling
man: fix typos
Jakub Wilk [Fri, 15 Jan 2016 11:46:08 +0000 (12:46 +0100)]
man: fix typos
Daniel Mack [Fri, 15 Jan 2016 08:03:34 +0000 (09:03 +0100)]
Merge pull request #2328 from evverx/fix-transient-units-memeory-leak
Fix transient units memory leak
Evgeny Vereshchagin [Fri, 15 Jan 2016 02:52:12 +0000 (02:52 +0000)]
tests: add function for valgrind wrapper creation
I used it for d9814c7 and bffd87b
Evgeny Vereshchagin [Fri, 15 Jan 2016 02:41:27 +0000 (02:41 +0000)]
core: fix memory leak in transient units
Fixes:
==1== HEAP SUMMARY:
==1== in use at exit: 67,182 bytes in 91 blocks
==1== total heap usage: 70,485 allocs, 70,394 frees, 42,184,635 bytes
allocated
==1==
==1== 5,742 (696 direct, 5,046 indirect) bytes in 29 blocks are
definitely lost in loss record 4 of 7
==1== at 0x4C2DD9F: realloc (in
/usr/lib/valgrind/vgpreload_memcheck-amd64-linux.so)
==1== by 0x21ADDD: realloc_multiply (alloc-util.h:67)
==1== by 0x21BFB0: strv_push (strv.c:448)
==1== by 0x21C245: strv_consume (strv.c:520)
==1== by 0x21C33C: strv_extend (strv.c:559)
==1== by 0x278AD7: unit_write_drop_in (unit.c:3352)
==1== by 0x278EEB: unit_write_drop_in_private (unit.c:3403)
==1== by 0x190C21: bus_service_set_transient_property
(dbus-service.c:254)
==1== by 0x190DBC: bus_service_set_property (dbus-service.c:284)
==1== by 0x18F00E: bus_unit_set_properties (dbus-unit.c:1226)
==1== by 0x186F6A: transient_unit_from_message (dbus-manager.c:683)
==1== by 0x1872B7: method_start_transient_unit (dbus-manager.c:763)
==1==
==1== LEAK SUMMARY:
==1== definitely lost: 696 bytes in 29 blocks
==1== indirectly lost: 5,046 bytes in 58 blocks
==1== possibly lost: 0 bytes in 0 blocks
==1== still reachable: 61,440 bytes in 4 blocks
==1== suppressed: 0 bytes in 0 blocks
Zbigniew Jędrzejewski-Szmek [Thu, 14 Jan 2016 17:33:19 +0000 (12:33 -0500)]
Merge pull request #2322 from fbuihuu/downgrade-warn-for-not-found-unit
transaction: downgrade warnings about wanted units which are not found
Tom Gundersen [Thu, 14 Jan 2016 16:02:57 +0000 (17:02 +0100)]
Merge pull request #2316 from poettering/dnssec14
Fourteenth DNSSEC PR
Lennart Poettering [Thu, 14 Jan 2016 15:09:54 +0000 (16:09 +0100)]
Merge pull request #2319 from walyong/log_msg_v04
[v4] bus-util: print "systemctl --user" on user service manager
Franck Bui [Thu, 14 Jan 2016 08:25:18 +0000 (09:25 +0100)]
transaction: downgrade warnings about wanted unit which are not found
If a unit was pulled by a Wants= dependency but its unit file was not
present then we logged this as an error.
However Wants= might be used to configure a soft/optional dependency
on another unit, ie. start an optional service only if it's installed
otherwise simply skip it. In this case emitting an error doesn't look
appropriate.
But it's still an error if the optional dependency exists but its
activation fails for any reasons.
Daniel Mack [Thu, 14 Jan 2016 09:35:45 +0000 (10:35 +0100)]
Merge pull request #2320 from evverx/fix-memory-leak-on-reload
Fix memory leak on daemon-reload