]>
git.ipfire.org Git - thirdparty/knot-resolver.git/log
Aleš [Wed, 19 Aug 2020 14:43:31 +0000 (16:43 +0200)]
sysrepo: common module to create/destroy context
Aleš [Tue, 18 Aug 2020 08:52:54 +0000 (10:52 +0200)]
cache_gc: build without sysrepo integration
Vasek Sraier [Thu, 13 Aug 2020 14:37:07 +0000 (16:37 +0200)]
model: code refactoring - rewritten to closure based OOP
Vasek Sraier [Thu, 13 Aug 2020 11:34:15 +0000 (13:34 +0200)]
sysrepo: build parts of resolver using sysrepo only if sysrepo build is enabled
Vasek Sraier [Thu, 13 Aug 2020 11:30:08 +0000 (13:30 +0200)]
sysrepo: added script to initialize sysrepo schema
Vasek Sraier [Thu, 13 Aug 2020 11:04:56 +0000 (13:04 +0200)]
sysrepo: added YANG schemas as a submodule in order to pin their version
Vasek Sraier [Thu, 13 Aug 2020 11:01:07 +0000 (13:01 +0200)]
sysrepo module removed, sysrepo-lua module renamed to sysrepo
Aleš [Mon, 3 Aug 2020 13:06:59 +0000 (15:06 +0200)]
watcher: sdbus module to control processes
Aleš [Wed, 29 Jul 2020 13:36:09 +0000 (15:36 +0200)]
sysrepo: cache_gc: xpath cutting removed
Aleš [Tue, 28 Jul 2020 15:01:32 +0000 (17:01 +0200)]
sysrepo: cache_gc: minor adjustments
- defined xpath for cache storage
- united return value variable
- break of long lines
Aleš [Fri, 24 Jul 2020 14:09:07 +0000 (16:09 +0200)]
sysrepo: meson files cleanup
Aleš [Mon, 20 Jul 2020 15:05:26 +0000 (17:05 +0200)]
Merge branch 'sysrepo' of gitlab.nic.cz:knot/knot-resolver into sysrepo
Vasek Sraier [Sun, 26 Apr 2020 21:17:23 +0000 (23:17 +0200)]
sysrepo-lua: changed StructuralBindNode into BindNode, removed unnecessary code, improved error reporting
Vasek Sraier [Sun, 26 Apr 2020 20:38:06 +0000 (22:38 +0200)]
sysrepo-lua: ported structural binding from sysrepo-bench branch
Vasek Sraier [Fri, 24 Apr 2020 13:26:01 +0000 (15:26 +0200)]
sysrepo-lua: fixed schema traversal
Vasek Sraier [Sun, 12 Apr 2020 10:05:48 +0000 (12:05 +0200)]
sysrepo-lua: dumping schema before module initialization for debugging
Vasek Sraier [Mon, 6 Apr 2020 13:38:11 +0000 (15:38 +0200)]
sysrepo-lua: proper generic list support in model
Vasek Sraier [Sun, 5 Apr 2020 15:05:00 +0000 (17:05 +0200)]
sysrepo-lua: helpers and TLS binding implementation (not tested)
Vasek Sraier [Sat, 4 Apr 2020 09:48:07 +0000 (11:48 +0200)]
sysrepo-lua: added listen interfaces binding
Vasek Sraier [Wed, 25 Mar 2020 13:11:50 +0000 (14:11 +0100)]
sysrepo-lua: fixed a bug so that our model does not have to mirror the real model as a whole
Aleš [Tue, 24 Mar 2020 23:25:08 +0000 (00:25 +0100)]
sysrepo-lua: model correction
Aleš [Mon, 23 Mar 2020 15:24:51 +0000 (16:24 +0100)]
meson.build: required libyang and sysrepo version
Vasek Sraier [Thu, 12 Mar 2020 10:27:49 +0000 (11:27 +0100)]
sysrepo-lua: improved code documentation, added few more bindings
Vasek Sraier [Wed, 11 Mar 2020 17:08:05 +0000 (18:08 +0100)]
sysrepo-lua: bind node simplification
Vasek Sraier [Wed, 11 Mar 2020 15:08:46 +0000 (16:08 +0100)]
sysrepo-lua: implemented bidirectional binding
Vasek Sraier [Wed, 11 Mar 2020 14:15:11 +0000 (15:15 +0100)]
sysrepo-lua: fixed validation
Vasek Sraier [Wed, 11 Mar 2020 14:10:56 +0000 (15:10 +0100)]
kres_cache_gc: upgrade to sysrepo 1.4
Vasek Sraier [Wed, 11 Mar 2020 14:00:59 +0000 (15:00 +0100)]
sysrepo: update of libsysrepo dependency to 1.4
Vasek Sraier [Wed, 11 Mar 2020 12:38:23 +0000 (13:38 +0100)]
sysrepo-lua: validating results of operational data serialization
Vasek Sraier [Thu, 5 Mar 2020 15:04:36 +0000 (16:04 +0100)]
sysrepo-lua: initial work on reporting to operational datastore
Vasek Sraier [Sat, 1 Feb 2020 13:45:24 +0000 (14:45 +0100)]
sysrepo-lua: created abstraction for reading and writing configuration to sysrepo
Vasek Sraier [Mon, 27 Jan 2020 13:42:15 +0000 (14:42 +0100)]
sysrepo-lua: fixed absolute path shared object loading from within Lua
Vasek Sraier [Sun, 26 Jan 2020 14:48:08 +0000 (15:48 +0100)]
sysrepo-lua: migrated module code from sysrepo-module-lua branch
Because this is not a merge commit, line history was lost. But
it does not matter that much, because pretty much all the commited
code was written by me and it's mainly a proof of concept. Nothing
important should have been lost.
Aleš [Thu, 23 Jan 2020 13:07:25 +0000 (14:07 +0100)]
sysrepo: C module integration to just print xpath of changed nodes
Aleš Mrázek [Fri, 17 Jan 2020 12:33:20 +0000 (13:33 +0100)]
sysrepo: integration to cache-gc
Aleš [Wed, 15 Jan 2020 15:42:48 +0000 (16:42 +0100)]
sysrepo: meson corrections
Aleš [Wed, 15 Jan 2020 14:23:00 +0000 (15:23 +0100)]
sysrepo: meson build for each part
Aleš [Wed, 15 Jan 2020 14:15:09 +0000 (15:15 +0100)]
sysrepo: module: added common files directory
Ales Mrazek [Wed, 15 Jan 2020 12:00:59 +0000 (13:00 +0100)]
sysrepo: update meson.build
Ales Mrazek [Wed, 8 Jan 2020 11:43:47 +0000 (12:43 +0100)]
sysrepo: meson configuration
Petr Špaček [Thu, 16 Jul 2020 08:30:26 +0000 (10:30 +0200)]
Merge branch '580-add-cache-usage-to-cache-stats' into 'master'
cache: add percentage usage to cache stats
Closes #580
See merge request knot/knot-resolver!1025
Lukáš Ježek [Fri, 10 Jul 2020 13:08:13 +0000 (15:08 +0200)]
cache: add percentage usage to cache stats
Petr Špaček [Wed, 15 Jul 2020 07:44:09 +0000 (09:44 +0200)]
Merge branch 'nitpicks' into 'master'
nitpicks: batch of tiny fixes collected over time
See merge request knot/knot-resolver!1024
Vladimír Čunát [Tue, 14 Jul 2020 15:34:30 +0000 (17:34 +0200)]
ci travis: move to the last stage
So that the overall pipeline time isn't extended too much.
Vladimír Čunát [Tue, 14 Jul 2020 15:13:35 +0000 (17:13 +0200)]
ci travis: tweak the timeouts a bit
In the past week the Travis runs have been consistently taking much more
time than before, usually around 20 minutes, leading to our CI timing out.
https://travis-ci.com/github/CZ-NIC/knot-resolver/builds
Vladimír Čunát [Tue, 14 Jul 2020 14:44:38 +0000 (16:44 +0200)]
lib/resolve nitpick: missing newline in verbose message
Vladimír Čunát [Mon, 29 Jun 2020 16:47:37 +0000 (18:47 +0200)]
.travis.yml: nit fixes according to the validator
- root: deprecated key sudo (The key `sudo` has no effect anymore.)
- root: key matrix is an alias for jobs, using jobs
Vladimír Čunát [Tue, 14 Apr 2020 10:06:44 +0000 (12:06 +0200)]
doc-comments: fix some of the complaints from doxygen
There's still frequent issue that documenting some parameters would be
mainly noise but doxygen will warn when not doing it.
WARN_IF_UNDOCUMENTED apparently doesn't cover this and
WARN_IF_DOC_ERROR would probably remove even some useful warnings.
Vladimír Čunát [Tue, 14 Apr 2020 06:43:59 +0000 (08:43 +0200)]
stats nitpick: nicer code
Vladimír Čunát [Tue, 14 Apr 2020 06:41:24 +0000 (08:41 +0200)]
remove more --forks=1 ocurrences
It's deprecated since 5.0.0.
Vladimír Čunát [Tue, 14 Jul 2020 16:15:10 +0000 (18:15 +0200)]
Merge !1023: pytests: fix tls session resumption tests
Tomas Krizek [Tue, 14 Jul 2020 15:25:24 +0000 (17:25 +0200)]
treewide: rename remaining deprecated labs URLs
Tomas Krizek [Tue, 14 Jul 2020 12:13:10 +0000 (14:13 +0200)]
pytests: fix tls session resumption tests
Tomas Krizek [Tue, 14 Jul 2020 08:35:00 +0000 (10:35 +0200)]
Merge branch 'tls_resumption_multi_process' into 'master'
tls: fix compilation to support net.tls_sticket_secret()
See merge request knot/knot-resolver!1021
Vladimír Čunát [Fri, 10 Jul 2020 12:35:36 +0000 (14:35 +0200)]
config tests: better test net.tls_sticket_secret()
The trick there is that it isn't supported (by us) on gnutls < 3.6.3.
I checked that the test fails before the fix in parent commit
and that it succeeds (is skipped) with gnutls 3.6.2.
Petr Špaček [Thu, 9 Jul 2020 14:51:54 +0000 (16:51 +0200)]
tls: enable net.tls_sticket_secret() for session resumption across processes
Tomas Krizek [Fri, 10 Jul 2020 12:16:36 +0000 (14:16 +0200)]
Merge branch 'gitlab-move' into 'master'
treewide: move to our new GitLab URL
See merge request knot/knot-resolver!1019
Vladimír Čunát [Wed, 1 Jul 2020 09:20:49 +0000 (11:20 +0200)]
treewide: move to our new GitLab URL
s/gitlab\.labs\.nic/gitlab.nic/g
Redirects are in place, so it shouldn't be required now, but why not.
Petr Špaček [Wed, 8 Jul 2020 06:46:37 +0000 (08:46 +0200)]
Merge branch 'test_cleanup' into 'master'
test cleanups
See merge request knot/knot-resolver!1017
Vladimír Čunát [Wed, 1 Jul 2020 16:27:23 +0000 (18:27 +0200)]
module tests (daf + prefill): switch to shared check_answer()
I think this eliminates the remaining copies. Most of the places don't
need all the features, but it still seems worth to deduplicate.
Petr Špaček [Tue, 30 Jun 2020 15:54:38 +0000 (17:54 +0200)]
config tests: log test file name at the beginning of output
Petr Špaček [Tue, 30 Jun 2020 15:50:35 +0000 (17:50 +0200)]
config tests: detect incorrect config test suite definitions
Old behavior where test definition without "return" was silently
skipped was very confusing.
Petr Špaček [Tue, 30 Jun 2020 15:26:35 +0000 (17:26 +0200)]
config tests: do not load and detect_time_skew and ta_update modules
It was only generating noise in test logs, especially when network is
not abvailable/is intentionally disabled.
Vladimír Čunát [Fri, 3 Jul 2020 16:09:20 +0000 (18:09 +0200)]
Merge !1012: don't drop capabilities when running as root
Tomas Krizek [Fri, 19 Jun 2020 10:33:40 +0000 (12:33 +0200)]
meson: add build options to disable libcapng
Tomas Krizek [Fri, 19 Jun 2020 09:47:33 +0000 (11:47 +0200)]
daemon: don't drop capabilities when running as root
When the effective user is root, no capabilities are dropped. This
change has no effect when running as non-privileged user or when
switching to non-privileged user via user() in config.
Dropping capabilities as a root user resulted in the following
unexpected behaviour:
1. When using trust anchor update, r/w access to root keys is neeeded.
These are typically owned by knot-resolver user. When kresd is
executed as root and capabilities are dropped, this file was no longer
writable, because it is owned by knot-resolver, not root.
2. It is impossible to recreate/resize cache due to the same permission
issue as above.
If you want to drop capabilities when starting kresd as a root user,
you can switch the user with the `user()` command. This changes the
effective user ID and drops any capabilities as well.
Tomas Krizek [Wed, 1 Jul 2020 12:26:03 +0000 (14:26 +0200)]
Merge branch 'release-5-1-2' into 'master'
release 5.1.2
See merge request knot/knot-resolver!1018
Tomas Krizek [Wed, 1 Jul 2020 08:30:03 +0000 (10:30 +0200)]
release 5.1.2
Tomas Krizek [Wed, 1 Jul 2020 11:26:23 +0000 (13:26 +0200)]
doc: default to ignoring warning during doc build
Add new target doc-strict for development to detect warnings, but avoid
failing package builds due to documentation warnings.
Tomas Krizek [Wed, 1 Jul 2020 08:24:42 +0000 (10:24 +0200)]
distro/rpm: use opensuse license format
opensuse and fedora/epel use different license strings, but the opensuse
value is used in Knot DNS, so let's be consistent.
Cherry picked from https://build.opensuse.org/request/show/817870
Tomas Krizek [Wed, 1 Jul 2020 07:58:46 +0000 (09:58 +0200)]
distro/tests: remove fedora30 (EOL)
Petr Špaček [Wed, 1 Jul 2020 06:25:14 +0000 (08:25 +0200)]
Merge branch 'rpz-warnings' into 'master'
policy.rpz: various fixes
See merge request knot/knot-resolver!1016
Vladimír Čunát [Tue, 30 Jun 2020 16:27:44 +0000 (18:27 +0200)]
NEWS: describe the RPZ fixes
Vladimír Čunát [Tue, 30 Jun 2020 14:11:18 +0000 (16:11 +0200)]
policy.rpz: support local data with larger RRsets
Vladimír Čunát [Tue, 30 Jun 2020 14:05:32 +0000 (16:05 +0200)]
tests check_answer(): support checking RDATA
Also allow using empty set as an alternative to NODATA pseudo-RCODE,
and migrate RPZ tests to this merged function.
Vladimír Čunát [Tue, 30 Jun 2020 13:45:08 +0000 (15:45 +0200)]
policy.ANSWER: allow multiple RRs
... by allowing .rdata to be a table. Larger RRsets seem useful.
Vladimír Čunát [Tue, 30 Jun 2020 12:37:12 +0000 (14:37 +0200)]
policy.rpz: fix $ORIGIN-related handling
- use parser-detected $ORIGIN instead of looking at SOA owner
- skip records outside $ORIGIN (and warn) instead of nesting them
- simplify a bit, and tweak warnings
Vladimír Čunát [Tue, 30 Jun 2020 08:51:08 +0000 (10:51 +0200)]
policy.rpz: don't warn on NS and SOA records
Also utilize table indexing.
This was a "regression" from extending RPZ support in 5.1.0.
NS and SOA are even mandatory, as RPZ is supposed to be a valid zone:
https://tools.ietf.org/html/draft-ietf-dnsop-dns-rpz-00#section-2
Tomas Krizek [Mon, 29 Jun 2020 15:05:08 +0000 (17:05 +0200)]
Merge branch '585-graphite-prevents-kresd-to-start-if-graphite-server-is-not-available' into 'master'
graphite: Reconnect to the graphite server when it was unavailable
Closes #585
See merge request knot/knot-resolver!1014
Petr Špaček [Mon, 29 Jun 2020 14:07:43 +0000 (16:07 +0200)]
graphite: NEWS
Petr Špaček [Mon, 29 Jun 2020 13:58:10 +0000 (15:58 +0200)]
graphite: use TCP timeout value derived from publish interval
It seems pointless to accumulate "late" connection attempts.
Petr Špaček [Mon, 29 Jun 2020 12:29:10 +0000 (14:29 +0200)]
graphite: make socket operations asynchronous
Petr Špaček [Mon, 29 Jun 2020 12:28:45 +0000 (14:28 +0200)]
graphite: more informative socket error messages
Lukáš Ježek [Fri, 26 Jun 2020 11:39:39 +0000 (13:39 +0200)]
graphite: reconnect to the graphite server when it was unavailable
Tomas Krizek [Mon, 29 Jun 2020 13:34:35 +0000 (15:34 +0200)]
Merge branch 'gc-fixes' into 'master'
GC fixes
See merge request knot/knot-resolver!1009
Vladimír Čunát [Tue, 9 Jun 2020 14:27:07 +0000 (16:27 +0200)]
gc: use kresd versioning instead of separate one
I don't think it makes sense to ever package it separately from kresd,
so why should it have separate versioning? (could be confusing)
Vladimír Čunát [Tue, 9 Jun 2020 14:19:24 +0000 (16:19 +0200)]
gc: print numbers in more readable way
Vladimír Čunát [Tue, 9 Jun 2020 06:09:32 +0000 (08:09 +0200)]
gc: fix integer overflow when computing how much to GC
On 32-bit systems the insufficient GC could commonly happen:
https://lists.nic.cz/pipermail/knot-resolver-users/2020/000265.html
The meaning of -f parameter got slightly changed, so that the buggy
computation could be greatly simplified. GC seems to make sense when
most of cache space is used, in which case the difference is small.
Vladimír Čunát [Tue, 9 Jun 2020 04:24:42 +0000 (06:24 +0200)]
gc: fix flushing of messages to logs
There were the usual effects: very incorrect timestamps (even by days),
and I suspect we could even lose some logs completely.
Vladimír Čunát [Thu, 25 Jun 2020 12:52:03 +0000 (14:52 +0200)]
Merge !1013: lib/cache: fix locking around cache preallocation
Fixes #582.
Petr Špaček [Thu, 25 Jun 2020 12:40:26 +0000 (14:40 +0200)]
cache: NEWS for prealloc lock fix
Lukáš Ježek [Thu, 25 Jun 2020 08:30:23 +0000 (10:30 +0200)]
lib/cache: fix locking around cache preallocation
Vladimír Čunát [Fri, 19 Jun 2020 12:12:52 +0000 (14:12 +0200)]
Merge !1010: meson: treat -Wattributes as error
Tomas Krizek [Wed, 17 Jun 2020 08:53:18 +0000 (10:53 +0200)]
meson: treat -Wattributes as error
Trie relies on __attribute__(cleanup) and if its missing, it could
lead to memory leaks such as:
https://sonarcloud.io/project/issues?id=dns-oarc%3Adnsjit&issues=AXDyskhlrWLe1VCMvmVe&open=AXDyskhlrWLe1VCMvmVe
Tomas Krizek [Wed, 27 May 2020 08:54:29 +0000 (10:54 +0200)]
Merge branch 'packaging-pytest' into 'master'
CI: optimize packaging tests
Closes #541
See merge request knot/knot-resolver!947
Lukáš Ježek [Tue, 26 May 2020 08:12:37 +0000 (10:12 +0200)]
CI: error handling when using gitlab api
Tomas Krizek [Mon, 18 May 2020 16:15:08 +0000 (18:15 +0200)]
tests/packaging: fix runtime dependencies
Tomas Krizek [Mon, 18 May 2020 12:30:04 +0000 (14:30 +0200)]
tests/packaging: add centos 8
Tomas Krizek [Mon, 18 May 2020 11:48:03 +0000 (13:48 +0200)]
tests/packaging: add ubuntu 20.04
Tomas Krizek [Mon, 18 May 2020 11:41:13 +0000 (13:41 +0200)]
tests/packaging: add f32, remove f30