- knot-resolver-users - lists.nic.cz

non-persistent cache during system reboot although stored on persistent storage, why?

by Michael Grimm

Hi, I am running knot-resolver-5.7.4 in a FreeBSD service jail (14.1-STABLE). Note: Because I am still pretty new to using knot resolver, I may miss something important besides [1]. MWN> sum /usr/home/jails/test/var/run/kresd/data.mdb 268 10240 /usr/home/jails/test/var/run/kresd/data.mdb MWN> ./_STATS [2] cache('['count_entries']'): 4953 cache('['usage_percent']'): 16.953125 ### stopping jail simulates shutdown server [3]: MWN> service jail stop test Stopping jails: test. MWN> sum /usr/home/jails/test/var/run/kresd/data.mdb 268 10240 /usr/home/jails/test/var/run/kresd/data.mdb ### Thus, data.mdb is preserved after shutdown! ### starting jail simulates booting server: MWN> service jail start test Starting jails: test. MWN> sum /usr/home/jails/test/var/run/kresd/data.mdb 15059 10240 /usr/home/jails/test/var/run/kresd/data.mdb MWN> ./_STATS cache('['count_entries']'): 87 cache('['usage_percent']'): 0.15625 1) After having stopped that jail, data.mdb is still available and hasn't been modified as shown by checksum. 2) After start of the jail including start of kresd data.mdb has been modified (checksum). 3) cache.stats() shows significantly lower numbers. Questions: #) become cache.stats() reset after a reboot? #) what am I missing? Thanks in advance and regards, Michael [1] https://knot-resolver.readthedocs.io/en/stable/daemon-bindings-cache.html#p… [2] _STATS (based on https://knot-resolver.readthedocs.io/en/stable/daemon-bindings-cache.html#c…) echo -n "cache('['count_entries']'): " ; echo "cache.stats()" | nc -NU /var/run/kresd/control/kresd.sock | grep count_entries echo -n "cache('['usage_percent']'): " ; echo "cache.stats()" | nc -NU /var/run/kresd/control/kresd.sock | grep usage_percent [3] a real server reboot shows the same issue with the cache

2 weeks, 5 days

2
4
0 0

Moving Knot Resolver release files

by Oto Šťáva

Dear Knot Resolver users, due to an internal infrastructure change, released sources for Knot Resolver have been moved from <https://secure.nic.cz/files/knot-resolver/> to <https://knot-resolver.nic.cz/release/>. Apart from this movement, the rest of the directory structure remains unchanged. Proper redirects (HTTP 301 Moved Permanently) have been put in place to make this change as painless and transparent as possible. These redirects can be expected to stay in place for the foreseeable future. Still, we do recommend to change any of your direct links from the secure.nic.cz server to knot-resolver.nic.cz, to avoid the extra indirection step and/or unforeseen issues in the future. Should any of you run into any issues or have any questions about this change, please do let us know, we will be happy to help you out. Best regards -- Oto Šťáva | Knot Resolver team lead | CZ.NIC z.s.p.o. PGP: 6DC2 B0CB 5935 EA7A 3961 4AA7 32B2 2D20 C9B4 E680

3 weeks, 2 days

1
0
0 0

Expired key in upstream repositories

by Vladimír Čunát

Hello. In case you're using our upstream repositories for Debian or Ubuntu, as suggested on https://www.knot-resolver.cz/download/ you'll be running into their signing key expiring since today. As we didn't update it in time, you'll have to update it manually by re-running: wgethttps://secure.nic.cz/files/knot-resolver/knot-resolver-release.deb dpkg -i knot-resolver-release.deb Ticket: https://gitlab.nic.cz/knot/knot-resolver/-/issues/747 We also forgot to add Ubuntu 22.04, so that is fixed now, too. --Vladimir

1 month

2
2
0 0

CNAME resolution across internal STUB zones

by Paul Furtado

Hello, I am attempting to migrate some internal bind servers to knot+knot-resolver. We have knot acting as the authoritative server for internal views of our public zones and we're attempting to use Knot Resolver to forward to Knot to provide recursive resolution on the internal network. In this configuration, when Knot is resolving a CNAME record from an internal stub zone, it doesn't chase the target of the CNAME for the client. To demonstrate, here are two example zones served by knot: # zone1.com test-a 30 IN A 192.168.1.1 # zone2.com test-cname 30 CNAME test-a.zone1.com. With these configured as stubs in Knot Resolver, if you query test-cname.zone2.com, knot resolver won't chase it across zones to get the A record value like it would for non-stub zones. Both Bind9 and Unbound chase this as expected. Is there any way to configure Knot Resolver to do the same? Thanks! - Paul

1 month, 1 week

2
2
0 0

Knot Resolver 5.7.4 and 6.0.8 (security release)

by Oto Šťáva

Dear Knot Resolver users, Knot Resolver versions 5.7.4 (stable) and 6.0.8 (early-access) have been released! Both releases include important security fixes, an update is strongly advised! Version 6.0.8 additionally brings some improvements, like faster policy reloads using a new policy-loader process, respecting system-wide crypto policies, JSON metrics, separate IPv6 and IPv4 metrics, and more. --- Knot Resolver 5.7.4: Security: - reduce buffering of transmitted data, especially TCP-based in userspace Also expose some of the new tweaks in lua: (require 'ffi').C.the_worker.engine.net.tcp.user_timeout = 1000 (require 'ffi').C.the_worker.engine.net.listen_{tcp,udp}_buflens.{snd,rcv} Improvements: - add the fresh DNSSEC root key "KSK-2024" already, Key ID 38696 (!1556) Incompatible changes: - libknot 3.0.x support is dropped (!1558) Upstream last maintained 3.0.x in spring 2022. Full changelog: https://gitlab.nic.cz/knot/knot-resolver/raw/v5.7.4/NEWS Sources: https://secure.nic.cz/files/knot-resolver/knot-resolver-5.7.4.tar.xz GPG signature: https://secure.nic.cz/files/knot-resolver/knot-resolver-5.7.4.tar.xz.asc Documentation: https://www.knot-resolver.cz/documentation/v5.7.4/ --- Knot Resolver 6.0.8: Security: - reduce buffering of transmitted data, especially TCP-based in userspace Also expose some of the new tweaks in lua: (require 'ffi').C.the_worker.engine.net.tcp.user_timeout = 1000 (require 'ffi').C.the_worker.engine.net.listen_{tcp,udp}_buflens.{snd,rcv} Packaging: - all packages: - remove unused dependency on `libedit` (!1553) - deb packages: - packages ``knot-resolver-core`` and ``knot-resolver-manager`` have been merged into a single ``knot-resolver6`` package. Suffix packages ``knot-resolver-*`` have been renamed to ``knot-resolver6-*``. This change _should_ be transparent, but please do let us know if you encounter any issues while updating. (!1549) - package ``python3-prometheus-client`` is now only an optional dependency - rpm packages: - packages ``knot-resolver-core`` and ``knot-resolver-manager`` have been merged into a single ``knot-resolver`` package. This change _should_ be transparent, but please do let us know if you encounter any issues while updating. (!1549) - bugfix: do not overwrite config.yaml (!1525) - package ``python3-prometheus_client`` is now only an optional dependency - arch package: - fix after they renamed a dependency (!1536) Improvements: - TLS (DoT, DoH): respect crypto policy overrides in OS (!1526) - manager: export metrics to JSON via management HTTP API (!1527) * JSON is the new default metrics output format * the ``prometheus-client`` Python package is now an optional dependency, required only for Prometheus export to work - cache: prefetching records * predict module: prefetching expiring records moved to prefetch module * prefetch module: new module to prefetch expiring records - stats: add separate metrics for IPv6 and IPv4 (!1545) - add the fresh DNSSEC root key "KSK-2024" already, Key ID 38696 (!1556) - manager: policy-loader: new component for separate loading of policy rules (!1540) The ``policy-loader`` ensures that configured policies are loaded into the rules database where they are made available to all running kresd workers. This loading is no longer done by all kresd workers as it was before, so this should significantly improve the resolver's startup/reload time when loading large sets of policy rules, e.g. large RPZs. Incompatible changes: - cache: the ``cache.prediction`` configuration property has been reorganized into ``cache.prefetch.expiring`` and ``cache.prefetch.prediction``, changing the default behaviour as well. See the `relevant documentation section <https://www.knot-resolver.cz/documentation/v6.0.8/config-cache-predict.html>`_ for more. - libknot <=3.2.x support is dropped (!1565) Bugfixes: - arch package: fix after they renamed a dependency (!1536) - fix startup with `dnssec: false` (!1548) - rpm packages: do not overwrite config.yaml (!1525) - fix NSEC3 records missing in answer for positive wildcard expansion with the NSEC3 having over-limit iteration count (#910, !1550) - views: fix a bug in subnet matching (!1562) Full changelog: https://gitlab.nic.cz/knot/knot-resolver/raw/v6.0.8/NEWS Sources: https://secure.nic.cz/files/knot-resolver/knot-resolver-6.0.8.tar.xz GPG signature: https://secure.nic.cz/files/knot-resolver/knot-resolver-6.0.8.tar.xz.asc Documentation: https://www.knot-resolver.cz/documentation/v6.0.8/ -- Oto Šťáva | Knot Resolver team lead | CZ.NIC z.s.p.o. PGP: 6DC2 B0CB 5935 EA7A 3961 4AA7 32B2 2D20 C9B4 E680

1 month, 3 weeks

3
2
0 0

Re: KNOT sharing cache or not for different instances (nat64 vs no nat64)

by Vladimír Čunát

On 31/05/2024 19.00, oui.mages_0w(a)icloud.com wrote: > we have different TLS domains/certificates for dns64 and non dns64 Oh, OK. Such a thing hasn't occurred to us, so it's not possible. In that case I expect you'll need to stay on 5.x for now, with separate processes for dns64 and non-dns64 (but they can share the cache). Overall I don't think the current code can support multiple certificates.

2 months, 3 weeks

2
3
0 0

Re: KNOT sharing cache or not for different instances (nat64 vs no nat64)

by Vladimír Čunát

On 31/05/2024 13.04, oui.mages_0w(a)icloud.com wrote: > Unless the policy module allows to filter by listened IP, I will still > need to use split instances: we don’t select on the server side which > client is to use dns64 or not, but as an ISP, we leave the choice to > the clients to decide which dns resolver they want to use (one of ours > with or without dns64, or a third party). That is possible, but with 5.x I probably wouldn't recommend it, as it's been left out of documentation (by mistake) and overall I don't have much trust in that part of 5.x policies. But after you migrate to >= 6.x, I would recommend just having a single shared configuration. And in views you can select dst-subnet paired with dns64 options. Such deployments were taken into account when designing 6.x views. https://www.knot-resolver.cz/documentation/latest/config-views.html#config-… In 6.x it would probably also be harder for you to run multiple configurations at once on a single machine, so that's another reason to unify this when you migrate. --Vladimir

3 months, 2 weeks

1
0
0 0

KNOT resolver and NS serverers with CNAME

by Vladimír Čunát

On 14/05/2024 08.10, Josef Karliak via knot-dns-users wrote: > > ISC bind is strict about CNAME of NS server: > > skipping nameserver 'aa.bb.cz' *because it is a CNAME*, while > resolving '9.4/4.3.2.1.in-addr.arpa/PTR'. > > How about Knot resolver ? > Hello. I believe there is neither effort to disallow them nor to make them work. Off the top of my head I'm not sure how it will be in practice; you could just try. Either way, please don't use them. (I replied to the correct mailing-list.) --Vladimir

3 months, 2 weeks

3
3
0 0

Knot Resolver 5.7.3

by Oto Šťáva

Dear Knot Resolver users, Knot Resolver version 5.7.3 (stable) has been released! This maintenance update brings a minor improvement to stats and a bugfix to missing NSEC3 records. An early-access version 6.0.8 including these changes and more will come at a later date - a few more things need to be finalized there before release. Improvements: - stats: add separate metrics for IPv6 and IPv4 (!1544) Bugfixes: - fix NSEC3 records missing in answer for positive wildcard expansion with the NSEC3 having over-limit iteration count (#910, !1550) Full changelog: https://gitlab.nic.cz/knot/knot-resolver/raw/v5.7.3/NEWS Sources: https://secure.nic.cz/files/knot-resolver/knot-resolver-5.7.3.tar.xz GPG signature: https://secure.nic.cz/files/knot-resolver/knot-resolver-5.7.3.tar.xz.asc Documentation: https://www.knot-resolver.cz/documentation/v5.7.3/ -- Oto Šťáva | Knot Resolver team lead | CZ.NIC z.s.p.o. PGP: 6DC2 B0CB 5935 EA7A 3961 4AA7 32B2 2D20 C9B4 E680

3 months, 2 weeks

1
0
0 0

Re: [knot-dns-users] Re: KNOT resolver and NS serverers with CNAME

by Vladimír Čunát

On 17/05/2024 22.43, Peter Thomassen wrote: > I think the question is whether Knot Resolver follows the letter of > the RFC, like BIND, or whether it is less strict. I'm not aware of RFCs saying that resolvers should fail in such situations. My understanding is more like it's allowed not to work. Anyway, it would be better to continue this thread on knot-resolver-users(a)lists.nic.cz (I already have posted there about this a couple days ago.) --Vladimir

4 months

1
0
0 0

2024

2023

2022

2021

2020

2019

2018

knot-resolver-users