Hi, Folks!
I want to use GoogleDNS 8.8.x.x as fallback forwarder when domain's NS'es
are down or inaccessible (for example, when network connectivity between me
and NS is broken).
Is it possible?
policy.add(policy.all(policy.FORWARD({ '8.8.8.8', '8.8.4.4' }))) works
well, but disables recursion at all. How to call it only after/when
recursion is failed?
WBR, Ilya
Dear Knot Resolver users,
Knot Resolver 2.3.0 has been released. This is a security release that
fixes CVE-2018-1110.
We're also introducing a new mailing list, knot-resolver-announce. Only
notifications about new releases or important announcements will be
posted there.
https://lists.nic.cz/cgi-bin/mailman/listinfo/knot-resolver-announce
Security
--------
- fix CVE-2018-1110: denial of service triggered by malformed DNS
messages (!550, !558, security!2, security!4)
- increase resilience against slow lorris attack (security!5)
Bugfixes
--------
- validation: fix SERVFAIL in case of CNAME to NXDOMAIN in a single
zone (!538)
- validation: fix SERVFAIL for DS . query (!544)
- lib/resolve: don't send unecessary queries to parent zone (!513)
- iterate: fix validation for zones where parent and child share
NS (!543)
- TLS: improve error handling and documentation (!536, !555, !559)
Improvements
------------
- prefill: new module to periodically import root zone into cache
(replacement for RFC 7706, !511)
- network_listen_fd: always create end point for supervisor supplied
file descriptor
- use CPPFLAGS build environment variable if set (!547)
Full changelog:
https://gitlab.labs.nic.cz/knot/knot-resolver/raw/v2.3.0/NEWS
Sources:
https://secure.nic.cz/files/knot-resolver/knot-resolver-2.3.0.tar.xz
GPG signature:
https://secure.nic.cz/files/knot-resolver/knot-resolver-2.3.0.tar.xz.asc
Documentation:
https://knot-resolver.readthedocs.io/en/v2.3.0/
--
Tomas Krizek
PGP: 4A8B A48C 2AED 933B D495 C509 A1FB A5F7 EF8C 4869
Hi,
I have a fresh installation of the Knot Resolver on my Fedora 27:
*Package version*:
$ rpm -q knot-resolver
knot-resolver-2.2.0-1.fc27.x86_64
but it does not work out of the box. The problem is, that the user
"knot-resolver" cannot bind to a privileged port. Why is the systemd
service file using knot-resolver user? It works just fine, when I remove
the "User=" option from service file and add this line into the
kres.conf file:
user('knot-resolver', 'knot-resolver')
The resulting process runs as knot-resolver and it binds successfully.
Best regards,
--
Martin Sehnoutka | Associate Software Engineer
PGP: 5FD64AF5
UTC+1 (CET)
RED HAT | TRIED. TESTED. TRUSTED.
Hello,
There is a website I need to use in a daily basis that uses DNSSEC,
however their keys have expired which causes validation to fail. I have
contacted their support but they failed to resolve the issue so far.
Since I can resolve the name when using `dig +cd`, I was hoping I could
configure `kresd` to skip validation when resolving that specific
domain. It seems that I should be able to do so by using the `policies`
module and the `FLAGS` action:
https://knot-resolver.readthedocs.io/en/stable/modules.html#actions
I am not sure with flag/flags to use. I inspected the source and tried
the following:
policy.add(policy.suffix(policy.FLAGS('DNSSEC_CD'),{todname('example.org.')}))
But this apparently had no effect. I also tried without the trailing dot
and played with other flags, but no success.
Does anybody know which flag I could set to bypass DNSSEC validation for
the specified domain? Or, if the policy module is not the way to achieve
that goal, is there any other way?
# kresd --version
Knot DNS Resolver, version 1.5.1
Any help will be greatly appreciated,
// Leonardo.
Greetings. I want to build Knot 2.1 on a Ubuntu 16.04.4 LTS. However, the
libknot from the packages is too old:
Makefile:87: *** libknot >= 2.6.4 required. Stop.
As compared to:
libknot-dev/xenial,now 2.1.1-1build1 amd64 [installed]
Is it possible to get the libknot package on Ubuntu 16.04.4 LTS updated
soon? If not, how do I install the latest libknot by hand?
--Paul Hoffman
Greetings. My kresd config file is:
net.listen('192.241.207.161', 5364)
trust_anchors.file = 'root.keys'
modules.load('ta_sentinel')
I wanted to test this with a zone I set up at this-is-signed.com. However,
I'm getting a positive result back for both the is-ta and the not-ta
records (it is properly giving me the SERVFAIL for the bogus record).
Have I configured Knot Resolver incorrectly? For Knot 2.2.1, do I need a
different form for the names in order to get the kskroll-sentinel effect to
kick in? From the DNSOP WG mailing list traffic, I thought I needed the
4f66 tag, but could have misinterpreted that.
--Paul Hoffman
# dig @192.241.207.161 -p 5364
kskroll-sentinel-is-ta-4f66.this-is-signed.com a
; <<>> DiG 9.10.3-P4-Ubuntu <<>> @192.241.207.161 -p 5364
kskroll-sentinel-is-ta-4f66.this-is-signed.com a
; (1 server found)
;; global options: +cmd
;; Got answer:
;; ->>HEADER<<- opcode: QUERY, status: NOERROR, id: 36153
;; flags: qr rd ra ad; QUERY: 1, ANSWER: 2, AUTHORITY: 0, ADDITIONAL: 1
;; OPT PSEUDOSECTION:
; EDNS: version: 0, flags:; udp: 4096
;; QUESTION SECTION:
;kskroll-sentinel-is-ta-4f66.this-is-signed.com. IN A
;; ANSWER SECTION:
kskroll-sentinel-is-ta-4f66.this-is-signed.com. 60 IN CNAME
this-is-signed.com.
this-is-signed.com. 60 IN A 192.241.207.161
;; Query time: 283 msec
;; SERVER: 192.241.207.161#5364(192.241.207.161)
;; WHEN: Sun Feb 25 00:09:54 UTC 2018
;; MSG SIZE rcvd: 105
# dig @192.241.207.161 -p 5364
kskroll-sentinel-not-ta-4f66.this-is-signed.com a
; <<>> DiG 9.10.3-P4-Ubuntu <<>> @192.241.207.161 -p 5364
kskroll-sentinel-not-ta-4f66.this-is-signed.com a
; (1 server found)
;; global options: +cmd
;; Got answer:
;; ->>HEADER<<- opcode: QUERY, status: NOERROR, id: 14466
;; flags: qr rd ra ad; QUERY: 1, ANSWER: 2, AUTHORITY: 0, ADDITIONAL: 1
;; OPT PSEUDOSECTION:
; EDNS: version: 0, flags:; udp: 4096
;; QUESTION SECTION:
;kskroll-sentinel-not-ta-4f66.this-is-signed.com. IN A
;; ANSWER SECTION:
kskroll-sentinel-not-ta-4f66.this-is-signed.com. 60 IN CNAME
this-is-signed.com.
this-is-signed.com. 54 IN A 192.241.207.161
;; Query time: 5 msec
;; SERVER: 192.241.207.161#5364(192.241.207.161)
;; WHEN: Sun Feb 25 00:10:00 UTC 2018
;; MSG SIZE rcvd: 106
# dig @192.241.207.161 -p 5364 bogus.this-is-signed.com a
; <<>> DiG 9.10.3-P4-Ubuntu <<>> @192.241.207.161 -p 5364
bogus.this-is-signed.com a
; (1 server found)
;; global options: +cmd
;; Got answer:
;; ->>HEADER<<- opcode: QUERY, status: SERVFAIL, id: 20810
;; flags: qr rd ra; QUERY: 1, ANSWER: 0, AUTHORITY: 0, ADDITIONAL: 0
;; QUESTION SECTION:
;bogus.this-is-signed.com. IN A
;; Query time: 9 msec
;; SERVER: 192.241.207.161#5364(192.241.207.161)
;; WHEN: Sun Feb 25 00:10:08 UTC 2018
;; MSG SIZE rcvd: 42
Dear Knot Resolver users,
Knot Resolver 2.1.0 is released.
Incompatible changes
--------------------
- stats: remove tracking of expiring records (predict uses another way)
- systemd: re-use a single kresd.socket and kresd-tls.socket
- ta_sentinel: implement protocol draft-ietf-dnsop-kskroll-sentinel-01
(our draft-ietf-dnsop-kskroll-sentinel-00 implementation had inverted
logic)
- libknot: require version 2.6.4 or newer to get bugfixes for DNS-over-TLS
Bugfixes
--------
- detect_time_jump module: don't clear cache on suspend-resume (#284)
- stats module: fix stats.list() returning nothing, regressed in 2.0.0
- policy.TLS_FORWARD: refusal when configuring with multiple IPs (#306)
- cache: fix broken refresh of insecure records that were about to expire
- fix the hints module on some systems, e.g. Fedora (came back on 2.0.0)
- build with older gnutls (conditionally disable features)
- fix the predict module to work with insecure records & cleanup code
Full changelog:
https://gitlab.labs.nic.cz/knot/knot-resolver/raw/v2.1.0/NEWS
Sources:
https://secure.nic.cz/files/knot-resolver/knot-resolver-2.1.0.tar.xz
GPG signature:
https://secure.nic.cz/files/knot-resolver/knot-resolver-2.1.0.tar.xz.asc
Documentation:
https://knot-resolver.readthedocs.io/en/v2.1.0/
--
Tomas Krizek
PGP: 4A8B A48C 2AED 933B D495 C509 A1FB A5F7 EF8C 4869
Hi,
I am running knot-resolver version 2.0.0 and I would like to redirect
(STUB) queries from a subnet to a specific nameserver. For that I
created a view configuration for that subnet (see below). A query
without being in that subnet caches the answer but unfortunately the
cache overrides the answer when doing the same query inside the view.
My configuration file looks like that:
---
modules = {
'hints > iterate',
'policy > hints',
'view < cache'
}
view:addr('${SUBNET}', policy.suffix(policy.STUB('${IP}'),
{todname('${DOMAIN_SUFFIX}')}))
---
Testing procedure:
1) Try to resolve A record being in the special subnet.
$kdig @${KNOT-RESOLVER} A www.${DOMAIN_SUFFIX}
-> STUB works and answer is correct
2) Try to resolve A record without being in the special subnet.
$kdig @${KNOT-RESOLVER} ${KNOT-RESOLVER}
-> view is not triggerd and answer is correct
3) Try to resolve A record again within the special subnet.
$kdig @${KNOT-RESOLVER} A www.${DOMAIN_SUFFIX}
-> The answer is satisfied from cache, which is not the right answer.
Is it even possible to do that?
Best wishes,
Sakirnth