The .delegating flag was only set, but never used in the dns_qpcache.
Remove it completely together with the code that was locking the node
to set the flag if the added type was DNAME.
Merge branch 'ondrej/remove-delegating-from-qpcache' into 'main'
See merge request isc-projects/bind9!10980
The .delegating flag was only set, but never used in the dns_qpcache.
Remove it completely together with the code that was locking the node
to set the flag if the added type was DNAME.
Upstream has removed the atomics implementation of CMM_LOAD_SHARED and
CMM_STORE_SHARED as these can be used also with non-stdatomics types.
As we only use the CMM api with stdatomics types, we can restore the
previous behaviour to prevent ThreadSanitizer warnings.
Closes#5660
Merge branch '5660-use-atomics-for-CMM-api-with-thread-sanitizer' into 'main'
See merge request isc-projects/bind9!11288
Upstream has removed the atomics implementation of CMM_LOAD_SHARED and
CMM_STORE_SHARED as these can be used also with non-stdatomics types.
As we only use the CMM api with stdatomics types, we can restore the
previous behaviour to prevent ThreadSanitizer warnings.
There are multiple reasons for the increased amount of differences we've
been seeing lately and for the raise of the threshold:
1. Recent hardening against cache poisoning (CVE-2025-40778) have
uncovered a few edge cases where the domain can't be properly
resolved with the new protections in place, but those are issues with
upstream configuration and DNS setup.
2. The same hardening magnified some behaviour differences between 9.21
and older versions. Some misconfigured domains, which can be resolved
with BIND 9.20 and older are no longer resolvable in 9.21+. This can
be again attributed to upstream DNS misconfiguration. See #5649.
3. A change in the respdiff CI job to include timeouts in the
comparison, or rather, increasing the timeouts to resolve the
previously timed out queries, which are typically failures. With the
previous job configuration, those were omitted from comparison,
because they were timeouts. Now, there should be no timeouts, but
there is a slight increase in the amount of differences for the
threshold evaluation.
This job is testing the current BIND implementation against the latest
released version. Unless there has been a behaviour change, there should
be no difference.
In practice, there is a small number of differences caused by upstream
discrepencies. Some of those cause "upstream unstable" answers which are
excluded from the results, but statistically, some of those will end up
being detected as differences on the resolver under test.
Currently, there seems to be about 300 upstream unstable answers with
typically around 50-60 differences. Setting the threshold to 0.1 should
be stable enough to pass if there are no changes, yet sensitive enough
to detect even fairly small changes to behaviour.
Disable test that fails for months on this platform.
Closes#5448
Merge branch '5448-disable-doh-unit-test-on-el8' into 'main'
See merge request isc-projects/bind9!11276
Change d5e4684b3d accidentally caused
ns_statscounter_edns0out to no longer be incremented. This has been
corrected.
Closes#5654
Merge branch '5654-restore-recording-ns_statscounter_edns0out' into 'main'
See merge request isc-projects/bind9!11267
When a zone is re-transferred, but the zone journal on an inline-signing secondary is out of sync, the zone could fall back to using NSEC records instead of NSEC3. This has been fixed.
Closes#5527
Merge branch '5527-retransfer-nsec3-bug' into 'main'
See merge request isc-projects/bind9!11226
When synchronizing the secure database, we skip DNSSEC records that
BIND 9 maintains with inline-signing. We should also skip private
RDATA type records that are used to track the current state of a
zone-signing process.
If the primary has been updated, but the secondary has not been
notified, the journal will go out of date. An 'rndc retransfer' causes
the zone to force an AXFR, removing and rebuilding zone and journal
files.
This test reproduces a bug that in such scenario, an NSEC3 signed zone
falls back to NSEC.
The "no root hints for view X" message must not be shown for the default
_bind/CH view. However, it is shown since 27c4f68dcc (part of effective
configuration changes).
The reason is that since 27c4f68dcc, `configure_views()` now processes
a single list of views, which contains both builtin and user views as
they are both part of the effective configuration. The changes omit the
(now removed, since 6b5f714e53) `need_hints` bool that was disabling the
warning for the builtin view (_bind/CH).
This disable that log again when configuring _bind/CH view.
Merge branch 'colin/nohintswarn-bindchaos' into 'main'
See merge request isc-projects/bind9!11264
The "no root hints for view X" message must not be shown for the default
_bind/CH view. However, it is shown since 27c4f68dcc (part of effective
configuration changes).
The reason is that since 27c4f68dcc, `configure_views()` now processes
a single list of views, which contains both builtin and user views as
they are both part of the effective configuration. Those changes omitted
the `need_hints` bool that disabled the warning for the builtin view.
This commit silences the log message again.
The new `dns_message_ednsinit()` and `dns_message_ednsaddopt()` functions
allow EDNS options to be added to a message one at a time; it is no
longer necessary to construct a full array of EDNS options and set
them all at once.
This allows us to simplify EDNS option handling code, and in the
future it wlil allow plugins to add EDNS options to existing
messages.
Merge branch 'each-refactor-message-edns' into 'main'
See merge request isc-projects/bind9!11261
now that the EDNS state is stored within dns_message_t, it's no longer
necessary to have a public API call to build an opt rdataset; we can
just have dns_message_setopt() build the opt record internally.
The new dns_message_ednsinit() and dns_message_ednsaddopt() functions
allow EDNS options to be added to a message one at a time; it is no
longer necessary to construct a full array of EDNS options and set
them all at once.
This allows us to simplify EDNS option handling code, and in the
future it wlil allow plugins to add EDNS options to existing
messages.
To check if the key managing for a zone is done, we check a log message.
On some machines, with many zones this may take some time.
Increate the timeout to a minute. Note that most zones will continue
their test case much faster, but with many zones there may be some
straggling.
When all zones are configured, regardless of whether the test module
actually uses them, it makes debugging the logs needlessly more
complicated, as there is a bunch of stuff going on that is completely
unrelated to the test.
Define a list of tested zones in each test module and only render the
named.conf with those zones defined.
This converts two test cases:
1. A zone that previously failed to load is now fixed. Make sure the
zone is signed correctly with the right NSEC3 parameters.
2. Test case to ensure the salt is the same after a restart, i.e. no
re-salting takes place. Previously we only tested with salt length
0, this commit adds a test case for salt length 8 as well.
This converts the nsec3 system test cases after to reconfiguring the
name server.
Two extra test for nsec3-change.kasp is updated. It depends on the
zone being updated, and a reconfig. This test code is moved to
tests_nsec3_reconfig.py.
Furthermore, an additional 'rndc signing -nsec3param' error test
case has been added.
Change the named.conf templating to make use of jinja template
rendering. The ns2 server is trivial. The ns3 server configuration
structure has changed:
The common configuration is moved out of named-fips.conf.
The main named.conf file is in named.conf.j2. It always includes the
common part, named-common.conf.j2, and the FIPS part,
named-fips.conf.j2.
The named-fips.conf.j2 and named-rsasha1.conf.j2 templates are
rendered differently depending on the reconfiged status. Mainly the
dnssec-policy for zones are different after reconfiguration, but there
are some other changes to, for example some zones change their
inline-signing setting.
Some zones only exist prior or after the configuration.
Finally, this is a bit hackish: If RSASHA1 is supported, named.conf
includes "named-rsasha1.conf", otherwise it includes the deliberately
empty "named-rsasha0.conf".
This converts all the nsec3 system test cases prior to reconfiguring the
name server. There are two main classes, one that tests the zone is
correctly signed with NSEC, the other with NSEC3.
Two extra tests for nsec3-dynamic-update-inline.kasp and
nsec3-change.kasp are also rewritten. For the former, we need to
change the 'nsupdate' definition to be able to set the expected RCODE.
The merging of the user options and defaults into the effective configuration broke the mutual inheritance of the `allow-recursion`, `allow-query`, and `allow-query-cache` ACLs, and of the `allow-recursion-on` and `allow-query-cache-on` ACLs. This has been fixed.
Closes#5647
Merge branch '5647-allow-recursion-inheritance' into 'main'
See merge request isc-projects/bind9!11254
when merging view objects into the effective configuration, add
allow-query-cache, allow-recursion, allow-query-cache-on and
allow-recursion-on ACLs as needed to reflect the way those
options inherit from each other.
this means the effective configuration is now correct for each
view. ACLs no longer need to be corrected when applying the
configuration, and the actual effective ACL values will be
displayed in "rndc showconf" and "named-checkconf -pe".
the merging of options and defaults into the effective configuration
broke the mutual inheritance of the allow-recursion, allow-query, and
allow-query-cache ACLs, and of the allow-recursion-on and
allow-query-cache-on ACLs.
this has been corrected by adding a 'cloned' flag to the cfg_obj
structure to indicate whether it was configured explicitly or
cloned from the defaults during parsing. we can then adjust the
ACLs while configuring a view, favoring user-configured values
when they're available over cloned defaults.
currently the adjustments to the ACLs are done in configure_view();
later they'll be moved into the effective configuration and this
special handling can be removed.
allow-recursion is set to "none" in the options block and to
"any" in the view. allow-query-cache in the view should inherit
the "any", not the "none". (currently this test does not pass.)
Call to `streamdns_resume_processing` is asynchronous but the socket
passed as argument is not attached when scheduling the call.
While there is no reproducible way (so far) to make the socket reference
number down to 0 before `streamdns_resume_processing` is called, attach
the socket before scheduling the call. This guard against an hypothetic
case where, for some reasons, the socket refcount would reach 0, and be
freed from memory when `streamdns_resume_processing` is called.
Closes#5620
Merge branch '5620-attach-socket-streamdns_resume_processing' into 'main'
See merge request isc-projects/bind9!11247
Call to `streamdns_resume_processing` is asynchronous but the socket
passed as argument is not attached when scheduling the call.
While there is no reproducible way (so far) to make the socket reference
number down to 0 before `streamdns_resume_processing` is called, attach
the socket before scheduling the call. This guard against an hypothetic
case where, for some reasons, the socket refcount would reach 0, and be
freed from memory when `streamdns_resume_processing` is called.
Reduces the number of outgoing queries when resolving the nameservers
for delegation points. This helps the DNS resolver with cold cache
resolve client queries with complex delegation chains and redirections.
Merge branch 'ondrej/fctx_getaddresses' into 'main'
See merge request isc-projects/bind9!11148
The fctx_getaddresses() was lengthy and little bit confusing with
goto statements. Split the single function into smaller parts:
one for forwarders, one for nameservers and one for alternates.
The dns_resolver mode of operation is to resolve all the domains as it
iterates the DNS tree to fill up the cache as quickly as possible.
This commit reduces the number of outgoing queries by reducing the
number of remote fetches started for the nameserver addresses resolution
via dns_adb_createfind() to a smaller number per depth of the recursion
since the delegation point (3 2 1 0) - where 0 means only create fetch
on demand if we don't have any addresses yet.
RFC 8777 specifies a placeholder value of "." for the gateway field when the
gateway type is 0 (no gateway). This was not being checked for nor emitted
when displaying the record. This has been corrected.
Instances of this record will need the placeholder period added to them when upgrading.
Closes#5639
Merge branch '5639-fix-atmrelay-type-0-support' into 'main'
See merge request isc-projects/bind9!11240
Add the OpenBSD and Clang combination to the existing kludge to
recognize size_t and uintXX_t types as same when using a generic.
Closes#5635
Merge branch '5635-openbsd-size_t-generics-fix' into 'main'
See merge request isc-projects/bind9!11235
