cfg_obj_t doesn't store a pointer to its a parser context anymore,
and does not depend on the parser's lifecycle. Instead, it stores a
reference to its own memory context (and in principle, each node
could have different memory context). This also slightly simplifies
the _destroy API as there is no need to pass a context through it
anymore.
In order to reduce the lifecycle dependency of a `cfg_obj_t` on its
parser, the `file` field needs its own reference count, so it isn't
deleted when the parser is. It is now stored as a subsidiary
`cfg_obj_t` object of type string.
When generating a new key, dnssec-keygen checks for possible
key ID collisions with existing keys. The dnssec.c:findmatchingkeys()
function, which is supposed to get the list of the existing keys,
fails to do that for the existing KEY rrtype keys (i.e. generated
using 'dnssec-keygen -T KEY') because it doesn't pass down to the
dst_key_fromnamedfile() -> dst_key_read_public() functions the type
of the keys it's interested in. Fix the issue by introducing a new
function parameter which tells in which type of keys the caller is
currently interested in.
the functions dns_dumpctx_db() and dns_dumpctx_version() are used in
only one place, to get the serial number of the version being dumped.
it's simpler to expose the serial number through its own call,
dns_dumpctx_serial(), and remove the others.
Add a magic number check to ensure the memory context validity before
destorying it.
This check is needed now as it was done before implicitly when
isc_mem_inuse was called, but isc_mem_inuse is now called later (to be
able to dump the outstanding allocations).
When a memory context is destroyed, if the `checkfree` property is set,
the program assert there is no remaining allocation. If there are and
assertions are enabled, the program immediately stops.
However, if memory trace/record debug is enabled, the dump of
outstanding allocation won't be printed as it is done after the
no remaining allocation assertion check.
This moves the no remaining allocation assertion check after the dump of
outstanding allocations, so it is still possible to figure out what's
still allocated by this memory context.
if a zone reload is already in progress when 'rndc reload <zone>' is
run, currently the message returned in "zone reload queued", which
is correct, but it's identical to the message returned when a reload
was *not* in progress, so the user can't easily tell what happened.
a user could reload a zone twice and not realize that only one
reload actually took place.
this has been addressed by changing the message returned to
"zone reload was already queued".
a new result code ISC_R_LOADING has been added to signal this
condition, taking the place of ISC_R_RELOAD, which was obsolete
and has been removed.
The selfsigned_dnskey() function can now return all the return codes
that dns_dnssec_keyfromrdata() can return and this would cause an
assertion failure as we were not expecting new isc_result_t codes.
It was possible for the result to be overwritten after a
validation failure, causing check_signer() to return success
when it should have returned an error.
Co-Authored-By: Ondřej Surý <ondrej@isc.org>
Since the "tkey-gssapi-credential" statement has been previously
deprecated, mark it as ancient and remove all code related to it:
- The code processing the "tkey-gssapi-credential" statement in the
configuration is the only user of the dst_gssapi_acquirecred() and
dst_gssapi_releasecred() functions, so remove them along with their
static helper functions and a backup definition of the
GSS_KRB5_MECHANISM macro.
- When calling gss_accept_sec_context(), pass GSS_C_NO_CREDENTIAL
instead of the credential acquired by gss_acquire_cred().
(Previously, NULL was passed when "tkey-gssapi-credential" was not
specified. Kerberos headers define GSS_C_NO_CREDENTIAL as
(gss_cred_id_t) 0, so the logic was effectively the same, but using
the GSS_C_NO_CREDENTIAL macro is more appropriate.) This renders
the 'cred' parameter for dst_gssapi_acceptctx() redundant, so remove
it from the prototype of the latter. (Contrary to what the
documentation for dst_gssapi_acceptctx() claims,
dst_gssapi_releasecred() does not need to subsequently be called to
free the GSS-API context; a dst_gssapi_deletectx() call in
gssapi_destroy() takes care of that when the dynamically generated
TSIG key is destroyed.)
- Remove the 'gsscred' member from struct dns_tkeyctx, along with its
related dns_gss_cred_id_t typedef.
Update the relevant sections of the ARM and code comments accordingly.
This makes the "tkey-gssapi-keytab" statement the only way to set up
GSS-TSIG in named.
Remove redundant code from bin/named/tkeyconf.c while at it.
Use arc4random on platforms where available. arc4random() provides high
quality cryptographically-secure pseudo-random numbers and is generally
recommended for application use.
The uv_random() call unfortunately uses getentropy() on platforms like
MacOS, OpenBSD or NetBSD which is not recommended for application use.
It was discovered in an upcoming academic paper that a xoshiro128**
internal state can be recovered by an external 3rd party allowing to
predict UDP ports and DNS IDs in the outgoing queries. This could lead
to an attacker spoofing the DNS answers with great efficiency and
poisoning the DNS cache.
Change the internal random generator to system CSPRNG with buffering to
avoid excessive syscalls.
Thanks Omer Ben Simhon and Amit Klein of Hebrew University of Jerusalem
for responsibly reporting this to us. Very cool research!
To prevent spoofed unsigned DNAME responses being accepted retry
response with unsigned DNAMEs over TCP if the response is not TSIG
signed or there isn't a good DNS CLIENT COOKIE.
Use the owner name of the NS record as the bailwick apex name
when determining which additional records to cache, rather than
the name of the delegating zone (or a parent thereof).
To prevent certain spoofing attacks, a new check has been added
to the existing rules for whether NS data can be cached: the owner
name of the NS RRset must be an ancestor of the name being queried.
If a matching but cryptographically invalid key was encountered during
the DNSSEC validation, the key would be just skipped and not counted
towards validation failures. Treat such DNSSEC keys as hard failures
and fail the DNSSEC validation immediatelly instead of continuing the
DNSSEC validation with the next DNSKEYs in the RRset.
Co-authored-by: Matthijs Mekking <matthijs@isc.org>
The type `ns_pluginregister_ctx_t` was initially added to pass plugin
contextual data when the plugin is registered, but this is also now
passed into `plugin_check`. Furthermore, those various data are not
specific to the registration in particular. Rename the type into
`ns_pluginctx_t` for clarity.
Names for reverse resolution are now shared and accessible from
dns/name.h header as those are used by ptr_12.c and byaddr.c files.
(Also, for consistency, move deprecated ip6.int. name in name.h, even
though it is used only by ptr_12.c).
Add an API to parse and extract either an IPv4 or IPv6 address from
a name using the reverse format. It takes care of family detection,
and returns a generic error in case of syntax error.
Functions hex_decode_init(), hex_decode_char() and hex_decode_finish()
are now exposed, as well as the context hex_decode_ctx_t. They now are
respectively called isc_hex_decodeinit(), isc_hex_decodechar(),
isc_hex_decodefinish() and isc_hex_decodectx_t.
This enable to re-implement the functionality of isc_hex_decodestring()
in contextes where the input is not a NULL-terminated string, but, for
example, individual characters extracted (and avoid creating an
intermediate buffer to store them). This also enable to decode a stream
of hex characters where only hex characters are expected (i.e. no white
spaces).
Make sure all zone's plugin are unloaded before the zone gets freed.
This makes passing zone metadata like its origin to the plugin
registering function safe, as this garantee that the origin would always
be valid from the plugin lifecycle.
This commit introduces a new type, ns_pluginregister_ctx_t,
which is passed to plugin_check() and plugin_register() in place of the
'source' parameter. The source value is now just part of the structure,
which also holds a pointer to the zone origin if the plugin is loaded at
a zone level.
This provides more contextual information, enabling the plugin to make
specific configuration decisions based on the name of the zone for which
it is loaded.
It's also flexible if more contextual data are needed in the future:
add a new field to ns_pluginregister_ctx_t, and new plugins can use
it without affecting compatibility with existing plugins.
In named_config_parsefile(), when checking the validity of
named.conf, the checking of plugin correctness was deliberately
postponed until the plugin is loaded and registered. However,
when the plugin was registered, the checking was never actually
done: the plugin_register() implementation was called, but
plugin_check() was not.
This made it necessary to duplicate the correctness checking in both
functions, so that both named-checkconf and named could catch errors.
That should not be required.
ns_plugin_register() now calls the check function before the register
function, and aborts if either one fails. ns_plugin_check() calls only
the check function. ns_plugin_check() is used by named-checkconf, and
ns_plugin_register() is used by named. (Note: this design has a
side effect that a call to ns_plugin_register() will result in the
plugin parameters being parsed twice at registration time.)
ns_plugin_check() now takes an additional argument for the hook
source: zone or view.
Macro CHECK_FOR_GLUE_IN_ANSWER is defined in `lib/dns/resolver.c` only,
documented nowhere and not exposed as build configuration. This is valid
at least for 9.21+, 9.20 and 9.18. Furthermore, it doesn't compile
anymore on 9.21+ with -DCHECK_FOR_GLUE_IN_ANSWER=1.
Considering it is very unlikely that anyone build named with this,
remove the code rather than fixing it.
The field `ns_hookasync_t` was initially named `hook_actx` and wrongly
renamed `hook_aclctx` during a mass-renaming of various names for the
config acl context into a consistent `aclctx` name (see !11003). Of
course this is wrong as `ns_hookasync_t` has nothing to do with ACL but
about _async_ context. This commit fixes the mistake by renaming this
field `hookasyncctx`
After a full sign we no longer have to need to take the sign delay into
account. Update the timing checks in keymgr_transition_time to determine
the start of the interval: Either the last change, or if SigPublish/
SigDelete is set. The latter case indicates a full sign was done and
so we no longer have to take the sign delay into account.
When introducing the kasp logic, a full sign of the zone did not
generate new signatures for the new active keys during a ZSK rollover.
The introduced kasp logic ensured that the rollover is performed
smoothly, as in the signatures are only replaced if the old signature
is close to expiring (depending on the signatures-refresh option).
Fix by maintaining a fullsign boolean value in the signing structure,
that will ensure the RRsets are signed with the correct key, rather
than a similar good key.
In case of a fullsign, we can also remove signatures from inactive
keys.
Remove the unused dns_zone_signwithkey function.
When disabling algorithms, use the signer name to determine if the
algorithm is disabled or not. This allows for algorithms to be
cleanly disabled on a zone level basis. Previously, just using the
records owner name, "disable-algorithms" could impact resolution of
names that where not disabled. This does now mean that
"disable-algorithms" can not be used to disable part of a zone anymore.
ACL configuration context variables are inconsistently named as `actx`,
`ac`, or `aclconfctx`, which caused confusion during code reviews. This
commit renames all `cfg_aclconfctx_t` variables to `aclctx`, which is
short, consistent, and unambiguous.
The DST_ALGORITHM_FORMATSIZE constant is unused. It could be used in
dst_kasp_key_format, but instead we will use DNS_NAME_FORMATSIZE
because it is used in other places too. Clean up the unused constant.
The configuration should also take into account the built-in
DNSSEC policies when verifying the keys in the key-directory match the
given policy. Update the code accordingly and add some good and
failure test cases.
With named-checkconf -k you can check your configuration including
checking the dnssec-policy keys against the configured keystores. If
there is a mismatch in the key files versus the policy, named-checkconf
will fail. This is useful for running before migrating to dnssec-policy.
For logging purposes, introduce a function that writes the identifying
information about a policy key into a string.
Allow a dnssec key to be initialized outside the keymgr code.
Add 'log_errors' to 'cfg_kasp_fromconfig' to avoid duplicate error
logs.
Enabling LTO in the subsequent commit requires the file names to be
unique and having same probes.d in each of the libraries breaks this
requirement. Rename probes.d to probes-{isc,dns,ns}.d files and adjust
the includes.
`named_g_actconfctx` is a global variable holding the ACL configuration
context alive (in particular, to dynamically load zones). However, this
object is build once per configuration (early) and is used only inside
server.c `apply_configuration` flow. (Two exceptions: the shutdown flow,
still in server.c and plugin check flow, which doesn't need it, so it's
NULL in such case).
Instead of leaving this global publicly exposed, it is now part of the
`named_server_t` object. This allows us to clearly see that, when
reconfigureing the server, the new instance of the ACL context is known
only by the newly built object and not currently used by "production"
object; and will help to move move logic before the exclusive mode is
taken.
The other advantage is that the ACL configuration context can now be
built before the exclusive lock as well.
Previously, the fetch contexts were stored inside rwlocked hashmap
table. This was one of the most contended places for the resolver,
especially in the cold cache situation.
Replace the locked hashmap with the lock-free hashtable from the RCU
library and protect the fetch contexts against reuse by replacing the
libisc reference counting with urcu_ref that can soft-fail in situation
where the reference count is already zero. This allows us to easily
skip re-using the fetch context if it is already in process of being
destroyed.
Previously, the slabtops for "type" and its signature was only loosely
coupled and the headers could expire at different time (both TTL and LRU
based expiry). This commit expires the headers in both related
headers.
Co-authored-by: Matthijs Mekking <matthijs@isc.org>
