The current dispatch code could reuse the TCP connection when
dns_dispatch_gettcp() would be used first. This is problematic as the
dns_resolver doesn't use TCP connection sharing, but dns_request could
get the TCP stream that was created outside of the dns_request.
Add new DNS_DISPATCHOPT_UNSHARED option to dns_dispatch_createtcp() that
would prevent the TCP stream to be reused. Use that option in the
dns_resolver call to dns_dispatch_createtcp() to prevent dns_request
from reusing the TCP connections created by dns_resolver.
Additionally, the dns_xfrin unit added TCP connection sharing for
incoming transfers. While interleaving *xfr streams on a TCP connection
should work this should be a deliberate change and be property of the
server that can be controlled. Additionally some level of parallel TCP
streams is desirable. Revert to the old behaviour by removing the
dns_dispatch_gettcp() calls from dns_xfrin and use the new option to
prevent from sharing the transfer streams with dns_request.
The dns__catz_update_cb() does not expect that 'catzs->zones'
can become NULL during shutdown.
Add similar checks in the dns__catz_update_cb() and dns_catz_zone_get()
functions to protect from such a case. Also add an INSIST in the
dns_catz_zone_add() function to explicitly state that such a case
is not expected there, because that function is called only during a
reconfiguration.
The dns_aclenv_t contains two dns_acl_t - localhost and localnets that
can be swapped with a different ACLs as we configure BIND 9. Instead of
protecting those two pointers with heavyweight read-write lock, use RCU
mechanism to dereference and swap the pointers.
In units that support detailed reference tracing via ISC_REFCOUNT
macros, we were doing:
/* Define to 1 for detailed reference tracing */
#undef <unit>_TRACE
This would prevent using -D<unit>_TRACE=1 in the CFLAGS.
Convert the above mentioned snippet with just a comment how to enable
the detailed reference tracing:
/* Add -D<unit>_TRACE=1 to CFLAGS for detailed reference tracing */
Keeping the information about lame server in the ADB was done in !322 to
fix following security issue:
[CVE-2021-25219] Disable "lame-ttl" cache
The handling of the lame servers needs to be redesigned and it is not
going to be enabled any time soon, and the current code is just dead
code that takes up space, code and stands in the way of making ADB work
faster.
Remove all the internals needed for handling the lame servers in the ADB
for now. It might get reintroduced later if and when we redesign ADB.
now that we have the QP chain mechanism, we can convert the
RPZ summary database to use a QP trie instead of an RBT.
also revised comments throughout the file accordingly, and
incidentally cleaned up calls to new_node(), which can no
longer fail.
dns_qp_findname_ancestor() now takes an optional 'predecessor'
parameter, which if non-NULL is updated to contain the DNSSEC
predecessor of the name searched for. this is done by constructing
an iterator stack while carrying out the search, so it can be used
to step backward if needed.
since dns_qp_findname_ancestor() can now return a chain object, it is no
longer necessary to provide a _NOEXACT search option. if we want to look
up the closest ancestor of a name, we can just do a normal search, and
if successful, retrieve the second-to-last node from the QP chain.
this makes ancestor lookups slightly more complicated for the caller,
but allows us to simplify the code in dns_qp_findname_ancestor(), making
it easier to ensure correctness. this was a fairly rare use case:
outside of unit tests, DNS_QPFIND_NOEXACT was only used in the zone
table, which has now been updated to use the QP chain. the equivalent
RBT feature is only used by the resolver for cache lookups of 'atparent'
types (i.e, DS records).
- make iterators reversible: refactor dns_qpiter_next() and add a new
dns_qpiter_prev() function to support iterating both forwards and
backwards through a QP trie.
- added a 'name' parameter to dns_qpiter_next() (as well as _prev())
to make it easier to retrieve the nodename while iterating, without
having to construct it from pointer value data.
dns_qp_findname_ancestor() now takes an optional 'chain' parameter;
if set, the dns_qpchain object it points to will be updated with an
array of pointers to the populated nodes between the tree root and the
requested name. the number of nodes in the chain can then be accessed
using dns_qpchain_length() and the individual nodes using
dns_qpchain_node().
add a 'foundname' parameter to dns_qp_findname_ancestor(),
and use it to set the found name in dns_nametree.
this required adding a dns_qpkey_toname() function; that was
done by moving qp_test_keytoname() from the test library to qp.c.
added some more test cases and fixed bugs with the handling of
relative and empty names.
When the given zone is not associated with a zone manager, the function
currently returns ISC_R_NOTFOUND, which is documented as the return
value for the case in which no incoming zone transfer is found. Make
the function return ISC_R_FAILURE in such a case instead.
Also update the description of the function as the value it returns is
not meant to indicate whether an ongoing incoming transfer for the given
zone exists. The boolean variables that the function sets via the
pointers provided as its parameters, combined with either keeping
'*xfrp' set to NULL or updating it to a valid pointer, can be used by
the caller to infer all the necessary information.
The 'end_serial' and some other members of the 'dns_xfrin_t'
structure can be accessed by the statistics channel, causing
a data race with the zone transfer process.
Use the existing 'statslock' mutex for protecting those members.
Instead of creating new memory pools for each new dns_message, change
dns_message_create() method to optionally accept externally created
dns_fixedname_t and dns_rdataset_t memory pools. This allows us to
preallocate the memory pools in ns_client and dns_resolver units for the
lifetime of dns_resolver_t and ns_clientmgr_t.
Currently in the statsistics channel's incoming zone transfers list
the local and remote addresses are shown only when the zone transfer
is already running. Since we have now introduced the "Refresh SOA"
state, which shows the state of the SOA query before the zone transfer
is started, this commit implements a feature to show the local and
remote addresses for the SOA query, when the state is "Refresh SOA".
Improve the "Duration (s)" field, so that it can show the duration of
all the major states of an incoming zone transfer process, while they
are taking place. In particular, it will now show the duration of the
"Pending", "Refresh SOA" and "Deferred" states too, before the actual
zone transfer starts.
With adding this state to the statistics channel, it can now show
the zone transfer in this state instead of as "Pending" when the
zone.c module is performing a refresh SOA request, before actually
starting the transfer process. This will help to understand
whether the process is waiting because of the rate limiter (i.e.
"Pending"), or the rate limiter is passed and it is now waiting for
the refresh SOA query to complete or time out.
Add a new field in the incoming zone transfers section of the
statistics channel to show the transport used for the SOA request.
When the transfer is started beginning from the XFRST_SOAQUERY state,
it means that the SOA query will be performed by xfrin itself, using
the same transport. Otherwise, it means that the SOA query was already
performed by other means (e.g. by zone.c:soa_query()), and, in that
case, we use the SOA query transport type information passed by the
'soa_transport_type' argument, when the xfrin object was created.
dns_xfrin_t state may be accessed from different threads when
when reporting transfer state. Ensure access is thread safe by
using atomics and locks where appropriate.
In some dns_transport getter functions it's possible to use a
const dns_transport_t as the first argument instead of just
dns_transport_t. Convert the function prototypes to use const.
Reusing TCP connections with dns_dispatch_gettcp() used linear linked
list to lookup existing outgoing TCP connections that could be reused.
Replace the linked list with per-loop cds_lfht hashtable to speedup the
lookups. We use cds_lfht because it allows non-unique node insertion
that we need to check for dispatches in different connection states.
Instead of high number of dispatches (4 * named_g_udpdisp)[1], make the
dispatches bound to threads and make dns_dispatchset_t create a dispatch
for each thread (event loop).
This required couple of other changes:
1. The dns_dispatch_createudp() must be called on loop, so the isc_tid()
is already initialized - changes to nsupdate and mdig were required.
2. The dns_requestmgr had only a single dispatch per v4 and v6. Instead
of using single dispatch, use dns_dispatchset_t for each protocol -
this is same as dns_resolver.
Add a configuration option, resolver-use-dns64, which when true
will cause named to map IPv4 address to IPv6 addresses using the
view's DNS64 mapping rules when making iterative queries.
1. Replace the "high-performance" macros that were only used if
DNS_NAME_USEINLINE was defined before including <dns/name.h> with
inline header functions with assertion checks and thus use them
everywhere.
2. Replace the old struct initializers with C99 designated initializers
for better understanding what is happening in these macros.
instead of allowing a NULL nametree in dns_nametree_covered(),
require nametree to exist, and ensure that the nametrees defined
for view and resolver objects are always created.
name trees can now also hold trees of counters. each time a name
dns_nametree_add() is called with a given name, the counter for that
name is incremented; the name is not deleted until dns_nametree_delete()
is called the same number of times.
this is meant to be used for synth-from-dnssec, which is incremented for
each key defined at a name, and decremented when a key is removed, the
name must continue to exist until the number of keys has reached zero.
name trees can now hold either boolean values or bit fields. the
type is selected when the name tree is created.
the behavior of dns_nametree_add() differs slightly beteween the types:
in a boolean tree adding an existing name will return ISC_R_EXISTS,
but in a bitfield tree it simply sets the specified bit in the bitfield
and returns ISC_R_SUCCESS.
replace the use of RBTs for deny-answer-aliases, the exclude
lists for deny-answer-aliases and deny-answer-addresses, and
dnssec-must-be-secure, with name trees.
this is a QP trie of boolean values to indicate whether a name is
included in or excluded from some policy. this can be used for
synth-from-dnssec, deny-answer-aliases, etc.
The dns_dispatchmgr object was only set in the dns_view object making it
prone to use-after-free in the dns_xfrin unit when shutting down named.
Remove dns_view_setdispatchmgr() and optionally pass the dispatchmgr
directly to dns_view_create() when it is attached and not just assigned,
so the dns_dispatchmgr doesn't cease to exist too early.
The dns_view_getdnsdispatchmgr() is now protected by the RCU lock, the
dispatchmgr reference is incremented, so the caller needs to detach from
it, and the function can return NULL in case the dns_view has been
already shut down.
Instead of an RBT for the forwarders table, use a QP trie.
We now use reference counting for dns_forwarders_t. When a forwarders
object is retrieved by dns_fwdtable_find(), it must now be explicitly
detached by the caller afterward.
QP tries require stored objects to include their names, so the
the forwarders object now has that. This obviates the need to
pass back a separate 'foundname' value from dns_fwdtable_find().
replace the red-black tree used by the negative trust anchor table
with a QP trie.
because of this change, dns_ntatable_init() can no longer fail, and
neither can dns_view_initntatable(). these functions have both been
changed to type void.
this function finds the closest matching ancestor, but the function
name could be read to imply that it returns the direct parent node;
this commit suggests a slightly less misleading name.
Make the `pval_r` and `ival_r` out arguments optional.
Add `pval_r` and `ival_r` out arguments to `dns_qp_deletekey()`
and `dns_qp_deletename()`, to return the deleted leaf.
Allow larger TTL values in zones that go insecure. This is necessary
because otherwise the zone will not be loaded due to the max-zone-ttl
of P1D that is part of the current insecure policy.
In the keymgr.c code, default back to P1D if the max-zone-ttl is set
to zero.
Add an option to enable/disable inline-signing inside the
dnssec-policy clause. The existing inline-signing option that is
set in the zone clause takes priority, but if it is omitted, then the
value that is set in dnssec-policy is taken.
The built-in policies use inline-signing.
This means that if you want to use the default policy without
inline-signing you either have to set it explicitly in the zone
clause:
zone "example" {
...
dnssec-policy default;
inline-signing no;
};
Or create a new policy, only overriding the inline-signing option:
dnssec-policy "default-dynamic" {
inline-signing no;
};
zone "example" {
...
dnssec-policy default-dynamic;
};
This also means that if you are going insecure with a dynamic zone,
the built-in "insecure" policy needs to be accompanied with
"inline-signing no;".
The updatenotify mechanism in dns_db relied on unlocked ISC_LIST for
adding and removing the "listeners". The mechanism relied on the
exclusive mode - it should have been updated only during reconfiguration
of the server. This turned not to be true anymore in the dns_catz - the
updatenotify list could have been updated during offloaded work as the
offloaded threads are not subject to the exclusive mode.
Change the update_listeners to be cds_lfht (lock-free hash-table), and
slightly refactor how register and unregister the callbacks - the calls
are now idempotent (the register call already was and the return value
of the unregister function was mostly ignored by the callers).
The dns_badcache unit had (yet another) own locked hashtable
implementation. Replace the hashtable used by dns_badcache with
lock-free cds_lfht implementation from liburcu.
When dns_request was canceled via dns_requestmgr_shutdown() the cancel
event would be propagated on different loop (loop 0) than the loop where
request was created on. In turn this would propagate down to isc_netmgr
where we require all the events to be called from the matching isc_loop.
Pin the dns_requests to the loops and ensure that all the events are
called on the associated loop. This in turn allows us to remove the
hashed locks on the requests and change the single .requests list to be
a per-loop list for the request accounting.
Additionally, do some extra cleanup because some race condititions are
now not possible as all events on the dns_request are serialized.
The isc_stats_create() can no longer return anything else than
ISC_R_SUCCESS. Refactor isc_stats_create() and its variants in libdns,
libns and named to just return void.
1. Change the _new, _add and _copy functions to return the new object
instead of returning 'void' (or always ISC_R_SUCCESS)
2. Cleanup the isc_ht_find() + isc_ht_add() usage - the code is always
locked with catzs->lock (mutex), so when isc_ht_find() returns
ISC_R_NOTFOUND, the isc_ht_add() must always succeed.
3. Instead of returning direct iterator for the catalog zone entries,
add dns_catz_zone_for_each_entry2() function that calls callback
for each catalog zone entry and passes two extra arguments to the
callback. This will allow changing the internal storage for the
catalog zone entries.
4. Cleanup the naming - dns_catz_<fn>_<obj> -> dns_catz_<obj>_<fn>, as an
example dns_catz_new_zone() gets renamed to dns_catz_zone_new().
store a pointer to the running loop when creating a dispatch entry
with dns_dispatch_add(), and use isc_loop_now() to get the timestamp for
the current event loop tick when we initialize the dispentry start time
and check for timeouts.
ultimately we want the slab implementation of dns_rdataset to
be usable by more database implementaions than just rbtdb. this
commit moves rdataset_methods to rdataslab.c, renamed
dns_rdataslab_rdatasetmethods.
new database methods have been added: locknode, unlocknode,
addglue, expiredata, and deletedata, allowing external functions to
perform functions that previously required internal access to the
database implementation.
database and heap pointers are now stored in the dns_slabheader object
so that header is the only thing that needs to be passed to some
functions; this will simplify moving functions that process slabheaders
out of rbtdb.c so they can be used by other database implementations.
to reduce the amount of common code that will need to be shared
between the separated cache and zone database implementations,
clean up unused portions of dns_db.
the methods dns_db_dump(), dns_db_isdnssec(), dns_db_printnode(),
dns_db_resigned(), dns_db_expirenode() and dns_db_overmem() were
either never called or were only implemented as nonoperational stub
functions: they have now been removed.
dns_db_nodefullname() was only used in one place, which turned out
to be unnecessary, so it has also been removed.
dns_db_ispersistent() and dns_db_transfernode() are used, but only
the default implementation in db.c was ever actually called. since
they were never overridden by database methods, there's no need to
retain methods for them.
in rbtdb.c, beginload() and endload() methods are no longer defined for
the cache database, because that was never used (except in a few unit
tests which can easily be modified to use the zone implementation
instead). issecure() is also no longer defined for the cache database,
as the cache is always insecure and the default implementation of
dns_db_issecure() returns false.
for similar reasons, hashsize() is no longer defined for zone databases.
implementation functions that are shared between zone and cache are now
prepended with 'dns__rbtdb_' so they can become nonstatic.
serve_stale_ttl is now a common member of dns_db.
in preparation for splitting up rbtdb.c, rename some types so they
can be defined in dns/types.h instead of only locally. these include:
- struct noqname, which is used to hold no-qname and closest-encloser
proofs, and is now named dns_proof_t;
- rbtdb_rdatatype_t, which is used to hold a pair of rdatatypes and
is now called dns_typepair_t and defined in rdatatype.h;
- rbtdb_serial_t, which is now just a uint32_t;
- rdatasetheader_t and rdatasetheaderlist_t, now called
dns_slabheader_t and dns_slabheaderlist_t;
- rbtdb_version_t, now called dns_rbtdb_version_t.
the helper functions header_from_raw() and raw_from_header() are
renamed dns_slabheader_fromrdataset() and dns_slabheader_raw().
also made further style changes:
- fixing uninitialized pointer variables throughout rbtdb.c;
- switching some initializations to struct literals;
- renaming some functions and struct members more descriptively;
- replacing dns_db_secure_t with a simple bool since it no longer needs
to be tri-valued.
the DNS_DBFIND_VALIDATEGLUE and DNS_DBFIND_FORCENSEC options
were never set, so the code implementing them (which in the case
of _VALIDATEGLUE appears to have been quite outdated anyway) was
never reached. they have now been removed.
BIND's rdataset structure is a view of some DNS records. It is
polymorphic, so the details of how the records are stored can vary.
For instance, the records can be held in an rdatalist, or in an
rdataslab in the rbtdb.
The dns_rdataset structure previously had a number of fields called
`private1` up to `private7`, which were used by the various rdataset
implementations. It was not at all clear what these fields were for,
without reading the code and working it out from context.
This change makes the rdataset inheritance hierarchy more clear. The
polymorphic part of a `struct dns_rdataset` is now a union of structs,
each of which is named for the class of implementation using it. The
fields of these structs replace the old `privateN` fields. (Note: the
term "inheritance hierarchy" refers to the fact that the builtin and
SDLZ implementations are based on and inherit from the rdatalist
implementation, which in turn inherits from the generic rdataset.
Most of this change is mechanical, but there are a few extras.
In keynode.c there were a number of REQUIRE()ments that were not
necessary: they had already been checked by the rdataset method
dispatch code. On the other hand, In ncache.c there was a public
function which needed to REQUIRE() that an rdataset was valid.
I have removed lots of "reset iterator state" comments, because it
should now be clear from `target->iter = NULL` where before
`target->private5 = NULL` could have been doing anything.
Initialization is a bit neater in a few places, using C structure
literals where appropriate.
The pointer arithmetic for translating between an rdataslab header and
its raw contents is now fractionally safer.
- use externally accessible functions for attachnode/detachnode
so these functions can be moved outside rbtdb.c
- simplify and tidy up some other functions
- use struct initializers when appropriate
- remove the flag RDATASET_ATTR_RETAIN; it was never being set
- renamed the rdataset attributes to
- remove the 'slab_methods' rdataset implementation. this was
a reduced set of slab rdataset methods, omitting 'setownercase()'
and 'getownercase()'. we can get the identical result by using
an DNS_RDATASETATTR_KEEPCASE attribute in rdatasets that
shouldn't have their case modified, and then we only need one
set of rdataset methods.
in the past there was overlap between the fields used
as resolver fetch options and ADB addrinfo flags. this has
mostly been eliminated; now we can clean up the rest of
it and remove some confusing comments.
The ability to read legacy HMAC-MD5 K* keyfile pairs using algorithm
number 157 was accidentally lost when the algorithm numbers were
consolidated into a single block, in commit
09f7e0607a.
The assumption was that these algorithm numbers were only known
internally, but they were also used in key files. But since HMAC-MD5
got renumbered from 157 to 160, legacy HMAC-MD5 key files no longer
work.
Move HMAC-MD5 back to 157 and GSSAPI back to 160. Add exception for
GSSAPI to list_hmac_algorithms.
The dns_zone_catz_enable_db() and dns_zone_catz_disable_db()
functions can race with similar operations in the catz module
because there is no synchronization between the threads.
Add catz functions which use the view's catalog zones' lock
when registering/unregistering the database update notify callback,
and use those functions in the dns_zone module, instead of doing it
directly.
view->adb may be referenced while the view is shutting down as the
zone uses a weak reference to the view and examines view->adb but
dns_view_detach call dns_adb_detach to clear view->adb.
- style fixes and general tidying-up in tkey.c
- remove the unused 'intoken' parameter from dns_tkey_buildgssquery()
- remove an unnecessary call to dns_tkeyctx_create() in ns_server_create()
(the TKEY context that was created there would soon be destroyed and
another one created when the configuration was loaded).
since it is not necessary to find partial matches when looking
up names in a TSIG keyring, we can use a hash table instead of
an RBT to store them.
the tsigkey object now stores the key name as a dns_fixedname
rather than allocating memory for it.
the `name` parameter to dns_tsigkeyring_add() has been removed;
it was unneeded since the tsigkey object already contains a copy
of the name.
the opportunistic cleanup_ring() function has been removed;
it was only slowing down lookups.
the prior practice of passing a dns_name containing the
expanded name of an algorithm to dns_tsigkey_create() and
dns_tsigkey_createfromkey() is unnecessarily cumbersome;
we can now pass the algorithm number instead.
- remove the 'ring' parameter from dns_tsigkey_createfromkey(),
and use dns_tsigkeyring_add() to add key objects to a keyring instead.
- add a magic number to dns_tsigkeyring_t
- change dns_tsigkeyring_dumpanddetach() to dns_tsigkeyring_dump();
we now call dns_tsigkeyring_detach() separately.
- remove 'maxgenerated' from dns_tsigkeyring_t since it never changes.
use the ISC_REFCOUNT attach/detach implementation in dns/tsig.c
so that detailed tracing can be used during refactoring.
dns_tsig_keyring_t has been renamed dns_tsigkeyring_t so the type
and the attach/detach function names will match.
- style cleanups.
- simplify the function parameters to dns_tsigkey_create():
+ remove 'restored' and 'generated', they're only ever set to false.
+ remove 'creator' because it's only ever set to NULL.
+ remove 'inception' and 'expiry' because they're only ever set to
(0, 0) or (now, now), and either way, this means "never expire".
+ remove 'ring' because we can just use dns_tsigkeyring_add() instead.
- rename dns_keyring_restore() to dns_tsigkeyring_restore() to match the
rest of the functions operating on dns_tsigkeyring objects.
When create_fetch() in the dns_validator unit detects deadlock, it
returns DNS_R_NOVALIDSIG, but it didn't attach to the validator. The
other condition to returning result != ISC_R_SUCCESS would be error from
dns_resolver_createfetch(). The caller (in two places out of three)
would detect the error condition and always detach from the validator.
Move the dns_validator_detach() on dns_resolver_createfetch() error
condition to create_fetch() function and cleanup the extra detaches in
seek_dnskey() and get_dsset().
Remove the code implementing nonstardard behaviors that were formerly
needed to allow GSS-TSIG to work with Windows 2000, which passed
End-of-Life in 2010.
Deprecate the "oldgsstsig" command and "-o" command line option
to nsupdate; these are now treated as synonyms for "gsstsig" and "-g"
respectively.
it was possible to add a TSIG key to more than one TSIG
keyring at a time, and this was in fact happening with the
session key, which was generated once and then added to the
keyrings for each view as it was configured.
this has been corrected and a REQUIRE added to dns_tsigkeyring_add()
to prevent it from happening again.
The following code block repeats quite often:
if (rdata.type == dns_rdatatype_dnskey ||
rdata.type == dns_rdatatype_cdnskey ||
rdata.type == dns_rdatatype_cds)
Introduce a new function to reduce the repetition.
The raw zone is not supposed to be signed. DNSKEY records in a raw zone
should not trigger zone signing. The update code needs to be able to
identify when it is working on a raw zone. Add dns_zone_israw() and
dns_zone_issecure() enable it to do this. Also, we need to check the
case for 'auto-dnssec maintain'.
A `dns_qmpulti_t` no longer needs to know about its loopmgr. We no
longer keep a linked list of `dns_qpmulti_t` that have reclamation
work, and we no longer mark chunks with the phase in which they are to
be reclaimed. Instead, empty chunks are listed in an array in a
`qp_rcu_t`, which is passed to call_rcu().
If the resolver received a FORMERR response to a request with
an DNS COOKIE option present that echoes the option back, resend
the request without an DNS COOKIE option present.
After the dns_xfrin was changed to use network manager, the maximum
global (max-transfer-time-in) and idle (max-transfer-idle-in) times for
incoming transfers were turned inoperational because of missing
implementation.
Restore this functionality by implementing the timers for the incoming
transfers.
It should be floor(DNS_NAME_MAXWIRE / 2) + 1 == 128
The mistake was introduced in c6bf51492d because:
* I was refactoring an existing `DNS_MAX_LABELS` defined as 127
* There was a longstanding bug in `dns_name_isvalid()` which
checked the number of labels against 127U instead of 128
* I mistakenly thought `dns_name_isvalid()` was correct and
`dns_name_countlabels()` was incorrect, but the reverse was true.
After this commit, occurrances of `DNS_NAME_MAXLABELS` with value
128 are consistent with the use of 127 or 128 before commit
c6bf51492d except for the mistake in `dns_name_isvalid()`.
This commit adds a test case that checks the MAXLABELS case
in `dns_name_fromtext()` and `dns_name_isvalid()`.
This change makes the zone table lock-free for reads. Previously, the
zone table used a red-black tree, which is not thread safe, so the hot
read path acquired both the per-view mutex and the per-zonetable
rwlock. (The double locking was to fix to cleanup races on shutdown.)
One visible difference is that zones are not necessarily shut down
promptly: it depends on when the qp-trie garbage collector cleans up
the zone table. The `catz` system test checks several times that zones
have been deleted; the test now checks for zones to be removed from
the server configuration, instead of being fully shut down. The catz
test does not churn through enough zones to trigger a gc, so the zones
are not fully detached until the server exits.
After this change, it is still possible to improve the way we handle
changes to the zone table, for instance, batching changes, or better
compaction heuristics.
Revert refcount debug tracing (commit a8b29f0365), there are better
ways to do it.
Use the dns_qpmethods_t typedef where appropriate.
Some stylistic improvements.
It is sometimes necessary to access a qp-trie outside an isc_loop,
such as in tests or an isc_work callback. The best option was to use
a `dns_qpmulti_write()` transaction, but that has overheads that are
not necessary for read-only access, such as committing a new version
of the trie even when nothing changed.
So this commit adds a `dns_qpmulti_read()` transaction, which is
nearly as lightweight as a query transaction, but it takes the mutex
like a write transaction.
This is the first of the "fancy" searches that know how the DNS
namespace maps on to the structure of a qp-trie. For example, it will
find the closest enclosing zone in the zone tree.
The dns_adbentry_overquota() was violating the layers accessing the
adbentry struct members directly. Change it to dns_adb_overquota() to
match the dns_adb API.
use the ISC_REFCOUNT implementation for dns_zone_attach() and
_detach(). (this applies only to external zone references, not
to dns_zone_iattach() and dns_zone_idetach().)
use dns_zone_ref() where previously a dummy zone object had been
used to increment the reference count.
My original idea had been that the core qp-trie code would be mostly
independent of the storage for keys, so I did not make it check at run
time that key lengths are sensible. However, the qp-trie search
routines need to get keys out of leaf objects, for which they provide
storage on the stack, which is particularly dangerous for unchecked
buffer overflows. So this change checks that key lengths are in bounds
at the API boundary between the qp-trie code and the rest of BIND, and
there is no more pretence that keys might be longer.
Add a new configuration option to set how the checkds method should
work. Acceptable values are 'yes', 'no', and 'explicit'.
When set to 'yes', the checkds method is to lookup the parental agents
by querying the NS records of the parent zone.
When set to 'no', no checkds method is enabled. Users should run
the 'rndc checkds' command to signal that DS records are published and
withdrawn.
When set to 'explicit', the parental agents are explicitly configured
with the 'parental-agents' configuration option.
This should have no functional effects.
The message size stats are specified by RSSAC002 so it's best not
to mess around with how they appear in the statschannel. But it's
worth changing the implementation to use general-purpose histograms,
to reduce code size and benefit from sharded counters.
Cleanup the remnants of MS Compiler bits from <isc/refcount.h>, printing
the information in named/main.c, and cleanup some comments about Windows
that no longer apply.
The bits in picohttpparser.{h,c} were left out, because it's not our
code.
The isc_fsaccess API was created to hide the implementation details
between POSIX and Windows APIs. As we are not supporting the Windows
APIs anymore, it's better to drop this API used in the DST part.
Moreover, the isc_fsaccess was setting the permissions in an insecure
manner - it operated on the filename, and not on the file descriptor
which can lead to all kind of attacks if unpriviledged user has read (or
even worse write) access to key directory.
Replace the code that operates on the private keys with code that uses
mkstemp(), fchmod() and atomic rename() at the end, so at no time the
private key files have insecure permissions.
The only place where dns_name_hash() was being used is the old hash
table in the dns_badcache unit. Squash the dns_name_fullhash() and
dns_name_hash() into single dns_name_hash() function that's always
case-insensitive as it doesn't make to do case-sensitive hashing of the
domain names and we were not using this anywhere.
Instead of marking the unused entities with UNUSED(x) macro in the
function body, use a `ISC_ATTR_UNUSED` attribute macro that expans to
C23 [[maybe_unused]] or __attribute__((__unused__)) as fallback.
Change the isc_job_run() to not-make any allocations. The caller must
make sure that it allocates isc_job_t - usually as part of the argument
passed to the callback.
For simple jobs, using isc_async_run() is advised as it allocates its
own separate isc_job_t.
for testing purposes, we need to be able to specify a library path from
which to load the dnsrps implementation. this can now be done with the
"dnsrps-library" option.
DNSRPS can now be enabled in configure regardless of whether librpz.so
is currently installed on the system.
the new dns_view_addtrustedkey() function allows a view's trust
anchors to be updated directly. this code was formerly in
dns_client_addtrustedkey(), which is now a wrapper around
dns_view_addtrustedkey().
stop and restart the server in the 'tsiggss' test, in order
to confirm that GSS negotiated TSIG keys are saved and restored
when named loads.
added logging to dns_tsigkey_createfromkey() to indicate whether
a key has been statically configured, generated via GSS negotiation,
or restored from a file.
REQUIRE that rdata->type is dns_rdatatype_svcb to detect when
dns_rdata_checksvcb is called with the wrong rdata type. There are
no code paths that currently pass the wrong rdata to dns_rdata_checksvcb.
This was found by GCC 12 static analysis.
without diffie-hellman TKEY negotiation, some other code is
now effectively dead or unnecessary, and can be cleaned up:
- the rndc tsig-list and tsig-delete commands.
- a nonoperational command-line option to dnssec-keygen that
was documented as being specific to DH.
- the section of the ARM that discussed TKEY/DH.
- the functions dns_tkey_builddeletequery(), processdeleteresponse(),
and tkey_processgssresponse(), which are unused.
Completely remove the TKEY Mode 2 (Diffie-Hellman Exchanged Keying) from
BIND 9 (from named, named.conf and all the tools). The TKEY usage is
fringe at best and in all known cases, GSSAPI is being used as it should.
The draft-eastlake-dnsop-rfc2930bis-tkey specifies that:
4.2 Diffie-Hellman Exchanged Keying (Deprecated)
The use of this mode (#2) is NOT RECOMMENDED for the following two
reasons but the specification is still included in Appendix A in case
an implementation is needed for compatibility with old TKEY
implementations. See Section 4.6 on ECDH Exchanged Keying.
The mixing function used does not meet current cryptographic
standards because it uses MD5 [RFC6151].
RSA keys must be excessively long to achieve levels of security
required by current standards.
We might optionally implement Elliptic Curve Diffie-Hellman (ECDH) key
exchange mode 6 if the draft ever reaches the RFC status. Meanwhile the
insecure DH mode needs to be removed.
There are leftovers from the previous refactoring effort, which left
some function declarations and comments in the header file unchanged.
Finish the renaming.
This implements node reference tracing that passes all the internal
layers from dns_db API (and friends) to increment_reference() and
decrement_reference().
It can be enabled by #defining DNS_DB_NODETRACE in <dns/trace.h> header.
The output then looks like this:
incr:node:check_address_records:rootns.c:409:0x7f67f5a55a40->references = 1
decr:node:check_address_records:rootns.c:449:0x7f67f5a55a40->references = 0
incr:nodelock:check_address_records:rootns.c:409:0x7f67f5a55a40:0x7f68304d7040->references = 1
decr:nodelock:check_address_records:rootns.c:449:0x7f67f5a55a40:0x7f68304d7040->references = 0
There's associated python script to find the missing detach located at:
https://gitlab.isc.org/isc-projects/bind9/-/snippets/1038
Now that we can configure a different digest type, update the code
to honor the configuration. Update 'dns_dnssec_syncupdate' so that
the correct CDS record is published, and also when deleting CDS records,
ensure that all possible CDS records are removed from the zone.
In general, it's better to do one thorough compaction when a batch of
work is complete, which is the way that `update` transactions work.
Conversely, `write` transactions are designed so that lots of little
transactions are not too inefficient, but they need explicit
compaction. This changes `dns_qp_compact()` so that it is easier to
compact any time that makes sense, if there isn't a better way to
schedule compaction. And `dns_qpmulti_commit()` only recycles garbage
when there is enough to make it worthwhile.
Add some qp-trie tracing macros which can be enabled by a
developer. These print a message when a leaf is attached or
detached, indicating which part of the qp-trie implementation
did so. The refcount methods must now return the refcount value
so it can be printed by the trace macros.
The first working multi-threaded qp-trie was stuck with an unpleasant
trade-off:
* Use `isc_rwlock`, which has acceptable write performance, but
terrible read scalability because the qp-trie made all accesses
through a single lock.
* Use `liburcu`, which has great read scalability, but terrible
write performance, because I was relying on `rcu_synchronize()`
which is rather slow. And `liburcu` is LGPL.
To get the best of both worlds, we need our own scalable read side,
which we now have with `isc_qsbr`. And we need to modify the write
side so that it is not blocked by readers.
Better write performance requires an async cleanup function like
`call_rcu()`, instead of the blocking `rcu_synchronize()`. (There
is no blocking cleanup in `isc_qsbr`, because I have concluded
that it would be an attractive nuisance.)
Until now, all my multithreading qp-trie designs have been based
around two versions, read-only and mutable. This is too few to
work with asynchronous cleanup. The bare minimum (as in epoch
based reclamation) is three, but it makes more sense to support an
arbitrary number. Doing multi-version support "properly" makes
fewer assumptions about how safe memory reclamation works, and it
makes snapshots and rollbacks simpler.
To avoid making the memory management even more complicated, I
have introduced a new kind of "packed reader node" to anchor the
root of a version of the trie. This is simpler because it re-uses
the existing chunk lifetime logic - see the discussion under
"packed reader nodes" in `qp_p.h`.
I have also made the chunk lifetime logic simpler. The idea of a
"generation" is gone; instead, chunks are either mutable or
immutable. And the QSBR phase number is used to indicate when a
chunk can be reclaimed.
Instead of the `shared_base` flag (which was basically a one-bit
reference count, with a two version limit) the base array now has a
refcount, which replaces the confusing ad-hoc lifetime logic with
something more familiar and systematic.
