Even though statistics are lockless they still use atomics which
might cause contention. Split stats counters into buckets, sharded
by an artificial thread identifier, to increase throughput.
isc_task_pause/unpause were inherently thread-unsafe - a task
could be paused only once by one thread, if the task was running
while we paused it it led to races. Fix it by making sure that
the task will pause if requested to, and by using a 'pause reference
counter' to count task pause requests - a task will be unpaused
iff all threads unpause it.
Don't remove from queue when pausing task - we lock the queue lock
(expensive), while it's unlikely that the task will be running -
and we'll remove it anyway in dispatcher
Add checks to the kasp system test to verify CDNSKEY publication.
This test is not entirely complete, because when there is a CDNSKEY
available but there should not be one for KEY N, it is hard to tell
whether the existing CDNSKEY actually belongs to KEY N or another
key.
The check works if we expect a CDNSKEY although we cannot guarantee
that the CDNSKEY is correct: The test verifies existence, not
correctness of the record.
When you do a restart or reconfig of named, or rndc loadkeys, this
triggers the key manager to run. The key manager will check if new
keys need to be created. If there is an active key, and key rollover
is scheduled far enough away, no new key needs to be created.
However, there was a bug that when you just start to sign your zone,
it takes a while before the KSK becomes an active key. An active KSK
has its DS submitted or published, but before the key manager allows
that, the DNSKEY needs to be omnipresent. If you restart named
or rndc loadkeys in quick succession when you just started to sign
your zone, new keys will be created because the KSK is not yet
considered active.
Fix is to check for introducing as well as active keys. These keys
all have in common that their goal is to become omnipresent.
In system tests on Windows tool's local port can sometimes clash with
'named'. On Unix the system is poked for the minimal local port,
otherwise is set to 32768 as a sane minimum. For Windows we don't
poke but set a hardcoded limit; this change aligns the limit with
Unix and changes it to 32768.
10067 cleanup:
CID 1452683 (#1 of 1): Dereference before null check (REVERSE_INULL)
check_after_deref: Null-checking dispatch suggests that it
may be null, but it has already been dereferenced on all
paths leading to the check.
10068 if (dispatch != NULL)
10069 isc_mem_put(server->mctx, dispatch, sizeof(*dispatch));
1549 cleanup:
1550 if (dctx->dbiter != NULL)
1551 dns_dbiterator_destroy(&dctx->dbiter);
1552 if (dctx->db != NULL)
1553 dns_db_detach(&dctx->db);
CID 1452686 (#1 of 1): Dereference before null check (REVERSE_INULL)
check_after_deref: Null-checking dctx suggests that it may
be null, but it has already been dereferenced on all paths
leading to the check.
1554 if (dctx != NULL)
1555 isc_mem_put(mctx, dctx, sizeof(*dctx));
707 complete_allnds:
CID 1452689 (#1 of 1): Dereference before null check (REVERSE_INULL)
check_after_deref: Null-checking dir_list suggests that it
may be null, but it has already been dereferenced on all
paths leading to the check.
708 if (dir_list != NULL) {
709 /* clean up entries from list. */
389 else
CID 1452695 (#1 of 1): Dereference before null check (REVERSE_INULL)
check_after_deref: Null-checking lcfg suggests that it may
be null, but it has already been dereferenced on all paths
leading to the check.
390 if (lcfg != NULL)
391 isc_logconfig_destroy(&lcfg);
122 cleanup:
CID 1452696 (#1 of 1): Dereference before null check (REVERSE_INULL)
check_after_deref: Null-checking s suggests that it may be
null, but it has already been dereferenced on all paths
leading to the check.
123 if (s != NULL)
124 isc_mem_free(mctx, s);
255 flag_fail:
256 /* get rid of what was build of the query list */
CID 1452697 (#1 of 1): Dereference before null check (REVERSE_INULL)
check_after_deref: Null-checking tql suggests that it may
be null, but it has already been dereferenced on all paths
leading to the check.
257 if (tql != NULL)
258 destroy_querylist(mctx, &tql);
6412 cleanup:
6413 dns_rdataset_disassociate(&neg);
6414 dns_rdataset_disassociate(&negsig);
CID 1452700 (#1 of 1): Dereference before null check (REVERSE_INULL)
check_after_deref: Null-checking closest suggests that it
may be null, but it has already been dereferenced on all
paths leading to the check.
6415 if (closest != NULL)
6416 free_noqname(mctx, &closest);
336 cleanup_mem:
337 /* cleanup memory */
338
339 /* free tmpPath memory */
CID 1452701 (#1 of 1): Dereference before null check (REVERSE_INULL)
check_after_deref: Null-checking tmpPath suggests that it
may be null, but it has already been dereferenced on all
paths leading to the check.
340 if (tmpPath != NULL && result != ISC_R_SUCCESS)
341 isc_mem_free(named_g_mctx, tmpPath);
342
343 /* free tmpPath memory */
344 return (result);
13429 cleanup:
13430 cancel_refresh(zone);
CID 1452702 (#1 of 1): Dereference before null check (REVERSE_INULL)
check_after_deref: Null-checking stub suggests that it may
be null, but it has already been dereferenced on all paths
leading to the check.
13431 if (stub != NULL) {
13432 stub->magic = 0;
6367cleanup:
6368 dns_rdataset_disassociate(&neg);
6369 dns_rdataset_disassociate(&negsig);
CID 1452704 (#1 of 1): Dereference before null check
(REVERSE_INULL) check_after_deref: Null-checking noqname
suggests that it may be null, but it has already been
dereferenced on all paths leading to the check.
6370 if (noqname != NULL)
6371 free_noqname(mctx, &noqname);
11030 cleanup:
CID 1452705 (#1 of 1): Dereference before null check
(REVERSE_INULL) check_after_deref: Null-checking dctx
suggests that it may be null, but it has already been
dereferenced on all paths leading to the check.
11031 if (dctx != NULL)
11032 dumpcontext_destroy(dctx);
11033 return (result);
1401 }
CID 1453455 (#1 of 1): Dereference before null check (REVERSE_INULL)
check_after_deref: Null-checking event suggests that it may be null,
but it has already been dereferenced on all paths leading to the check.
1402 if (event != NULL)
1403 isc_event_free(ISC_EVENT_PTR(&event));
13836 if (zone != NULL)
13837 dns_zone_detach(&zone);
null: At condition dz != NULL, the value of dz must be NULL.
dead_error_condition: The condition dz != NULL cannot be true.
13838 if (dz != NULL) {
CID 1453456 (#1 of 1): Logically dead code (DEADCODE)
dead_error_begin: Execution cannot reach this statement:
dns_zone_detach(&dz->zone);.
13839 dns_zone_detach(&dz->zone);
13840 isc_mem_put(named_g_mctx, dz, sizeof(*dz));
13841 }
128 return (ISC_R_SUCCESS);
129
CID 1456146 (#1 of 1): Structurally dead code (UNREACHABLE)
unreachable: This code cannot be reached: {
if (dst->labels[i] != N....
130 do {
402 ctx->serve_stale_ttl = 0;
notnull: At condition indentctx, the value of indentctx
cannot be NULL. dead_error_condition: The condition indentctx
must be true.
CID 1456147 (#1 of 1): Logically dead code (DEADCODE)
dead_error_line: Execution cannot reach the expression
default_indent inside this statement: ctx->indent = (indentctx
? ....
403 ctx->indent = indentctx ? *indentctx : default_indent;
1636 cleanup:
CID 1458130 (#1 of 1): Dereference before null check (REVERSE_INULL)
check_after_deref: Null-checking buffer suggests that it may be
null, but it has already been dereferenced on all paths leading to
the check.
1637 if (buffer != NULL)
1638 isc_buffer_free(&buffer);
Found by LGTM.com (see below for description), and while it should not
happen as EDNS OPT RDLEN is uint16_t, the fix is easy. A little bit
of cleanup is included too.
> In a loop condition, comparison of a value of a narrow type with a value
> of a wide type may result in unexpected behavior if the wider value is
> sufficiently large (or small). This is because the narrower value may
> overflow. This can lead to an infinite loop.
Increase the short lived record TTL and negative SOA TTL to make
this test less vulnerable to timing issues. The drawback is that we
also have to sleep longer in this test.
Add queries and checks for CAA RRtype in the serve-stale test.
Ensure that the "Others" rrtype stat counter is incremented and
decremented properly if the RRset becomes stale/ancient.
The low max-stale-ttl config option needs to be increased in order
to match the timing when things expire (aka become ancient).
This commit simplifies the cachedb rrset statistics in two ways:
- Introduce new rdtypecounter arithmetics, allowing bitwise
operations.
- Remove the special DLV statistic counter.
New rdtypecounter arithmetics
-----------------------------
"The rdtypecounter arithmetics is a brain twister". Replace the
enum counters with some defines. A rdtypecounter is now 8 bits for
RRtypes and 3 bits for flags:
0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15
+--+--+--+--+--+--+--+--+--+--+--+--+--+--+--+--+
| | | | | | S |NX| RRType |
+--+--+--+--+--+--+--+--+--+--+--+--+--+--+--+--+
If the 8 bits for RRtype are all zero, this is an Other RRtype.
Bit 7 is the NXRRSET (NX) flag and indicates whether this is a
positive (0) or a negative (1) RRset.
Then bit 5 and 6 mostly tell you if this counter is for an active,
stale, or ancient RRtype:
S = 0x00 means Active
S = 0x01 means Stale
S = 0x10 means Ancient
Since a counter cannot be stale and ancient at the same time, we
treat S = 0x11 as a special case to deal with NXDOMAIN counters.
S = 0x11 indicates an NXDOMAIN counter and in this case the RRtype
field signals the expiry of this cached item:
RRType = 0 means Active
RRType = 1 means Stale
RRType = 2 means Ancient
This also removes counting the DLV RRtype separately. Since we have
deprecated the lookaside validation it makes no sense to keep this
special statistic counter.
Since OpenBSD 6.6 is the current OpenBSD release, replace OpenBSD 6.5
GitLab CI jobs with their up-to-date counterparts.
As CI jobs for OpenBSD 6.6 will be run by a generalized libvirt executor
rather than an OpenBSD-specific one, make the necessary tag and variable
adjustments as well.
- Add quotes before and after zone name when generating "addzone"
input so avoid "unexpected token" errors.
- Use a hex digest for zone filenames when the zone or view name
contains a slash.
- Test with a domain name containing a slash.
- Incidentally added 'catzhash.py' to contrib/scripts to generate
hash labels for catalog zones, as it was needed to write the test.
The isc_buffer_allocate() function now cannot fail with ISC_R_MEMORY.
This commit removes all the checks on the return code using the semantic
patch from previous commit, as isc_buffer_allocate() now returns void.
The isc_mempool_create() function now cannot fail with ISC_R_MEMORY.
This commit removes all the checks on the return code using the semantic
patch from previous commit, as isc_mempool_create() now returns void.
Each system test can be marked as failed not only due to some tested
component(s) not behaving as expected, but also because of core dumps,
assertion failures, and/or ThreadSanitizer reports being found among its
artifacts. Make the system test summary list the tests which exhibit
such atypical symptoms to more clearly present the nature of problems
found.
The `rndc signing -clear` command cleans up the private-type records
that keep track of zone signing activity, but before this change it
did not tell the secondary servers that the zone has changed.
Resolve "bind 9.14.8 and 9.14.9 aborts when queried for non-existing domain in chaos class"
Closes#1569 and #1540
See merge request isc-projects/bind9!2843
Added test to ensure that NXDOMAIN is returned when BIND is queried for a
non existing domain in CH class (if a view of CHAOS class is configured)
and that it also doesn't crash anymore in those cases.
Function dns_view_findzonecut in view.c wasn't correctly handling
classes other than IN (chaos, hesiod, etc) whenever the name being
looked up wasn't in cache or in any of the configured zone views' database.
That resulted in a NULL fname being used in resolver.c:4900, which
in turn was triggering abort.
When both 'broken' and 'failed' test cases appear in unit test output
...
===> Broken tests
lib/isc/tests/socket_test:main -> broken: Test case timed out [300.022s]
===> Failed tests
lib/isc/tests/time_test:main -> failed: 2 of 6 tests failed [0.006s]
===> Summary
...
spurious '===>' string gets matched, that results in the following
error:
Usage error for command debug: '===>' is not a test case identifier (missing ':'?).
Following change makes sure the string is omitted.
I checked on FreeBSD and OpenBSD that the AWK construct is supported.
If we created a key, mark its SyncPublish time as 'now' and started
bind the key might not be published if the SyncPublish time is in
the same second as the time the zone is loaded. This is mostly
for dnssec system test, as this kind of scenario is very unlikely
in a real world environment.
To reproduce the race - create a task, send two events to it, first one
must take some time. Then, from the outside, pause(), unpause() and detach()
the task.
When the long-running event is processed by the task it is in
task_state_running state. When we called pause() the state changed to
task_state_paused, on unpause we checked that there are events in the task
queue, changed the state to task_state_ready and enqueued the task on the
workers readyq. We then detach the task.
The dispatch() is done with processing the event, it processes the second
event in the queue, and then shuts down the task and frees it (as it's not
referenced anymore). Dispatcher then takes the, already freed, task from
the queue where it was wrongly put, causing an use-after free and,
subsequently, either an assertion failure or a segmentation fault.
The probability of this happening is very slim, yet it might happen under a
very high load, more probably on a recursive resolver than on an
authoritative.
The fix introduces a new 'task_state_pausing' state - to which tasks
are moved if they're being paused while still running. They are moved
to task_state_paused state when dispatcher is done with them, and
if we unpause a task in paused state it's moved back to task_state_running
and not requeued.
This is a bug I encountered when trying to schedule an algorithm
rollover. My plan, for a zone whose maximum TTL is 48h, was to sign
with the new algorithm and schedule a change of CDS records for more
than 48 hours in the future, roughly like this:
$ dnssec-keygen -a 13 -fk -Psync now+50h $zone
$ dnssec-keygen -a 13 $zone
$ dnssec-settime -Dsync now+50h $zone_ksk_old
However the algorithm 13 CDS was published immediately, which could
have made the zone bogus.
To reveal the bug using the `smartsign` test, this change just adds a
KSK with all its times in the future, so it should not affect the
existing checks at all. But the final check (that there are no CDS or
CDSNSKEY records after -Dsync) fails with the old `syncpublish()`
logic, because the future key's sync records appear early. With the
new `syncpublish()` logic the future key does not affect the test, as
expected, and it now passes.
When two threads unreferenced handles coming from one socket while
the socket was being destructed we could get a use-after-free:
Having handle H1 coming from socket S1, H2 coming from socket S2,
S0 being a parent socket to S1 and S2:
Thread A Thread B
Unref handle H1 Unref handle H2
Remove H1 from S1 active handles Remove H2 from S2 active handles
nmsocket_maybe_destroy(S1) nmsocket_maybe_destroy(S2)
nmsocket_maybe_destroy(S0) nmsocket_maybe_destroy(S0)
LOCK(S0->lock)
Go through all children, figure
out that we have no more active
handles:
sum of S0->children[i]->ah == 0
UNLOCK(S0->lock)
destroy(S0)
LOCK(S0->lock)
- but S0 is already gone
We weren't consistent about who should unreference the handle in
case of network error. Make it consistent so that it's always the
client code responsibility to unreference the handle - either
in the callback or right away if send function failed and the callback
will never be called.
If taskmgr is shutting down ns_client_setup will fail to create
a task for the newly created client, we weren't cleaning up already
created/attached things (memory context, server, clientmgr).
In tcp and udp stoplistening code we accessed libuv structures
from a different thread, which caused a shutdown crash when named
was under load. Also added additional DbC checks making sure we're
in a proper thread when accessing uv_ functions.
We had a race in which n UDP socket could have been already closing
by libuv but we still sent data to it. Mark socket as not-active
when stopping listening and verify that socket is not active when
trying to send data to it.
if validator_start() is called with validator->event->message set to
NULL, we can't use message->rcode to decide which negative proofs are
needed, so we use the rdataset attributes instead to determine whether
the rdataset was cached as NXDOMAIN or NODATA.
We pass interface as an opaque argument to tcpdns listening socket.
If we stop listening on an interface but still have in-flight connections
the opaque 'interface' is not properly reference counted, and we might
hit a dead memory. We put just a single source of truth in a listening
socket and make the child sockets use that instead of copying the
value from listening socket. We clean the callback when we stop listening.
- isc__netievent_storage_t was to small to contain
isc__netievent__socket_streaminfo_t on Windows
- handle isc_uv_export and isc_uv_import errors properly
- rewrite isc_uv_export and isc_uv_import on Windows
hp implementation requires an object for each thread accessing
a hazard pointer. previous implementation had a hardcoded
HP_MAX_THREAD value of 128, which failed on machines with lots of
CPU cores (named uses 3n threads). We make isc__hp_max_threads
configurable at startup, with the value set to 4*named_g_cpus.
It's also important for this value not to be too big as we do
linear searches on a list.
it now removes matching trust anchors from from the dslist while leaving
the other trust anchors in place.
also cleaned up the API to remove functions that were never being used.
NOTE: the keytable test is still failing because dns_keytable_deletekey()
is looking for exact matches in keynodes containing dst_key objects,
which no keynode has anymore.
the internal keytable structure has not yet been changed, but
insertion of DS anchors is the only method now available.
NOTE: the keytable unit test is currently failing because of tests
that expect individual keynode objects to contain single DST key
objects.
as initial-key and static-key trust anchors will now be stored as a
DS rrset, code referencing keynodes storing DNSKEY trust anchors will
no longer be reached.
this function is used by dns_view_untrust() to handle revoked keys, so
it will still be needed after the keytable/validator refactoring is
complete, even though the keytable will be storing DS trust anchors
instead of keys. to simplify the way it's called, it now takes a DNSKEY
rdata struct instead of a DST key.
The isc_refcount API that provides reference counting lost DbC checks for
overflows and underflows in the isc_refcount_{increment,decrement} functions.
The commit restores the overflow check in the isc_refcount_increment and
underflows check in the isc_refcount_decrement by checking for the previous
value to not be on the boundary.
This commits removes superfluous checks when using the isc_refcount API.
Examples of superfluous checks:
1. The isc_refcount_decrement function ensures there was not underflow,
so this check is superfluous:
INSIST(isc_refcount_decrement(&r) > 0);
2 .The isc_refcount_destroy() includes check whether the counter
is zero, therefore this is superfluous:
INSIST(isc_refcount_decrement(&r) == 1 && isc_refcount_destroy(&r));
If a connection was closed early (right after accept()) an assertion
that assumed that the connection was still alive could be triggered
in accept_connection. Handle those errors properly and not with
assertions, free all the resources afterwards.
- the socket stat counters have been moved from socket.h to stats.h.
- isc_nm_t now attaches to the same stats counter group as
isc_socketmgr_t, so that both managers can increment the same
set of statistics
- isc__nmsocket_init() now takes an interface as a paramter so that
the address family can be determined when initializing the socket.
- based on the address family and socket type, a group of statistics
counters will be associated with the socket - for example, UDP4Active
with IPv4 UDP sockets and TCP6Active with IPv6 TCP sockets. note
that no counters are currently associated with TCPDNS sockets; those
stats will be handled by the underlying TCP socket.
- the counters are not actually used by netmgr sockets yet; counter
increment and decrement calls will be added in a later commit.
If pktinfo were supported then we could listen on :: for ipv6 and get
the information about the destination address from pktinfo structure passed
in recvmsg but this method is not portable and libuv doesn't support it - so
we need to listen on all interfaces.
We should verify that this doesn't impact performance (we already do it for
ipv4) and either remove all the ipv6pktinfo detection code or think of fixing
libuv.
- use UV_{TC,UD}P_IPV6ONLY for IPv6 sockets, keeping the pre-netmgr
behaviour.
- add a new listening_error bool flag which is set if the child
listener fails to start listening. This fixes a bug where named would
hang if, e.g., we failed to bind to a TCP socket.
For multithreaded TCP listening we need to pass a bound socket to all
listening threads. Instead of using uv_pipe handle passing method which
is quite complex (lots of callbacks, each of them with its own error
handling) we now use isc_uv_export() to export the socket, pass it as a
member of the isc__netievent_tcpchildlisten_t structure, and then
isc_uv_import() it in the child thread, simplifying the process
significantly.
These functions can be used to pass a uv handle between threads in a
safe manner. The other option is to use uv_pipe and pass the uv_handle
via IPC, which is way more complex. uv_export() and uv_import() functions
existed in libuv at some point but were removed later. This code is
based on the original removed code.
The Windows version of the code uses two functions internal to libuv;
a patch for libuv is attached for exporting these functions.
Ensure BIND is continuously tested on Tumbleweed, a pure rolling release
version of openSUSE. This will allow BIND incompatibilities with latest
upstream versions of its dependencies to be caught more quickly.
5339. [bug] With some libmaxminddb versions, named could erroneously
match an IP address not belonging to any subnet defined
in a given GeoIP2 database to one of the existing
entries in that database. [GL #1552]
Only comparing the value of the integer passed as the last argument to
MMDB_lookup_sockaddr() against MMDB_SUCCESS is not enough to ensure that
an MMDB lookup was successful - the 'found_entry' field of the
MMDB_lookup_result_s structure returned by that function also needs to
be true or else the remaining contents of that structure should be
ignored as the lookup failed. Extend the relevant logical condition in
get_entry_for() to ensure the latter does not return incorrect MMDB
entries for IP addresses which do not belong to any subnet defined in a
given GeoIP2 database.
Before this change, there was a missing blank line between the
negative trust anchors for one view, and the heading line for the next
view. This is because dns_ntatable_totext() omits the last newline.
There is an example of the incorrect output below; the fixed output
has a blank line before "Start view auth".
secure roots as of 21-Oct-2019 12:03:23.500:
Start view rec
Secure roots:
./RSASHA256/20326 ; managed
Negative trust anchors:
example.com: expiry 21-Oct-2019 13:03:15.000
Start view auth
Secure roots:
./RSASHA256/20326 ; managed
Negative trust anchors:
example.com: expiry 21-Oct-2019 13:03:07.000
Some unit tests need various managers to be created before they are run.
The interface manager spawned during libns tests listens on a fixed port
number, which causes intermittent issues when multiple tests using an
interface manager are run concurrently. Make the interface manager
listen on a randomized port number to greatly reduce the risk of
multiple unit tests using the same port concurrently.
"rndc signing -serial <value>" could take longer than a second to
complete. Loop waiting for update to succeed.
For tests where "rndc signing -serial <value>" is supposed to not
succeed, repeatedly test that we don't get the new serial, then
test that we have the old value. This should prevent false negatives.
GitLab issue and merge request numbers placed in release notes (in the
form of "#1234" for issues and "!5678" for merge requests) should not be
split across two lines. Extend the shell pipeline generating
doc/arm/notes.txt with a sed invocation which prevents such splitting.
The initial tcp statistics test was not testing tcp-highwater counter,
but only initial number of current TCP clients, so this missing test was
added to ensure initial tcp-highwater value is correct.
After the network manager rewrite, tcp-higwater stats was only being
updated when a valid DNS query was received over tcp.
It turns out tcp-quota is updated right after a tcp connection is
accepted, before any data is read, so in the event that some client
connect but don't send a valid query, it wouldn't be taken into
account to update tcp-highwater stats, that is wrong.
This commit fix tcp-highwater to update its stats whenever a tcp connection
is established, independent of what happens after (timeout/invalid
request, etc).
During BIND startup it scans for network interfaces available, in this
process it ensures that for every interface it will bind and listen to,
at least one socket will be always available accepting connections on
that interface, this way avoiding some DOS attacks that could exploit
tcp quota on some interface and make others unavailable.
In the previous network implementation this initial "reserved" tcp-quota
used by BIND was already been added to the tcp-highwater stats, but with
the new network code it was necesary to add this workaround to ensure
tcp-highwater stats reflect the tcp-quota used by BIND after startup.
2019-12-12 11:23:10 -08:00
373 changed files with 8617 additions and 6925 deletions
<holder>Internet Systems Consortium, Inc. ("ISC")</holder>
</copyright>
</docinfo>
Some files were not shown because too many files have changed in this diff
Show More
Reference in New Issue
Block a user
Blocking a user prevents them from interacting with repositories, such as opening or commenting on pull requests or issues. Learn more about blocking a user.
Witold Kręcicki
Michał Kępień
Matthijs Mekking
Michal Nowak
Mark Andrews
Ondřej Surý
Matthijs MekkingandMatthijs Mekking
Evan Hunt
Tony FinchandMark Andrews
Tinderbox User
Ondřej Surý
Logan CamposandOndřej Surý
Petr MenšíkandMark Andrews