ComputerSurge/bind9
3
0
Fork 0
Code Pull Requests 1 Activity
31,092 Commits 589 Branches 805 Tags
0abcface49789c5fdc28fec062cdf692c43a3e7a
Commit Graph

31092 Commits

This Branch
This Branch
All Branches
Author SHA1 Message Date
Mark Andrews
0abcface49 properly wait for zone to be loaded 2020-03-04 15:39:08 +11:00
Mark Andrews
5cc912b031 Merge branch 'marka-check-touched' into 'master' 
Check that inline master zones that are only touched are correctly reloaded.

Closes #1627

See merge request isc-projects/bind9!3135
 2020-03-03 22:06:39 +00:00
Mark Andrews
64474db14e Add CHANGES entry 2020-03-04 08:46:33 +11:00
Mark Andrews
f171347b5f Restart zone maintenance if it had been stalled. 2020-03-04 08:45:34 +11:00
Mark Andrews
50e079d6c3 Check inline master zones that are touched reload correctly. 2020-03-04 08:45:34 +11:00
Ondřej Surý
bb74fdcaf0 Merge branch 'ondrej/fix-notes-9.17.0-xml' into 'master' 
fixup the version number in 9.17.0 release notes

See merge request isc-projects/bind9!3145
 2020-03-03 14:24:17 +00:00
Ondřej Surý
3357574843 fixup the version number in 9.17.0 release notes 2020-03-03 09:31:34 +01:00
Witold Krecicki
3a21933586 Merge branch 'wpk/socket-destroy-race' into 'master' 
Fix a race in isc_socket destruction.

Closes #1651

See merge request isc-projects/bind9!3004
 2020-03-03 08:20:44 +00:00
Witold Kręcicki
81ba0fe0e6 Fix a race in isc_socket destruction. 
There was a very slim chance of a race between isc_socket_detach and
process_fd: isc_socket_detach decrements references to 0, and before it
calls destroy gets preempted. Second thread calls process_fd, increments
socket references temporarily to 1, and then gets preempted, first thread
then hits assertion in destroy() as the reference counter is now 1 and
not 0.
 2020-03-03 08:34:19 +01:00
Mark Andrews
3c15372c2b Merge branch '1627-call-set_resigntime-and-zone_settimer' into 'master' 
Call set resigntime and zone settimer

Closes #1627

See merge request isc-projects/bind9!3137
 2020-03-03 05:07:58 +00:00
Mark Andrews
23ce0f32e5 add release note entry 2020-03-03 15:37:42 +11:00
Mark Andrews
905549f061 Add CHANGES note 2020-03-03 15:30:52 +11:00
Ondřej Surý
aaaa7e67fa Adjust lock/unlock in zone_rekey() 2020-03-03 15:26:31 +11:00
Mark Andrews
a6626594aa Ensure that resigintime and the zone timer are set 2020-03-03 15:26:27 +11:00
Matthijs Mekking
143d1c9767 Add more zone locks 
Add more zone locks around code that touches zone timer setting
in failure modes.
 2020-03-03 15:26:27 +11:00
Mark Andrews
7212961849 Always call set_resigntime with the zone lock held 2020-03-03 15:26:27 +11:00
Mark Andrews
5ec57f31b0 Always call zone_settimer() 
zone_needdump() could potentially not call zone_settimer() so
explitly call zone_settimer() as zone->resigntime could have
gone backward.
 2020-03-03 15:26:27 +11:00
Mark Andrews
5d1611afdc Call set_resigntime() in receive_secure_serial() 
With RRSIG records no longer being signed with the full
sig-validity-interval we need to ensure the zone->resigntime
as it may need to be set to a earlier time.
 2020-03-03 15:26:27 +11:00
Witold Krecicki
9aa84dabf9 Merge branch 'wpk/perfwork-7-multilocked-badcache' into 'master' 
Badcache with multiple locks.

See merge request isc-projects/bind9!3076
 2020-03-02 17:18:35 +00:00
Witold Kręcicki
6c8f309745 Workaround for clang static analyzer bug. 2020-03-02 14:12:05 +01:00
Witold Kręcicki
47e5f5564c Badcache with multiple locks. 
Previously badcache used one single mutex for everything, which
was causing performance issues. Use one global rwlock for the whole
hashtable and per-bucket mutexes.
 2020-03-02 12:13:09 +01:00
Michał Kępień
c309fcabec Merge branch '1640-add-ZLIB_LIBS-to-ISCLIBS' into 'master' 
Add ZLIB_LIBS to ISCLIBS

Closes #1640

See merge request isc-projects/bind9!3117
 2020-02-28 14:30:04 +00:00
Michał Kępień
fc967ba092 Add ZLIB_LIBS to ISCLIBS 
When --with-zlib is passed to ./configure (or when the latter
autodetects zlib's presence), libisc uses certain zlib functions and
thus libisc's users should be linked against zlib in that case.  Adjust
Makefile variables appropriately to prevent shared build failures caused
by underlinking.
 2020-02-28 15:22:29 +01:00
Witold Krecicki
74030a590f Merge branch 'wpk/perfwork-3-small-nits' into 'master' 
Perfwork 3/6 - Various small nits and tunables

See merge request isc-projects/bind9!3067
 2020-02-28 08:28:31 +00:00
Evan Hunt
0b76d8a490 comments 2020-02-28 08:46:16 +01:00
Witold Kręcicki
4b6a064972 Don't define NS_CLIENT_TRACE by default 2020-02-28 08:46:16 +01:00
Witold Kręcicki
4791263def Increase inactivehandles and inactivereqs size for better reuse. 2020-02-28 08:46:16 +01:00
Witold Kręcicki
0344684385 Increase nodelock count for both cache and regular db. 2020-02-28 08:46:16 +01:00
Witold Kręcicki
0d80266f7e Use RESOLVER_NTASKS_PERCPU - 32 for regular tuning, 8 for small 2020-02-28 08:46:16 +01:00
Witold Kręcicki
517e6eccdf use SO_INCOMING_CPU for UDP sockets 2020-02-28 08:46:16 +01:00
Witold Kręcicki
8c6c07286f Remove some stale fields from ns_client_t; make sendbuf allocated on heap 2020-02-28 08:46:16 +01:00
Witold Kręcicki
fe584c01cc Don't update LRU if the node was recently used. 
Updating LRU requires write-locking the node, which causes contention.
Update LRU only if time difference is large enough.
 2020-02-28 08:46:16 +01:00
Witold Kręcicki
a658f7976c We don't need to fill udp local address every time since we are bound to it. 2020-02-28 08:46:16 +01:00
Witold Kręcicki
938b61405b Don't check if the client is on recursing list (requires locking) if it's not RECURSING 2020-02-28 08:46:16 +01:00
Witold Kręcicki
eb874608c1 Use the original threadid when sending a UDP packet to decrease probability of context switching 2020-02-28 08:46:16 +01:00
Mark Andrews
c324230c6f Merge branch '1638-rrsig-soa-and-re-signing' into 'master' 
Resolve "RRSIG(SOA) and re-signing."

Closes #1638

See merge request isc-projects/bind9!3114
 2020-02-27 22:48:48 +00:00
Mark Andrews
88c828cb9f capture named-journalprint output 2020-02-27 22:30:14 +00:00
Mark Andrews
a24fd55836 sort RRSIG(SOA) to be last of RRSIGs with a common re-resign time 2020-02-27 22:30:14 +00:00
Mark Andrews
660dc3eba7 use the full sig-validity-interval for RRSIG(SOA) 2020-02-27 22:30:14 +00:00
Mark Andrews
563aff110e Merge branch 'marka-coverity-stop-taint' into 'master' 
Simplify hash computation to prevent pointer being classed as tainted.

See merge request isc-projects/bind9!3081
 2020-02-27 20:01:04 +00:00
Mark Andrews
8c983a7ebd Simplify hash computation to prevent pointer being classed as tainted. 
mem.c:add_trace_entry() -> isc_hash_function() -> isc_siphash24()

129        for (; in != end; in += 8) {

	6. byte_swapping: Performing a byte swapping operation on
	in implies that it came from an external source, and is
	therefore tainted.

130                uint64_t m = U8TO64_LE(in);
 2020-02-27 19:41:36 +00:00
Witold Krecicki
ebfb0f7392 Merge branch 'wpk/perfwork-5-isc-result-rwlock' into 'master' 
Perfwork 5/6 - Use isc_rwlock for isc_result tables

See merge request isc-projects/bind9!3069
 2020-02-27 09:48:11 +00:00
Witold Kręcicki
00f2146265 Use isc_rwlock for isc_result tables 2020-02-27 07:58:48 +00:00
Evan Hunt
d434ec89ce Merge branch 'each-speedup-serve-stale-test' into 'master' 
Speed up serve-stale test

See merge request isc-projects/bind9!3090
 2020-02-27 06:57:56 +00:00
Evan Hunt
195d25b222 send bursts of queries in parallel to reduce test runtime 
sending each group of queries simultaneously, and then checking the
output after the last one finishes, reduces the runtime of the
serve-stale test by about six minutes.
 2020-02-26 22:28:33 -08:00
Evan Hunt
027601cd3e stop testing 'serve-stale yes' and 'serve-fail no' alternate syntax 
"yes" and "no" are permissible synonyms for "on" and "off", which
use exactly the same code paths. making sure they work isn't a good
use of 80 seconds of test time.
 2020-02-26 22:28:33 -08:00
Evan Hunt
770bf4980f use "on|off" in rndc -h output to match the man page 2020-02-26 22:28:33 -08:00
Witold Krecicki
027f445c66 Merge branch '1623-assertion-failure-ns-client-endrequest' into 'master' 
Don't issue ns_client_endrequest on a NS_CLIENTSTATE_READY client.

Closes #1623

See merge request isc-projects/bind9!3077
 2020-02-26 12:35:19 +00:00
Witold Kręcicki
b0888ff039 Don't issue ns_client_endrequest on a NS_CLIENTSTATE_READY client. 
Fix a potential assertion failure on shutdown in ns__client_endrequest.
Scenario:
1. We are shutting down, interface->clientmgr is gone.
2. We receive a packet, it gets through ns__client_request
3. mgr == NULL, return
4. isc_nmhandle_detach calls ns_client_reset_cb
5. ns_client_reset_cb calls ns_client_endrequest
6. INSIST(client->state == NS_CLIENTSTATE_WORKING ||
          client->state == NS_CLIENTSTATE_RECURSING) is not met
   - we haven't started processing this packet so
   client->state == NS_CLIENTSTATE_READY.
As a solution - don't do anything in ns_client_reset_cb if the client
is still in READY state.
 2020-02-26 12:15:01 +00:00
Witold Krecicki
6d67e06665 Merge branch 'wpk/perfwork-4-compressctx-arena' into 'master' 
Perfwork 4/6 - add an arena to compressctx

See merge request isc-projects/bind9!3068
 2020-02-26 08:17:28 +00:00