Commit Graph
3881 Commits
Author SHA1 Message Date
Witold KręcickiandEvan Hunt af5f10fcf6 Disable qname minimization in RPZ test for now 2018-10-23 22:41:29 -04:00
Mark AndrewsandWitold Kręcicki 15c3bf9e82 stop.pl: report test being stopped 2018-10-23 12:15:04 +00:00
Mark AndrewsandWitold Kręcicki 23766ff690 checkpoint 2018-10-23 12:15:04 +00:00
Witold Kręcicki f2af336dc4 Fix looping issues 2018-10-23 12:15:04 +00:00
Witold Kręcicki 0246ea14c4 Make resolver tests more civilized 2018-10-23 12:15:04 +00:00
Mark AndrewsandWitold Kręcicki 98ab65dbe9 qmin: fixup expected queries for query for .ugly succeds when qname-minimization is in relaxed mode 2018-10-23 12:15:04 +00:00
Mark AndrewsandWitold Kręcicki 4754f53979 qmin: fix 'query for .ugly fails when qname-minimization is in strict mode'; remove debugging echos 2018-10-23 12:15:04 +00:00
Witold Kręcicki 70a1ba20ec QNAME miminimization should create a separate fetch context for each fetch -
this makes the cache more efficient and eliminates duplicates queries.
2018-10-23 12:15:04 +00:00
Witold KręcickiandMichał Kępień b5c9a8caad Set result to SERVFAIL if upstream responded with FORMERR
Commit ba91243542 causes the resolver to
respond to a client query with FORMERR when all upstream queries sent to
the servers authoritative for QNAME elicit FORMERR responses.  This
happens because resolver code returns DNS_R_FORMERR in such a case and
dns_result_torcode() acts as a pass-through for all arguments which are
already a valid RCODE.

The correct RCODE to set in the response returned to the client in the
case described above is SERVFAIL.  Make sure this happens by overriding
the RCODE in query_gotanswer(), on the grounds that any format errors in
the client query itself should be caught long before execution reaches
that point.  This change should not reduce query error logging accuracy
as the resolver code itself reports the exact reason for returning a
DNS_R_FORMERR result using log_formerr().
2018-10-23 13:50:27 +02:00
Ondřej Surý 09fd5c442b Convert the system tests that were using DSA to use a default algorithm defined in conf.sh 2018-10-05 11:16:13 +02:00
Ondřej Surý 3994b1f9c2 Remove support for obsoleted and insecure DSA and DSA-NSEC3-SHA1 algorithms 2018-10-05 09:21:43 +02:00
Evan Hunt 2f0897caff add a test case 2018-10-04 23:33:18 -07:00
Mark Andrews a216135a15 reclimit: delegate to ans4 so that ans2 can behave like a real name server 2018-10-04 12:59:38 +10:00
Mark Andrews 848e42227e chain: change nameserver name so that the delegation matches 2018-10-04 12:26:28 +10:00
Mark Andrews 524c1317f7 chain: add 'start test' messages to named.run 2018-10-04 12:26:28 +10:00
Evan Hunt ae8026171b update system test to add a check for UPDATE messages 2018-10-03 01:03:56 -07:00
Mark Andrews a6507af9e3 zero: send grep output to /dev/null; set ret=0 at start of 'check repeated recursive lookups of non recurring TTL=0 responses get new values' test so the failed subtest count is correct 2018-10-03 01:28:27 -04:00
Mark Andrews 0fc1b1bffa improve case presevation 2018-10-02 15:09:32 +10:00
Evan Hunt b026663c62 use an example domain 2018-09-28 09:58:07 -07:00
Evan Hunt b4dca44ad2 fix chaosnet address comparison 2018-09-28 00:42:03 -07:00
Evan Hunt 72b49ce869 tests for allow-recursion-on and allow-query-cache-on, default and inherited 2018-09-27 21:48:44 -07:00
Evan Hunt 6d05ec7dc1 fix problems running dnsrps tests in rpz and rpzrecurse
- set up server correctly when running the second test pass
- reset stats counters before the second pass
- clean up test output
2018-09-27 13:11:44 -07:00
Evan Hunt bf9c60078a don't use $() in system tests 2018-09-11 14:02:25 -04:00
Evan Hunt cc0abf62a1 add -class option to rndc nta 2018-09-10 12:04:04 -07:00
Evan Hunt 83dc5a704a report when NTAs added to multiple views
- the text returned by "rndc nta" when adding NTAs to multiple views
  was incorrectly terminated after the first line, so users only saw
  on NTA added unless they checked the logs.
2018-09-10 12:02:49 -07:00
Mark Andrews d1f8e116db add more missing name in update rule checks 2018-09-10 09:55:24 +10:00
Mark Andrews 54c4db569b add krb5-* tests to nsupdate system test. 2018-09-10 09:32:32 +10:00
Mark Andrews 5fb75a3d75 check that name field is not a valid type 2018-09-10 09:32:31 +10:00
Ondřej Surý 2750799132 Assume socklen_t is always available on all supported platforms (except on Windows where it is typedefed to int32_t) 2018-09-07 12:17:40 +02:00
Ondřej Surý 7716bdd400 Remove dummy PATH_RANDOMDEV and CHECK_DSA defines 2018-09-07 12:17:40 +02:00
Ondřej Surý 4014bc42dd According to POSIX.1-2001, POSIX.1-2008 #include <sys/select.h> is correct way to get fd_set 2018-09-07 12:17:30 +02:00
Evan Hunt 881a868dc0 the allow-query test was in conf.sh.win32 as allow_query, so was skipped 2018-09-05 18:02:40 -07:00
Mark Andrews 05eb9fd338 simplify 2018-08-31 18:47:29 +10:00
Mark AndrewsandWitold Krecicki feaf381adb use pack to construct the reply as Net::DNS just get it wrong 2018-08-31 03:47:56 -04:00
Mark AndrewsandEvan Hunt 615ebc39e3 remove EDNS workarounds, update legacy test 2018-08-30 21:17:00 -07:00
Mark AndrewsandEvan Hunt e7e4603f9b bring up interfaces 9 and 10 2018-08-30 21:16:59 -07:00
Mark Andrews 8e5abc6cdf report initial RRSIG expiry field values 2018-08-30 22:37:08 -04:00
Witold Kręcicki 79ce86090a rndc reconfig should not touch already loaded zones, some refactoring of dns_{zone,view,zt}_{async,}load 2018-08-28 17:04:45 +02:00
Ondřej Surý fecbc7923a Remove isc_keyboard family of functions as they were not used anywhere 2018-08-28 14:37:30 +02:00
Ondřej Surý 0f24c55d38 Refactor *_destroy and *_detach functions to unified order of actions.
This properly orders clearing the freed pointer and calling isc_refcount_destroy
as early as possible to have ability to put proper memory barrier when cleaning
up reference counting.
2018-08-28 13:15:59 +02:00
Ondřej Surý bef8ac5bae Rewrite isc_refcount API to fetch_and_<op>, instead of former <op>_and_<fetch> 2018-08-28 12:15:39 +02:00
Ondřej Surý 0a7535ac81 isc_refcount_init() now doesn't return isc_result_t and asserts on failed initialization 2018-08-28 12:15:39 +02:00
Ondřej Surý efd613e874 memmove, strtoul, and strcasestr functions are part of ISO C90, remove the compatibility shim 2018-08-28 10:31:48 +02:00
Ondřej Surý 00ca487fec We always want IPv6 2018-08-28 10:31:47 +02:00
Ondřej Surý 02e349c051 Remove legacy support for HP-UX 2018-08-28 10:31:47 +02:00
Ondřej Surý 8b3878dc7d Run rsabigexponent test only when OpenSSL is used as cryptography provider 2018-08-27 11:24:32 +02:00
Michał Kępień a23162ab28 Set "use-alt-transfer-source no;" in the "mirror" system test
Force ns3 to use a constant source address (10.53.0.3) when sending
transfer requests for the "initially-unavailable" zone to prevent
failures of transfers not triggered by bin/tests/system/mirror/tests.sh
from causing fallback to using a source address for which transfers of
that zone are refused throughout the entire "mirror" system test since
that might yield false positives.
2018-08-25 21:48:17 +02:00
Michał Kępień e36c869e85 Prevent a race in the "inline" system test
A short time window exists between logging the addition of an NSEC3PARAM
record to a zone and committing it to the current version of the zone
database.  If a query arrives during such a time window, an unsigned
response will be returned.  One of the checks in the "inline" system
test requires NSEC3 records to be present in an answer - that check
would fail in the case described above.  Use rndc instead of log
watching for checking whether zone signing and NSEC3 chain modifications
are complete in order to prevent intermittent "inline" system test
failures.
2018-08-24 09:37:37 +02:00
Mark Andrews 1056376d10 when filter-aaaa and dns64 are both configured a assertion failure could occur 2018-08-24 09:26:54 +10:00
Evan Hunt 0d6f504ba3 specify "dnssec-validation yes" instead of default "auto" 2018-08-23 13:52:57 -04:00