Commit Graph

3583 Commits

Author SHA1 Message Date
David Lawrence
96ab25294b 307. [bug] When canceling a query, the resolver didn't check for
isc_socket_sendto() calls that did not yet have their
			completion events posted, so it could (rarely) end up
			destroying the query context and then want to use
			it again when the send event posted, triggering an
			assertion as it tried to cancel an already-canceled
			query.  (RT #77)

... so now the resolver counts the number of outstanding isc_socket_sendto()s
and only destroys the query context when there are no more.
2000-07-04 01:19:20 +00:00
Brian Wellington
e68de4a7db 306. [bug] Reading HMAC-MD5 private key files didn't work. 2000-07-03 23:27:45 +00:00
Andreas Gustafsson
b3e4ce3059 alternative fix to 303. [RT #157] 2000-07-03 22:42:36 +00:00
Andreas Gustafsson
d8560fb7ff [RT #155]
305. [bug]             When reloading the server with a config file
                        containing a syntax error, it could catch an
                        assertion failure trying to perform zone
                        maintenance on tentatively created zones whose
                        views were never fully configured and lacked
                        an address database.
2000-07-03 20:21:53 +00:00
Brian Wellington
e2894b0e50 304. [bug] If more than LWRES_CONFMAXNAMESERVERS servers
are listed in resolv.conf, silently ignore them
                        instead of returning failure.
2000-07-03 20:13:12 +00:00
Andreas Gustafsson
855bda3de3 new joint ISC/BSDI copyright message as requested by
Paul Borman <prb@bsdi.com>
2000-07-03 17:44:22 +00:00
Mark Andrews
845e0b5f96 303. [bug] add additional sanity checks to differentiate a AXFR
response vs a IXFR response. (RT #157)
2000-07-03 13:33:05 +00:00
David Lawrence
de8e189332 300. [bug] Using both <isc/net.h> and <lwres/net.h> didn't work
on platforms lacking IPv6 because each included their
			own ipv6 header file for the missing definitions.  Now
			each library's ipv6.h defines the wrapper symbol of
			the other (ISC_IPV6_H and LWRES_IPV6_H).
2000-07-01 01:41:54 +00:00
David Lawrence
641689bff4 note in the comment that isc_time_microdiff returns 0 if t1 < t2.
no need to pull up.
2000-06-30 22:40:32 +00:00
Brian Wellington
e405739af2 297. [bug] Specifying a key name that wasn't fully qualified
in certain parts of the config file could cause
                        an assertion.
2000-06-30 15:39:23 +00:00
Andreas Gustafsson
fe8d130967 make install didn't work when using separate build dir 2000-06-30 00:29:29 +00:00
Andreas Gustafsson
428b78e56d RT #153: when invoked with type==CNAME and a message
not constructed by dns_message_parse(), dns_message_findname() failed to
find anything due to checking for attribute bits set only by dns_message_parse().
This caused an infinite loop when constructing the response to an ANY query
at a CNAME in a secure zone.
2000-06-29 19:49:43 +00:00
Andreas Gustafsson
824068b85b cut-and-paste-o: dns_message_findname() tested for
dns_rdatatype_cname twice instead of cname+dname
2000-06-29 18:58:32 +00:00
Andreas Gustafsson
26be84cdfc spelling, grammar, and formatting; no functional change 2000-06-29 16:48:53 +00:00
Mark Andrews
b74e73c5b4 294. [bug] If we run out of space in while processing glue
when reading a master file and commit "current name"
			reverts to "name_current" instead of staying as
			"name_glue".
2000-06-29 15:55:19 +00:00
Michael Graff
2e0e781ba4 use the correct function name in debugging statements 2000-06-29 07:40:58 +00:00
Brian Wellington
089b7bf0a5 a missing #include "assert_p.h" caused a compile failure. 2000-06-28 22:57:42 +00:00
Brian Wellington
09aea877e7 An lwres context was being destroyed while still in use 2000-06-28 21:55:33 +00:00
Brian Wellington
f51647808f Using 0 as an error value tends to conflict with using 0 to represent
success.
2000-06-28 21:43:46 +00:00
Jim Reid
084d809e48 fixed some typos
removed references to ENI_ error codes because they're not
documented in RFC2133 and are defined in getnameinfo.c rather
than some include file anyway
2000-06-28 21:43:41 +00:00
David Lawrence
5eae30a019 in fact, ISC_LANG_BEGINDECLS and ISC_LANG_ENDDECLS are not needed at all since
rdatastructpre.h and rdatastructsuf.h already provide the wrapper.
(no need to pull up.)
2000-06-28 19:07:19 +00:00
David Lawrence
778d24f5ba do not include isc/lang.h; rdatastructpre.h already gets it.
no need to pull these changes up to the 9.0.0 branch.
2000-06-28 19:03:24 +00:00
Michael Graff
0cd1ba32c1 put edns0 restarts back to 3, and fix up a few trace messages to be standard ones -- lower case first letter. 2000-06-28 16:30:29 +00:00
David Lawrence
4a44b85e6e Disable the requirement that all messages be signed, since this
will not work until the "controls" statement is implemented in
named.conf post-9.0.0.
2000-06-28 05:24:41 +00:00
David Lawrence
de88422aec only define ISC_MEM_DEBUG if it is not already defined, or if ISC_MEM_DEBUGOFF
is not defined.  no need to pull this one up to v9_0.
2000-06-28 03:46:37 +00:00
David Lawrence
13c32cb589 285. [bug] A change made to the dst API for beta4 inadvertently
broke OMAPI's creation of a dst key from an incoming
			message, causing an assertion to be triggered.  Fixed.
2000-06-28 03:09:44 +00:00
Brian Wellington
171d0db7f9 Return NO_RECOVERY, not HOST_NOT_FOUND if the error is not LWRES_NOTFOUND 2000-06-27 23:20:34 +00:00
Brian Wellington
2ef7b73fd5 Missing #include "assert_p.h" 2000-06-27 23:13:26 +00:00
Jim Reid
a07cfacd85 initial draft of lwres manpages 2000-06-27 21:53:10 +00:00
Andreas Gustafsson
c4bf0b3906 don't issue warnings saying max-ncache-ttl and max-cache-ttl are
not yet implemented, because they are
2000-06-27 21:33:20 +00:00
Andreas Gustafsson
33e482fa3e edited comment, no functional change 2000-06-27 21:04:12 +00:00
Brian Wellington
d364e89bda The off by one error wasn't. 2000-06-27 18:51:51 +00:00
Michael Sawyer
d8afbf2f30 Force SIGHUP to be in state SIG_DFL when starting, since Solaris would break
that, if run from under cron.
2000-06-27 18:49:14 +00:00
Brian Wellington
9006ddc64a Fix off-by-one error and move a few lines of code. 2000-06-27 18:03:52 +00:00
Brian Wellington
031ce3bc62 Return better errors (TRAILINGDATA, not UNEXPECTEDEND), and style changes. 2000-06-27 18:03:23 +00:00
Brian Wellington
f356f1bb5d style changes 2000-06-27 18:02:12 +00:00
Brian Wellington
db118c5f26 Add a comment saying that _parseheader() isn't the right place to verify
that the buffer containing the entire packet is long enough.
2000-06-27 18:01:16 +00:00
Brian Wellington
c614882bd8 typo 2000-06-27 00:41:13 +00:00
Brian Wellington
6f4317150b remove dead code 2000-06-27 00:24:27 +00:00
Brian Wellington
50719689ef coding style 2000-06-27 00:24:12 +00:00
Brian Wellington
c1ecf4ace2 Remove declaration of REQUIRE 2000-06-26 23:35:00 +00:00
Brian Wellington
347ccc2716 Fixed a few errors in error recovery and uses of int instead of lwres_result_t 2000-06-26 23:23:28 +00:00
Andreas Gustafsson
a7d4703552 commented 2000-06-26 22:59:22 +00:00
Andreas Gustafsson
aa57fcf820 two more unchecked calls to strdup() found by Brian 2000-06-26 22:50:20 +00:00
Andreas Gustafsson
f264d3cb2c sense of strcmp() wrong; missing test of strdup() return value
(find by Brian by inspection)
2000-06-26 22:30:32 +00:00
Andreas Gustafsson
95e7cd1ecb commented 2000-06-26 22:09:49 +00:00
Michael Graff
7618ec31e3 bump restarts to 4. After 3, we change the retry times. This lets the query take longer, but it is more likely to succeed with EDNS0 if the server or the network is just slow, not broken. 2000-06-26 21:46:25 +00:00
David Lawrence
2298aec5a1 test for high bit being set in isc_file_settime needed to use an
unsigned long long constant to work correctly on platforms that have
a 64 bit timeval.tv_sec and 32 bit ints.
2000-06-26 21:33:57 +00:00
Brian Wellington
bc9470d240 *** empty log message *** 2000-06-26 21:27:14 +00:00
Michael Graff
57ebb859ee After N restarts (currently 3) try again with EDNS0 disabled. 2000-06-26 21:07:36 +00:00