Commit graph

5319 commits

Author SHA1 Message Date
Andreas Gustafsson
08a44b06ef the length of the NULL record RDATA can be zero;
do not assert the contrary
2001-06-20 17:43:31 +00:00
Andreas Gustafsson
ac299c4f5c Removed statements to the effect that certain functions
perform assertion checks.  Such statements are inappropriate as they
document the implementation rather than the public interface.  The
functions are not required to perform assertion checks, but the caller
is required to pass arguments that conform to the API requirements.
2001-06-18 22:56:35 +00:00
Andreas Gustafsson
d748172fd7 added one end tag too many 2001-06-18 22:11:59 +00:00
Andreas Gustafsson
86b0285d7e missing end tags; other cleanup 2001-06-18 22:07:02 +00:00
Andreas Gustafsson
474a81662f missing end tags; other cleanup 2001-06-18 17:56:54 +00:00
Andreas Gustafsson
19872fdfb7 convert more socket errors to specific ISC error codes, notably ECONNRESET 2001-06-15 22:07:51 +00:00
Brian Wellington
2196128fc2 When doing best-effort parsing (in dig, for example) on a message with multiple
questions where an unrecoverable error is seen after 2 or more questions have
been parsed, memory would be leaked from msg->namepool, and would trigger an
ENSURE() in dns_message_destroy().
2001-06-15 05:24:24 +00:00
Brian Wellington
0a9a3d8c6d 904. [bug] The server would leak memory if attempting to use
an expired TSIG key. [RT #1406]
2001-06-15 02:24:02 +00:00
Andreas Gustafsson
d1ecee17a5 remove the other blank line before .TH, too 2001-06-14 19:10:14 +00:00
Andreas Gustafsson
23ac0d8259 comments for dns_message_parse() referred to
nonexistent arguments
2001-06-14 18:38:48 +00:00
Mark Andrews
9a6573e16b sys/select.h not sys/socket.h 2001-06-12 01:35:42 +00:00
Andreas Gustafsson
509ef0c871 Keep track of the number of times the system malloc() has
failed, to help determine whether a server crash resulted
from a bug in the handling of an out-of-memory condition
or not.
2001-06-11 20:27:16 +00:00
Andreas Gustafsson
8de7014e56 removed CVS Id from machine-generated man page files -- it is no longer needed
now that they are ignored by update_copyrights
2001-06-11 19:02:45 +00:00
Andreas Gustafsson
c44072c299 uninitialized pointer could potentially (but not likely)
be passed to free_adbfind(); if it happened, it would look like RT #1393
(caught by bwelling)
2001-06-11 04:26:41 +00:00
David Lawrence
eda73d8aad redo 1.64 "potential null pointer dereference" more consistent with the style
of the function and file (i.e., not returning from the middle of a function)
2001-06-10 17:06:45 +00:00
David Lawrence
95e8c0f8f5 removed blank line before .TH 2001-06-10 13:58:29 +00:00
David Lawrence
2b403f22d4 Removed an obsolete XXXMLG comment, with the assent of Michael Graff. 2001-06-08 23:53:41 +00:00
David Lawrence
04260c5c48 895. [func] New function, isc_dir_current(), akin to POSIX's
getcwd().
2001-06-08 23:50:32 +00:00
Brian Wellington
ba9e9a88c8 The last change didn't compile due to faulty cut&paste. 2001-06-08 22:48:28 +00:00
David Lawrence
504f7802d4 893. [func] Removed isc_file_test() and added isc_file_exists()
for the basic functionality that was being added
			with isc_file_test().
2001-06-08 21:53:49 +00:00
Brian Wellington
3b6bcedffe 891. [bug] Return an error when a SIG(0) signed response to
an unsigned query is seen.  This should actually
			do the verification, but it's not currently
			possible. [RT #1391]
2001-06-08 19:37:29 +00:00
Andreas Gustafsson
841179549b 889. [port] Eliminated blank lines before .TH in nroff man
pages since they cause problems with some versions
                        of nroff. [RT #1390]
2001-06-08 19:33:02 +00:00
Brian Wellington
01535cfb3a removed erroneous UNUSED macro 2001-06-08 19:16:15 +00:00
Brian Wellington
9af6e1c548 Don't die when trying to delete a nonexistent key [RT #1392] 2001-06-08 18:29:31 +00:00
Mark Andrews
4585aeb2cc 866. [func] Close debug only file channels when debug is set to
zero. [RT #1246]
2001-06-08 02:57:13 +00:00
Andreas Gustafsson
56877490bc 865. [bug] The new configuration parser did not allow
the optional debug level in a "severity debug"
                        clause of a logging channel to be omitted.
                        This is now allowed and treated as "severity
                        debug 1;" like it did in BIND 8.2.4, not as
                        "severity debug 0;" like it did in BIND 9.1.
                        [RT #1367]
2001-06-08 01:03:00 +00:00
Andreas Gustafsson
11c8a04f9b grammar 2001-06-07 20:11:30 +00:00
Andreas Gustafsson
d358eea93e megabytes is abbreviated MB, not mb 2001-06-07 18:31:54 +00:00
Mark Andrews
c69d61c69b realloc -> malloc 2001-06-07 02:14:52 +00:00
Andreas Gustafsson
8b9fa5b005 redid RT #1353 fix in 1.53 so that it does not free
file names while they are still referenced from the parse tree
2001-06-07 01:58:49 +00:00
Brian Wellington
90c845947a #include <time.h> 2001-06-07 00:45:35 +00:00
Mark Andrews
6cefb60af5 pullup from BIND 8
862.   [bug]           Use after realloc(), non portable pointer arithmetic in
                        grmerge().
2001-06-07 00:22:35 +00:00
Brian Wellington
224e84bc0e misc. stuff related to [RT #1293] 2001-06-07 00:21:53 +00:00
Brian Wellington
571688b02f Revert part of the ISC_MAGIC/ISC_MAGIC_VALID change. Because
common.impmagic is not the first four bytes of the struct,
ISC_MAGIC_VALID() cannot be used.
2001-06-06 22:03:58 +00:00
Brian Wellington
338d559b3e remove extraneous blank line 2001-06-06 21:02:43 +00:00
Mark Andrews
cd8c92da37 pullup from BIND8:
DE_CONST bind(arg 2)
2001-06-06 01:56:32 +00:00
Mark Andrews
99a8d30559 860. [func] Drop cross class glue in zone transfers. 2001-06-05 23:43:15 +00:00
David Lawrence
f991461dc1 The most notable change is that incremental_cleaning_action() will now
wait at least six seconds between each pass of the tree.  This keeps
the CPU from being bogged down in a persistent overlimit condition, and
because it uses timers also lets the cache cleaner know right away when
the server is being shut down, so it can stop trying to clean.

Also fixed an unlocked access to cache->live_tasks in dns_cache_detach().

Also fixed a problem with repeated calls to dns_cache_setcachesize() setting
the database as not overmem, even though it might have still been in
the process of cleaning from a previous limit.  Just set the new
isc_mem_setwater() limits instead and let the mem.c call water() if cleaning
needs to be either started or stopped because of the new limits.

Minor code cleanups.
2001-06-05 22:27:51 +00:00
David Lawrence
65640f401a Removed the requirement for isc_mem_setwater() that when the callback function
is not null then hi_water must exceed lo_water and both must be > 0.
Now requires hi_water >= lo_water, and they can both be 0.  Not allowing
them to be equal was just an unnecessary restriction, and letting them be 0
is useful for the case where the context has had non-zero values set, went
over hi_water and called the callback, and then wanted to remove limits
(perhaps because of a reconfiguration).  This allows the callback to
be signaled as ISC_MEM_LOWATER on the next isc_mem_put(), which the caller
might need to terminate any outstanding action that was triggered by
the hi_water.
2001-06-05 22:14:20 +00:00
Mark Andrews
92296c744e complete the rollback 2001-06-05 09:23:14 +00:00
Mark Andrews
8af4b7042d rollback individual RR. 2001-06-05 09:17:21 +00:00
Mark Andrews
03f91269f5 856. [func] Allow partial rdatasets to be returned in answer and
authority sections to help non-TCP capable clients
                        recover from truncation.  [RT #1301]
2001-06-05 09:02:16 +00:00
Brian Wellington
7e93abafe6 spelling & style 2001-06-05 06:34:45 +00:00
Brian Wellington
33e49e8fdc #include <isc/magic.h> 2001-06-05 06:34:11 +00:00
Mark Andrews
afb0a628ef 855. [bug] Stop spurious "using RFC 1035 TTL semantics" warnings. 2001-06-05 05:12:47 +00:00
Brian Wellington
17dba29ba5 854. [bug] The config parser didn't properly handle config
options that were specified in units of time other
			than seconds.  [RT #1372]
2001-06-04 21:51:27 +00:00
Brian Wellington
fe61fbe14e ignore Makefiles and autoconf stuff 2001-06-04 21:48:52 +00:00
David Lawrence
163e27991b Revert part of the ISC_MAGIC/ISC_MAGIC_VALID change. Because
common.impmagic is not the first four bytes of the struct,
ISC_MAGIC_VALID() cannot be used.
2001-06-04 20:57:25 +00:00
David Lawrence
92ef1a9b9d use ISC_MAGIC for all magic numbers, for our friends in EBCDIC land 2001-06-04 19:33:39 +00:00
Brian Wellington
deb588e256 typo that caused an assertion failure [RT #1368] 2001-06-03 23:53:50 +00:00