Compare commits

...

4 Commits

Author SHA1 Message Date
cvs2git
36b886ae36 This commit was manufactured by cvs2git to create branch
'v9_6_ESV_R4_patch'.
2011-03-28 08:49:25 +00:00
Mark Andrews
20a5a0874f 9.6-ESV-R4 2011-03-28 08:49:24 +00:00
Mark Andrews
c3ab09c9a1 9.6-ESV-R4 2011-03-28 08:41:26 +00:00
cvs2git
5eb117549e This commit was manufactured by cvs2git to create branch
'v9_6_esv_branch_r4'.
2011-01-31 00:15:37 +00:00
2 changed files with 106 additions and 95 deletions

193
CHANGES
View File

@@ -1,3 +1,6 @@
--- 9.6-ESV-R4 released ---
--- 9.6.3 released ---
3009. [bug] clients-per-query code didn't work as expected with
@@ -50,51 +53,9 @@
wrong lock which could lead to server deadlock.
[RT #22614]
2972. [bug] win32: address windows socket errors. [RT #21906]
2971. [bug] Fixed a bug that caused journal files not to be
compacted on Windows systems as a result of
non-POSIX-compliant rename() semantics. [RT #22434]
2970. [security] Adding a NO DATA negative cache entry failed to clear
any matching RRSIG records. A subsequent lookup of
of NO DATA cache entry could trigger a INSIST when the
unexpected RRSIG was also returned with the NO DATA
cache entry.
CVE-2010-3613, VU#706148. [RT #22288]
2969. [security] Fix acl type processing so that allow-query works
in options and view statements. Also add a new
set of tests to verify proper functioning.
CVE-2010-3615, VU#510208. [RT #22418]
2968. [security] Named could fail to prove a data set was insecure
before marking it as insecure. One set of conditions
that can trigger this occurs naturally when rolling
DNSKEY algorithms.
CVE-2010-3614, VU#837744. [RT #22309]
2967. [bug] 'host -D' now turns on debugging messages earlier.
[RT #22361]
2966. [bug] isc_print_vsnprintf() failed to check if there was
space available in the buffer when adding a left
justified character with a non zero width,
(e.g. "%-1c"). [RT #22270]
2965. [func] Test HMAC functions using test data from RFC 2104 and
RFC 4634. [RT #21702]
2964. [bug] view->queryacl was being overloaded. Seperate the
usage into view->queryacl, view->cacheacl and
view->queryonacl. [RT #22114]
2962. [port] win32: add more dependencies to BINDBuild.dsw.
[RT #22062]
2960. [func] Check that named accepts non-authoritative answers.
[RT #21594]
@@ -114,13 +75,6 @@
exact match" message when returning a wildcard
no data response. [RT #21744]
2952. [port] win32: named-checkzone and named-checkconf failed
to initialise winsock. [RT #21932]
2951. [bug] named failed to generate a correct signed response
in a optout, delegation only zone with no secure
delegations. [RT #22007]
2950. [bug] named failed to perform a SOA up to date check when
falling back to TCP on UDP timeouts when
ixfr-from-differences was set. [RT #21595]
@@ -139,27 +93,6 @@
2941. [bug] sdb and sdlz (dlz's zone database) failed to support
DNAME at the zone apex. [RT #21610]
2939. [func] Check that named successfully skips NSEC3 records
that fail to match the NSEC3PARAM record currently
in use. [RT# 21868]
2937. [bug] Worked around an apparent race condition in over
memory conditions. Without this fix a DNS cache DB or
ADB could incorrectly stay in an over memory state,
effectively refusing further caching, which
subsequently made a BIND 9 caching server unworkable.
This fix prevents this problem from happening by
polling the state of the memory context, rather than
making a copy of the state, which appeared to cause
a race. This is a "workaround" in that it doesn't
solve the possible race per se, but several experiments
proved this change solves the symptom. Also, the
polling overhead hasn't been reported to be an issue.
This bug should only affect a caching server that
specifies a finite max-cache-size. It's also quite
likely that the bug happens only when enabling threads,
but it's not confirmed yet. [RT #21818]
2935. [bug] nsupdate: improve 'file not found' error message.
[RT #21871]
@@ -189,17 +122,11 @@
smaller)
[RT #19737]
2925. [bug] Named failed to accept uncachable negative responses
from insecure zones. [RT# 21555]
2923. [bug] 'dig +trace' could drop core after "connection
timeout". [RT #21514]
2922. [contrib] Update zkt to version 1.0.
2921. [bug] The resolver could attempt to destroy a fetch context
too soon. [RT #19878]
2918. [maint] Add AAAA address for I.ROOT-SERVERS.NET.
2916. [func] Add framework to use IPv6 in tests.
@@ -229,10 +156,6 @@
2901. [port] Use AC_C_FLEXIBLE_ARRAY_MEMBER. [RT #21316]
2900. [bug] The placeholder negative caching element was not
properly constructed triggering a INSIST in
dns_ncache_towire(). [RT #21346]
2899. [port] win32: Support linking against OpenSSL 1.0.0.
2898. [bug] nslookup leaked memory when -domain=value was
@@ -243,9 +166,6 @@
2891. [maint] Update empty-zones list to match
draft-ietf-dnsop-default-local-zones-13. [RT# 21099]
2890. [bug] Handle the introduction of new trusted-keys and
DS, DLV RRsets better. [RT #21097]
2889. [bug] Elements of the grammar where not properly reported.
[RT #21046]
@@ -272,9 +192,6 @@
2877. [bug] The validator failed to skip obviously mismatching
RRSIGs. [RT #21138]
2876. [bug] Named could return SERVFAIL for negative responses
from unsigned zones. [RT #21131]
2875. [bug] dns_time64_fromtext() could accept non digits.
[RT #21033]
@@ -284,9 +201,6 @@
2870. [maint] Add AAAA address for L.ROOT-SERVERS.NET.
2869. [bug] Fix arguments to dns_keytable_findnextkeynode() call.
[RT #20877]
2868. [cleanup] Run "make clean" at the end of configure to ensure
any changes made by configure are integrated.
Use --with-make-clean=no to disable. [RT #20994]
@@ -322,11 +236,108 @@
2853. [bug] add_sigs() could run out of scratch space. [RT #21015]
2852. [bug] Handle broken DNSSEC trust chains better. [RT #15619]
2851. [doc] nslookup.1, removed <informalexample> from the docbook
source as it produced bad nroff. [RT #21007]
--- 9.6-ESV-R3 released ---
2972. [bug] win32: address windows socket errors. [RT #21906]
2971. [bug] Fixed a bug that caused journal files not to be
compacted on Windows systems as a result of
non-POSIX-compliant rename() semantics. [RT #22434]
2970. [security] Adding a NO DATA negative cache entry failed to clear
any matching RRSIG records. A subsequent lookup of
of NO DATA cache entry could trigger a INSIST when the
unexpected RRSIG was also returned with the NO DATA
cache entry.
CVE-2010-3613, VU#706148. [RT #22288]
2969. [security] Fix acl type processing so that allow-query works
in options and view statements. Also add a new
set of tests to verify proper functioning.
CVE-2010-3615, VU#510208. [RT #22418]
2968. [security] Named could fail to prove a data set was insecure
before marking it as insecure. One set of conditions
that can trigger this occurs naturally when rolling
DNSKEY algorithms.
CVE-2010-3614, VU#837744. [RT #22309]
2967. [bug] 'host -D' now turns on debugging messages earlier.
[RT #22361]
2966. [bug] isc_print_vsnprintf() failed to check if there was
space available in the buffer when adding a left
justified character with a non zero width,
(e.g. "%-1c"). [RT #22270]
2964. [bug] view->queryacl was being overloaded. Seperate the
usage into view->queryacl, view->cacheacl and
view->queryonacl. [RT #22114]
2962. [port] win32: add more dependencies to BINDBuild.dsw.
[RT #22062]
2952. [port] win32: named-checkzone and named-checkconf failed
to initialise winsock. [RT #21932]
2951. [bug] named failed to generate a correct signed response
in a optout, delegation only zone with no secure
delegations. [RT #22007]
--- 9.6-ESV-R2 released ---
2939. [func] Check that named successfully skips NSEC3 records
that fail to match the NSEC3PARAM record currently
in use. [RT# 21868]
2937. [bug] Worked around an apparent race condition in over
memory conditions. Without this fix a DNS cache DB or
ADB could incorrectly stay in an over memory state,
effectively refusing further caching, which
subsequently made a BIND 9 caching server unworkable.
This fix prevents this problem from happening by
polling the state of the memory context, rather than
making a copy of the state, which appeared to cause
a race. This is a "workaround" in that it doesn't
solve the possible race per se, but several experiments
proved this change solves the symptom. Also, the
polling overhead hasn't been reported to be an issue.
This bug should only affect a caching server that
specifies a finite max-cache-size. It's also quite
likely that the bug happens only when enabling threads,
but it's not confirmed yet. [RT #21818]
2925. [bug] Named failed to accept uncachable negative responses
from insecure zones. [RT# 21555]
2921. [bug] The resolver could attempt to destroy a fetch context
too soon. [RT #19878]
2900. [bug] The placeholder negative caching element was not
properly constructed triggering a INSIST in
dns_ncache_towire(). [RT #21346]
2890. [bug] Handle the introduction of new trusted-keys and
DS, DLV RRsets better. [RT #21097]
2869. [bug] Fix arguments to dns_keytable_findnextkeynode() call.
[RT #20877]
--- 9.6-ESV-R1 released ---
2876. [bug] Named could return SERVFAIL for negative responses
from unsigned zones. [RT #21131]
--- 9.6-ESV released ---
2852. [bug] Handle broken DNSSEC trust chains better. [RT #15619]
--- 9.6.2 released ---
2850. [bug] If isc_heap_insert() failed due to memory shortage

View File

@@ -1,10 +1,10 @@
# $Id: version,v 1.43.12.11 2011/01/30 06:38:13 marka Exp $
# $Id: version,v 1.43.12.11.2.2 2011/03/28 08:49:24 marka Exp $
#
# This file must follow /bin/sh rules. It is imported directly via
# configure.
#
MAJORVER=9
MINORVER=6
PATCHVER=3
RELEASETYPE=
RELEASEVER=
PATCHVER=
RELEASETYPE=-ESV
RELEASEVER=-R4