/contrib/bind9/RELEASE-NOTES-BIND-9.8.1.txt

https://bitbucket.org/freebsd/freebsd-head/ · Plain Text · 268 lines · 245 code · 23 blank · 0 comment · 0 complexity · ca9b29a8caf429cd87d2bff834059278 MD5 · raw file

  1. __________________________________________________________________
  2. Introduction
  3. BIND 9.8.1 is the current production release of BIND 9.8.
  4. This document summarizes changes from BIND 9.8.0 to BIND 9.8.1. Please
  5. see the CHANGES file in the source code release for a complete list of
  6. all changes.
  7. Download
  8. The latest versions of BIND 9 software can always be found on our web
  9. site at http://www.isc.org/downloads/all. There you will find
  10. additional information about each release, source code, and some
  11. pre-compiled versions for certain operating systems.
  12. Support
  13. Product support information is available on
  14. http://www.isc.org/services/support for paid support options. Free
  15. support is provided by our user community via a mailing list.
  16. Information on all public email lists is available at
  17. https://lists.isc.org/mailman/listinfo.
  18. New Features
  19. 9.8.1
  20. * Added a new include file with function typedefs for the DLZ
  21. "dlopen" driver. [RT #23629]
  22. * Added a tool able to generate malformed packets to allow testing of
  23. how named handles them. [RT #24096]
  24. * The root key is now provided in the file bind.keys allowing DNSSEC
  25. validation to be switched on at start up by adding
  26. "dnssec-validation auto;" to named.conf. If the root key provided
  27. has expired, named will log the expiration and validation will not
  28. work. More information and the most current copy of bind.keys can
  29. be found at http://www.isc.org/bind-keys. *Please note this feature
  30. was actually added in 9.8.0 but was not included in the 9.8.0
  31. release notes. [RT #21727]
  32. Security Fixes
  33. 9.8.1
  34. * If named is configured with a response policy zone (RPZ) and a
  35. query of type RRSIG is received for a name configured for RRset
  36. replacement in that RPZ, it will trigger an INSIST and crash the
  37. server. RRSIG. [RT #24280]
  38. * named, set up to be a caching resolver, is vulnerable to a user
  39. querying a domain with very large resource record sets (RRSets)
  40. when trying to negatively cache the response. Due to an off-by-one
  41. error, caching the response could cause named to crash. [RT #24650]
  42. [CVE-2011-1910]
  43. * Using Response Policy Zone (RPZ) to query a wildcard CNAME label
  44. with QUERY type SIG/RRSIG, it can cause named to crash. Fix is
  45. query type independant. [RT #24715]
  46. * Using Response Policy Zone (RPZ) with DNAME records and querying
  47. the subdomain of that label can cause named to crash. Now logs that
  48. DNAME is not supported. [RT #24766]
  49. * Change #2912 populated the message section in replies to UPDATE
  50. requests, which some Windows clients wanted. This exposed a latent
  51. bug that allowed the response message to crash named. With this
  52. fix, change 2912 has been reduced to copy only the zone section to
  53. the reply. A more complete fix for the latent bug will be released
  54. later. [RT #24777]
  55. Feature Changes
  56. 9.8.1
  57. * Merged in the NetBSD ATF test framework (currently version 0.12)
  58. for development of future unit tests. Use configure --with-atf to
  59. build ATF internally or configure --with-atf=prefix to use an
  60. external copy. [RT #23209]
  61. * Added more verbose error reporting from DLZ LDAP. [RT #23402]
  62. * The DLZ "dlopen" driver is now built by default, no longer
  63. requiring a configure option. To disable it, use "configure
  64. --without-dlopen". (Note: driver not supported on win32.) [RT
  65. #23467]
  66. * Replaced compile time constant with STDTIME_ON_32BITS. [RT #23587]
  67. * Make --with-gssapi default for ./configure. [RT #23738]
  68. * Improved the startup time for an authoritative server with a large
  69. number of zones by making the zone task table of variable size
  70. rather than fixed size. This means that authoritative servers with
  71. lots of zones will be serving that zone data much sooner. [RT
  72. #24406]
  73. * Per RFC 6303, RFC 1918 reverse zones are now part of the built-in
  74. list of empty zones. [RT #24990]
  75. Bug Fixes
  76. 9.8.1
  77. * During RFC5011 processing some journal write errors were not
  78. detected. This could lead to managed-keys changes being committed
  79. but not recorded in the journal files, causing potential
  80. inconsistencies during later processing. [RT #20256]
  81. * A potential NULL pointer deference in the DNS64 code could cause
  82. named to terminate unexpectedly. [RT #20256]
  83. * A state variable relating to DNSSEC could fail to be set during
  84. some infrequently-executed code paths, allowing it to be used
  85. whilst in an unitialized state during cache updates, with
  86. unpredictable results. [RT #20256]
  87. * A potential NULL pointer deference in DNSSEC signing code could
  88. cause named to terminate unexpectedly [RT #20256]
  89. * Several cosmetic code changes were made to silence warnings
  90. generated by a static code analysis tool. [RT #20256]
  91. * When using the -x (sign with only KSK) option on dnssec-signzone,
  92. it could incorrectly count the number of ZSKs in the zone. (And in
  93. 9.9.0, some code cleanup and improved warning messages). [RT
  94. #20852]
  95. * When using _builtin in named.conf, named.conf changes were not
  96. found when reloading the config file. Now checks _builtin zone
  97. arguments to see if the zone is re-usable or not. [RT #21914]
  98. * Running dnssec-settime -f on an old-style key will now force the
  99. key to be rewritten to the new key format even if no other change
  100. has been specified, using "-P now -A now" as default values. [RT
  101. #22474]
  102. * After an external code review, a code cleanup was done. [RT #22521]
  103. * Cause named to terminate at startup or rndc reconfig reload to
  104. fail, if a log file specified in the conf file isn't a plain file.
  105. (RT #22771]
  106. * named now forces the ADB cache time for glue related data to zero
  107. instead of relying on TTL. This corrects problematic behavior in
  108. cases where a server was authoritative for the A record of a
  109. nameserver for a delegated zone and was queried to recursively
  110. resolve records within that zone. [RT #22842]
  111. * When a validating resolver got a NODATA response for DNSKEY, it was
  112. not caching the NODATA. Fixed and test added. [RT #22908]
  113. * Fixed a bug in which zone keys that were published and but not
  114. immediately activated, automatic signing could fail to trigger. [RT
  115. #22911]
  116. * Fixed precedence order bug with NS and DNAME records if both are
  117. present. (Also fixed timing of autosign test in 9.7+) [RT #23035]
  118. * When a DNSSEC signed dynamic zone's signatures need to be
  119. refreshed, named would first delete the old signatures in the zone.
  120. If a private key of the same algorithm isn't available to named,
  121. the signing would fail but the old signatures would already be
  122. deleted. named now checks if it can access the private key before
  123. deleting the old signatures and leaves the old signature if no
  124. private key is found. [RT #23136]
  125. * When using "auto-dnssec maintain" and rolling to a new key, a
  126. private-type record (only used internally by named) could be
  127. created and not marked as complete. [RT #23253]
  128. * Fixed last autosign test report. [RT #23256]
  129. * named didn't save gid at startup and later assumed gid 0. named now
  130. saves/restores the gid when creating creating named.pid at startup.
  131. [RT #23290]
  132. * If the server has an IPv6 address but does not have IPv6
  133. connectivity to the internet, dig +trace could fail attempting to
  134. use IPv6 addresses. [RT #23297]
  135. * If named is configured with managed zones, the managed key maint
  136. timer can exercise a race condition that can crash the server. [RT
  137. #23303]
  138. * Changing TTL did not cause dnssec-signzone to generate new
  139. signatures. [RT #23330]
  140. * Have the validating resolver use RRSIG original TTL to compute
  141. validated RRset and RRSIG TTL. [RT #23332]
  142. * In "make test" bin/tests/resolver, hold the socket manager lock
  143. while freeing the socket. [RT #23333]
  144. * If named encountered a CNAME instead of a DS record when walking
  145. the chain of trust down from the trust anchor, it incorrectly
  146. stopped validating. [RT #23338]
  147. * dns/view.h needed dns/rpz.h but it wasn't in the Makfile.in HEADERS
  148. variable. [RT #23342]
  149. * RRSIG records could have time stamps too far in the future. [RT
  150. #23356]
  151. * named stores cached data in an in-memory database and keeps track
  152. of how recently the data is used with a heap. The heap is stored
  153. within the cache's memory space. Under a sustained high query load
  154. and with a small cache size, this could lead to the heap exhausting
  155. the cache space. This would result in cache misses and SERVFAILs,
  156. with named never releasing the cache memory the heap used up and
  157. never recovering. This fix removes the heap into its own memory
  158. space, preventing the heap from exhausting the cache space and
  159. allowing named to recover gracefully when the high query load
  160. abates. [RT #23371]
  161. * Fully separated key management on a per view basis. [RT #23419]
  162. * If running on a powerpc CPU and with atomic operations enabled,
  163. named could lock up. Added sync instructions to the end of atomic
  164. operations. [RT #23469]
  165. * If OpenSSL was built without engine support, named would have
  166. compile errors and fail to build. [RT #23473]
  167. * If ./configure finds GOST but not elliptic curve, named fails to
  168. build. Added elliptic curve support check in GOST OpenSSL engine
  169. detection. [RT #23485]
  170. * "rndc secroots" would abort on the first error and so could miss
  171. remaining views. [RT #23488]
  172. * Handle isc_event_allocate failures in t_tasks test. [RT #23572]
  173. * ixfr-from-differences {master|slave}; failed to select the
  174. master/slave zones, resulting in on diff/journal file being
  175. created. [RT #23580]
  176. * If a DNAME substitution failed, named returned NOERROR. The correct
  177. response should be YXDOMAIN. [RT #23591]
  178. * dns_dnssec_findzonekeys{2} used a inconsistant timestamp when
  179. determining which keys are active. This could result in some RRsets
  180. not being signed/re-signed. [RT #23642]
  181. * Remove bin/tests/system/logfileconfig/ns1/named.conf and add
  182. setup.sh in order to resolve changing named.conf issue. [RT #23687]
  183. * NOTIFY messages were not being sent when generating a NSEC3 chain
  184. incrementally. [RT #23702]
  185. * DDNS updates using SIG(0) with update-policy match type "external"
  186. could cause a crash. Also fixed nsupdate core dump on shutdown when
  187. using a SIG(0) key, due to the key not being freed. [RT #23735]
  188. * Zones using automatic key maintenance could fail to check the key
  189. repository for updates. named now checks once per hour and the
  190. automatic check bug has been fixed. [RT #23744]
  191. * named now uses the correct strtok/strtok_r/strtok_s based on OS.
  192. [RT #23747]
  193. * Signatures for records at the zone apex could go stale due to an
  194. incorrect timer setting. [RT #23769]
  195. * The autosign tests attempted to open ports within reserved ranges.
  196. Test now avoids those ports. [RT #23957]
  197. * GSS TGIS test was failing, since log_cred() caused KRB5_KTNAME to
  198. be cached. Now sets KRB5_KTNAME before calling log_cred() in
  199. dst_gssapi_acceptctx(). [RT #24004]
  200. * named, acting as authoritative server for DLZ zones, was not
  201. correctly setting the authoritative (AA) bit. [RT #24146]
  202. * Clean up some cross-compiling issues and added two undocumented
  203. configure options, --with-gost and --with-rlimtype, to allow
  204. over-riding default settings (gost=no and rlimtype="long int") when
  205. cross-compiling. [RT #24367]
  206. * When trying sign with NSEC3, if dnssec-signzone couldn't find the
  207. KSK, it would give an incorrect error "NSEC3 iterations too big for
  208. weakest DNSKEY strength" rather than the correct "failed to find
  209. keys at the zone apex: not found" [RT #24369]
  210. * Configuring 'dnssec-validation auto' in a view instead of in the
  211. options statement could trigger an assertion failure in
  212. named-checkconf. [RT #24382]
  213. * Improved consistency checks for dnssec-enable and
  214. dnssec-validation, added test cases to the checkconf system test.
  215. [RT #24398]
  216. * If named is configured to be both authoritative and recursive and
  217. receives a recursive query for a CNAME in a zone that it is
  218. authoritative for, if that CNAME also points to a zone the server
  219. is authoritative for, the recursive part of name will not follow
  220. the CNAME change and the response will not be a complete CNAME
  221. chain. [RT #24455]
  222. * nsupdate could dump core on shutdown when using SIG(0) keys. [RT
  223. #24604]
  224. * Named could fail to validate zones list in a DLV that validated
  225. insecure without using DLV and had DS records in the parent zone.
  226. [RT #24631]
  227. * dnssec-signzone now records timestamps just before and just after
  228. signing, improving the accuracy of signing statistics. [RT #16030]
  229. * If allow-new-zones was set to yes and name-based ACLs were used,
  230. named could crash when "rndc reconfig" was issued. [RT #22739]
  231. * RT #23136 fixed a problem where named would delete old signatures
  232. even when the private key wasn't available to re-sign the zone,
  233. resulting in a zone with missing signatures. This fix (CHANGES
  234. 3114) did not completely fix all issues. [RT #24577]
  235. * A bug in FreeBSD kernels causes IPv6 UDP responses greater than
  236. 1280 bytes to not fragment as they should. Until there is a kernel
  237. fix, named will work around this by setting IPV6_USE_MIN_MTU on a
  238. per packet basis. [RT #24950]
  239. Known issues in this release
  240. * None.
  241. Thank You
  242. Thank you to everyone who assisted us in making this release possible.
  243. If you would like to contribute to ISC to assist us in continuing to
  244. make quality open source software, please visit our donations page at
  245. http://www.isc.org/supportisc.