$Id: TODO,v 4.9.1.2 1993/05/17 09:59:01 vixie Exp $ Things to do. Each entry should contain the proposer, date proposed, and an explaination of what's being proposed. New ones are added at the bottom. Note that the author/coordinator of BIND does not neccessarily endorse all of the proposals listed herein; if you did not get explicit "buy-in" then your changes may not be accepted even if they appear in proposal form here in this file. [vixie@pa.dec.com 06Jan93]: syntax checking in db_load.c currently there is little or no syntax checking in db_load.c. you can see this in action if you add an HINFO RR with only one word of info -- the second field in the database is trash, and is sent out in zone transfers or responses as trash. such records should be ignored as errors much as "cname and other data" is. (kre@munnari.oz.au comments: This should turn into "rewrite db_load.c" - probably using yacc or something (though that doesn't matter) so that it approximates the syntax its supposed to implement.) [mark@comp.vuw.ac.nz 26apr93]: support classes other than IN. Allow specification of zones in classes other than "IN" in the named.boot file---allow an optional "/class" on the "primary" and "secondary" directives. Also fix zone transfers so only data in the class requested is transfered. [vixie@pa.dec.com 25apr93]: clean up debugging replace all "#ifdef DEBUG...fprintf(...)...#endif" with dprintf(...) which would be a macro that only expands to an fprintf() if DEBUG is set. dprintf(x, (args)) with x as the log level. perhaps change log levels to be symbolic, and perhaps make them a mask instead of a limit. [vixie@pa.dec.com 25apr93]: clean up #ifdef's and portability feature #ifdef's should be limited to whole functions, which will be called no matter what and would only be non-empty if the feature is enabled. allow feature ifdef's in .h files, though. portability #ifdef's should be limited to whole functions, too. add a new portability.c module that implements anything which varies from system to system. add a second portability.h-like file that is included _before_ all the system includes. portability.h as it stands is included _after_ all system includes, which is convenient for most things but not all. add and use function prototypes. make everything static that can be. externs should only be in .h files (add more .h files, per module if needed, to cover these). add "export" keyword (null define) to make it clear which names are exported and which are static. all top-blevel names in a module must be "export" or "static". [sater@cs.vu.nl 26apr93]: sortlist improvement Improve the code around the sortlist area to better cope with parallel networks of different speeds. The -i hack I sent to you could function as inspiration only. [kre@munnari.oz.au 26apr93]: add an INN style control interface to replace sending signals. With that expand debugging to permit monitoring of actions taken on a single query (query through control port, full traced as it occurs) or all queries that reference some particular name or zone, or which are forwarded, or asked, of some particluar server. Allow reloads & dumps of a single zone, rather than the whole universe. Allow selective cache pruning (to edit away bad data that's been obtained from somewhere) [kre@munnari.oz.au 26apr93]: add a syntax to zone files (non rfc standard, but I don't care) to permit RR's to age away at some particular time, and others to become active at some particular time (probably with a syntax something like "<[date]" or "@[date]" preceding, or in the former case, replacing, the TTL field of the record). Approaching "date" in the "<[date]" case, the TTL's on the record would be decreased, so no data cached anywhere will remain valid after "date", after "date", this RR would simply be inoperative (essentially identical to a comment). In the "@[date]" case (or perhaps ">[date]" for symmetry) the RR would be ignored until "date" at which time the "@[date]" field would simply be ignored. Both annotations could be used together (with appropriate interpretations depending on which date is earlier than the other). Annotations on RR's in a zone would cause the SOA parameters to be automatically adjusted in zone transfers (and SOA requests) so that secondary servers would also hand out the same values (dropping the TTL down low as a "<[date]" approaches, and forcing a new zone transfer at "date"). [Paul: I know you said that you'd like to wait for the IETF DNS WG to "bless" an official load balancing scheme, but I'll be adding my shuffle A records to BIND 4.9 for use here at U-M anyhow. The code mods to existing source files are minimal since the bulk of the work is done in a separate .c I added. If you don't want SA records to move into 4.9.1 unless they become official, please just toss this first entry. --bryan@umich.edu] [bryan@umich.edu 25apr93]: add "shuffle A" records There are several schemes for adding some kind of load balancing capability to the DNS. Our "Shuffle Address" (SA) records are one stab at this, and since they're in use at U-M, I need to add them so we can use BIND 4.9 here. [bryan@umich.edu 25apr93]: add AFSDB records AFSDB records were proposed in RFC xxxx. We use them here at the University of Michigan, so I need to add them for our copy of BIND 4.9. [bryan@umich.edu 25apr93]: small fix to resolver's p_cdname() The current copy of p_cdname() in the resolver does not work for query responses larger than 512 bytes (which can happen when using TCP). A very small modification changes the "sanity check" argument (the second one) to dn_expand() from "msg + 512" to "cp + MAXCDNAME". (This showed up very recently.) [gshapiro@wpi.wpi.edu and vixie@pa.dec.com 26apr93]: access control "xfrnets" is ok but what we really need is full access control per zone rather than a global list of acceptable client nets. this is especially important if you send /etc/passwd via zone transfer. [gdmr@dcs.ed.ac.uk 26Apr93]: hesiod support Get Hesiod working properly: add HSprimary and primary/HS, HSsecondary and secondary/HS; patch up inter-class leaks (4.8.3 had LOTS!); add -C class-specifier to named-xfer. [Aside: I see you have some HS "support" in named-xfer already, but it won't work when there are IN- and HS-class zones at the same level in the name tree. My fix to 4.8.3 had named pass named-xfer a -C <class-number> option.] [gdmr@dcs.ed.ac.uk 26Apr93]: SA RR's Add shuffle-address (SA) support. We need this, so I'll attempt to add it in if nobody else does, but it would be better if the original SA author did the work... [jaffe@noc.rutgers.edu 26apr93]: negative caching Negative caching - I've already implemented this in 4.9<beta> and am willing to send you the code any time you tell me you're ready to accept it. It's basically used to keep track of "unknown" hosts for a short time so that you don't waste time processing a lot of requests for a host you already know doesn't exist. Mostly useful on machines that support large mailing lists... [postel@isi.edu anant@isi.edu 28apr93]: negative caching Paul: We'ed like to have included in 4.9.1 our implemention for negative caching. --jon & Anant. [steve@uunet.uu.net 26apr93]: TXT RR improvements - fix TXT records so that they can deal properly with multiple strings (e.g., ``foo IN TXT "aaa" "bbb"''). This results in a fair number of smallish changes throughout the code and also throughout various tools (e.g., nslookup). [steve@uunet.uu.net 26apr93]: X25, ISDN, RT support - add X25, ISDN, and RT records, for sake of completeness (I already have code that should do this, at least for 4.8.3). At least, I figure this should go in unless someone nuked these RR types while I wasn't looking... (-: [steve@uunet.uu.net 26apr93]: core dump on invalid address formats - dotted quints (A.B.C.D.E) in A records may make named dump core. [vixie@pa.dec.com 02may93]: better nameserver tracking, plus data tagging Currently every A RR in the online database (cache or zones in memory) has some fields which are used to track attributes of the name server if this A RR happens to be that of a name server. This is neccessary since we need to keep track of nameserver RTT and sort our queries appropriately to try "fast" or "close" nameservers before we try distant or slow ones. However, due to the way C structures work, this information is maintained for other RR types and for A RR's which are not addresses of nameservers. I propose adding a separate database that tracks neighboring nameservers; no field would be needed in a databuf (RR) to point to this since we can search it by address. Multi-homed nameservers (i.e., with nameservers on hosts which have more than one network interface) would be single-datum, multiple-key. Information to be retained about other nameservers would include RTT as well as error statistics, packet statistics, and so on. This database would be dumpable in the same way that the main cache is now dumpable -- that is, using a signal. At some point we can consider adding SNMP query capability to this database. But most importantly, we can keep in each databuf (RR) the source address of the nameserver which sent us the data -- thus allowing the cache dump to include a tag on each RR that says where it came from. This will help a lot in tracking down corrupt data. Other than continuing to manage the source pool and integrate other folks' patches, this feature may be my only contribution to BIND 4.9.1. [kyle@uunet.uu.net 16may93]: need an option to die if primary zone file missing as of 4.9, a server will not forward a query if it is itself on the NS list for the relevant domain. this means that if a primary server cannot load its zone file, it will not be able to answer queries in that zone -- it won't even forward them. this is arguably correct, since it prevents bad forwarding loops when two or more servers are all unable to load the zone (primary or secondary, with secondary failures being the more common). what is needed is real loop detection such that reasonable non-looping queries can be forwarded. what we're likely to actually get is an option that causes named to just syslog and die if it can't load a primary zone file. note that at present, named is running somewhat bare-assed since an expired zone in a secondary (or missing zone file in a primary) will cause that named to return SERVFAIL for all queries to that zone. if your screwed up primary/secondary server is also the forwarding server for a collection of hosts, those hosts will get SERVFAIL's back from queries to the affected domains, and depending on the age of their resolvers, they might not try other servers after they get the first SERVFAIL. [ this entry was written by Paul Vixie after getting a problem report from Kyle after uu.net disappeared in a brief but ugly way. --vix ] [vixie@pa.dec.com 16may93]: inet_addr needs to die to be replaced by calls to inet_aton, which doesn't confuse the broadcast address with bad addresses. ## ++Copyright++ 1993 ## - ## Copyright (c) 1993 Regents of the University of California. ## All rights reserved. ## ## Redistribution and use in source and binary forms, with or without ## modification, are permitted provided that the following conditions ## are met: ## 1. Redistributions of source code must retain the above copyright ## notice, this list of conditions and the following disclaimer. ## 2. Redistributions in binary form must reproduce the above copyright ## notice, this list of conditions and the following disclaimer in the ## documentation and/or other materials provided with the distribution. ## 3. All advertising materials mentioning features or use of this software ## must display the following acknowledgement: ## This product includes software developed by the University of ## California, Berkeley and its contributors. ## 4. Neither the name of the University nor the names of its contributors ## may be used to endorse or promote products derived from this software ## without specific prior written permission. ## ## THIS SOFTWARE IS PROVIDED BY THE REGENTS AND CONTRIBUTORS ``AS IS'' AND ## ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE ## IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ## ARE DISCLAIMED. IN NO EVENT SHALL THE REGENTS OR CONTRIBUTORS BE LIABLE ## FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL ## DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS ## OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) ## HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT ## LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY ## OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF ## SUCH DAMAGE. ## - ## Portions Copyright (c) 1993 by Digital Equipment Corporation. ## ## Permission to use, copy, modify, and distribute this software for any ## purpose with or without fee is hereby granted, provided that the above ## copyright notice and this permission notice appear in all copies, and that ## the name of Digital Equipment Corporation not be used in advertising or ## publicity pertaining to distribution of the document or software without ## specific, written prior permission. ## ## THE SOFTWARE IS PROVIDED "AS IS" AND DIGITAL EQUIPMENT CORP. DISCLAIMS ALL ## WARRANTIES WITH REGARD TO THIS SOFTWARE, INCLUDING ALL IMPLIED WARRANTIES ## OF MERCHANTABILITY AND FITNESS. IN NO EVENT SHALL DIGITAL EQUIPMENT ## CORPORATION BE LIABLE FOR ANY SPECIAL, DIRECT, INDIRECT, OR CONSEQUENTIAL ## DAMAGES OR ANY DAMAGES WHATSOEVER RESULTING FROM LOSS OF USE, DATA OR ## PROFITS, WHETHER IN AN ACTION OF CONTRACT, NEGLIGENCE OR OTHER TORTIOUS ## ACTION, ARISING OUT OF OR IN CONNECTION WITH THE USE OR PERFORMANCE OF THIS ## SOFTWARE. ## - ## --Copyright--