4.4BSD/usr/src/contrib/bind-4.9/TODO
$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--