Dents -- Bind alternative?

Just saw an announcement for Dents ( http://www.dents.org/ ) on
comp.os.linux.announce. This is supposedly a (future) replacement for
BIND, as a DNS server. It has some cool ideas (dynamic zones, open storage
interface, control interface, etc), but it looks like it's a work in
progress.

Has anyone looked at this? It seems like it could (theoretically) address
several of the DNS issues that have been discussed here and on inet-access
lately.

I'm not quite sure I understand why these features weren't just
contributed to BIND, but maybe it is time for a fresh start...

Pete.

P.S. Looks like one of the participants works for Mindspring, but there
doesn't appear to be any association between this project and Mindspring.

Just saw an announcement for Dents ( http://www.dents.org/ ) on
comp.os.linux.announce. This is supposedly a (future) replacement for
BIND, as a DNS server. It has some cool ideas (dynamic zones, open storage
interface, control interface, etc), but it looks like it's a work in
progress.

let the record show that:

1. the authors attempted to contact me before their release, but i was
   out of town on a thanksgiving family holiday. they suggested that
   "some friendly competition" would be good for everybody, and i cannot
   argue with that!

2. bind 8.1.2++, now in preparation, will have a control interface (like
   ctlinnd) rather than signals.

3. bind 9.0, also now in preparation but not due 'til next year, will have
   dynamic zones, an open storage interface, and a lot of other stuff.

Has anyone looked at this? It seems like it could (theoretically) address
several of the DNS issues that have been discussed here and on inet-access
lately.

since bind is funded by ISP's, i can only imagine that it is also going to
address most of the DNS issues that have been brought up here and elsewhere.

I'm not quite sure I understand why these features weren't just
contributed to BIND, but maybe it is time for a fresh start...

perhaps the authors think ISC's source isn't "open" enough, or perhaps they
didn't know that ISC was hiring DNS software engineers. or perhaps they
just had some ideas they wanted to try out. (bind 9.0 is a fresh start,
other than the configuration file parser and some of the underlying
libraries.)

coming, as it does, on the heels of a recent announcement from philips (.se)
of a load balancing web proxying dns serving open source content server, it
seems that folks everywhere at once are finding dns service an interesting
area to work in. hopefully there will be a lot of code sharing, to the
betterment of all.

Just saw an announcement for Dents ( http://www.dents.org/ ) on
comp.os.linux.announce. This is supposedly a (future) replacement for
BIND, as a DNS server. It has some cool ideas (dynamic zones, open storage
interface, control interface, etc), but it looks like it's a work in
progress.

It is.

P.S. Looks like one of the participants works for Mindspring, but there
doesn't appear to be any association between this project and Mindspring.

Johannes is a former MindSpring employee; we started the project when
he worked here. He has since gone on to another job. No, MindSpring is
not sanctioning the project, and we are both doing it in our spare time.
MindSpring is donating the web site and the mailing list. MindSpring is
not at present running Dents anywhere, but I hope for that to change once
we get it cleaned up a little. In answer to your question, though, this
is very much a personal project and not an official MindSpring project.

perhaps the authors think ISC's source isn't "open" enough, or perhaps they
didn't know that ISC was hiring DNS software engineers. or perhaps they
just had some ideas they wanted to try out. (bind 9.0 is a fresh start,
other than the configuration file parser and some of the underlying
libraries.)

It was the last option. Neither of us is really interested in working
for the ISC, and bind is certainly very open software (three cheers for
the ISC). We had some ideas that we wanted to try out, so we did.

coming, as it does, on the heels of a recent announcement from philips (.se)
of a load balancing web proxying dns serving open source content server, it
seems that folks everywhere at once are finding dns service an interesting
area to work in.

Well, we found DNS service to be an interesting problem a year and a
half ago when we started this project; it's just now that we've told
the rest of the world. 8^)

Our goal is to advance the state of the art in server design and to give
Paul a run for his money (sic). Like Paul said, we certainly intend for
this to be a friendly competition, but it is a competition, and we do
aim to be serious players in the DNS space.

Look for some announcements in the next few months about dents-enabled
solutions which allow network architects to perform previously-impossible
tricks based on DNS; that's the real meat of this project. If you want
to see your needs met faster, or if you just want to have fun, then

  % echo "subscribe dents-list"|mail majordomo@lists.mindspring.com

If you want to see bind beat the pants off of dents, then I am sure that
Paul would be more than happy to have the help implementing the list he
posted to NANOG earlier.

Let the games begin!

AT&T seems to be completely gone, and Sprint connectivity is getting
progressively worse. Anyone know what's going on?

For instance:

[ Mon Nov 30 12:24:21 ]
[ root@pdx-s02 -> traceroute sprintlink.net
traceroute to sprintlink.net (199.0.232.90), 30 hops max, 40 byte packets
1 pdx-pm01 (209.188.52.11) 1.055 ms 0.801 ms 0.774 ms
2 209.104.198.73 (209.104.198.73) 3.968 ms 4.637 ms 4.334 ms
3 a0-0-17.edge0.pao.priori.net (209.104.192.77) 111.857 ms 50.848 ms
40.709
ms
4 f0-0-0.plalic01.eli.net (207.173.115.249) 40.981 ms 40.323 ms
40.550 ms
5 a0-0--5.scrlic01.eli.net (207.173.113.141) 45.199 ms 48.857 ms
48.961 ms
6 sl-gw8-stk-12-0.sprintlink.net (144.228.107.17) 46.588 ms 276.005 ms
277.
638 ms
7 * * *

etc.

Jim

Jim,

Traceroute from one of my boxes with connectivity to VERIO (NYC).

[techno@amber techno]$ traceroute sprintlink.net
traceroute to sprintlink.net (199.0.232.90), 30 hops max, 40 byte packets
1 209.227.6.17 (209.227.6.17) 2.077 ms 2.020 ms 2.101 ms
2 209.227.17.73 (209.227.17.73) 4.778 ms 4.763 ms 4.705 ms
3 agis-mxol.newyork2.agis.net (206.185.152.29) 23.676 ms 38.185 ms
23.271 ms
4 ga016.philadelphia1.agis.net (206.84.233.10) 15.310 ms 32.279 ms
11.992 ms
5 a2-0.58.pennsauken1.agis.net (206.185.158.10) 39.808 ms 169.346 ms
149.762 ms
6 192.157.69.9 (192.157.69.9) 45.326 ms 34.370 ms 36.452 ms
7 sl-bb10-pen-5-3.sprintlink.net (144.232.5.61) 31.292 ms 35.377 ms
24.426 ms
8 sl-bb12-pen-9-0.sprintlink.net (144.232.5.153) 30.801 ms 29.427 ms
31.403 ms
9 sl-bb10-rly-5-0.sprintlink.net (144.232.8.153) 141.616 ms 77.702 ms
94.215 ms
10 sl-bb5-dc-0-0-0.sprintlink.net (144.232.7.162) 94.506 ms 71.646 ms
81.569 ms
11 208.28.7.19 (208.28.7.19) 120.286 ms 106.853 ms 106.874 ms
12 sl-noc-1-4-1-T1.sprintlink.net (144.228.129.130) 51.890 ms 36.726 ms
64.077 ms
13 * tiny.sprintlink.net (199.0.232.90) 44.001 ms *

AT&T looks better than it ever has for us..

  1 nac-nyc6-e0.nac.net (207.99.10.206) 4 msec 4 msec 8 msec
  2 s2-0.core1.whi.nac.net (209.123.11.145) 4 msec 4 msec 8 msec
  3 nac-globalcenter-Fa2-1-100mb.nac.net (207.99.5.191) 8 msec 4 msec 0 msec
  4 vc37.atm1-0.cr1.DCA.globalcenter.net (206.132.191.162) 12 msec 12 msec 12 msec
  5 fe2-0.br1.DCA.globalcenter.net (204.152.166.13) 16 msec 12 msec 12 msec
  6 pos6-0-0-155M.cr1.IAD.globalcenter.net (204.152.166.6) 16 msec 48 msec 16 msec
  7 fe0-0.br1.IAD.globalcenter.net (204.152.166.133) 20 msec 16 msec 16 msec
  8 12.127.45.129 12 msec 20 msec 16 msec
  9 br3-a3110s1.wswdc.ip.att.net (12.127.1.18) 16 msec 20 msec 20 msec
10 br1-h11.cgcil.ip.att.net (12.127.15.193) 96 msec 88 msec 104 msec
11 br2-a350s2.cgcil.ip.att.net (12.127.14.234) 104 msec 92 msec 92 msec
12 br1-h11.sl9mo.ip.att.net (12.127.15.210) 108 msec 112 msec 112 msec
13 dc2-h100.bgtmo.ip.att.net (12.127.15.33) 108 msec 104 msec 104 msec

AT&T seems to be completely gone, and Sprint connectivity is getting
progressively worse. Anyone know what's going on?

For instance:

[ Mon Nov 30 12:24:21 ]
[ root@pdx-s02 -> traceroute sprintlink.net
traceroute to sprintlink.net (199.0.232.90), 30 hops max, 40 byte packets
1 pdx-pm01 (209.188.52.11) 1.055 ms 0.801 ms 0.774 ms
2 209.104.198.73 (209.104.198.73) 3.968 ms 4.637 ms 4.334 ms
3 a0-0-17.edge0.pao.priori.net (209.104.192.77) 111.857 ms 50.848 ms
40.709
ms
4 f0-0-0.plalic01.eli.net (207.173.115.249) 40.981 ms 40.323 ms
40.550 ms
5 a0-0--5.scrlic01.eli.net (207.173.113.141) 45.199 ms 48.857 ms
48.961 ms
6 sl-gw8-stk-12-0.sprintlink.net (144.228.107.17) 46.588 ms 276.005 ms
277.
638 ms
7 * * *

etc.

Jim
  __________________________________________________________________
  Jim Dawson jdawson@navi.net
  GCN Communications, Inc. http://www.navi.net
  618 NW Glisan St. Ste. 407 voice: +1.503.641.1449
  Portland, Or 97209 USA fax: +1.503.641.3634
  ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~

-- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- --
     Atheism is a non-prophet organization. I route, therefore I am.
       Alex Rubenstein, alex@nac.net, KC2BUO, ISP/C Charter Member
               Father of the Network and Head Bottle-Washer
     Net Access Corporation, 9 Mt. Pleasant Tpk., Denville, NJ 07834
Don't choose a spineless ISP; we have more backbone! http://www.nac.net
-- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- --

Well, it would make sense that those who can see at&t will have better
response when most other tier 1's can't see it at all. No traffic from the
big players on the at&t network = less traffic and better performance.

AT&T looks better than it ever has for us..

1 nac-nyc6-e0.nac.net (207.99.10.206) 4 msec 4 msec 8 msec
2 s2-0.core1.whi.nac.net (209.123.11.145) 4 msec 4 msec 8 msec
3 nac-globalcenter-Fa2-1-100mb.nac.net (207.99.5.191) 8 msec 4 msec 0 msec
4 vc37.atm1-0.cr1.DCA.globalcenter.net (206.132.191.162) 12 msec 12 msec
12 msec
5 fe2-0.br1.DCA.globalcenter.net (204.152.166.13) 16 msec 12 msec 12 msec
6 pos6-0-0-155M.cr1.IAD.globalcenter.net (204.152.166.6) 16 msec 48 msec
16 msec
7 fe0-0.br1.IAD.globalcenter.net (204.152.166.133) 20 msec 16 msec 16 msec
8 12.127.45.129 12 msec 20 msec 16 msec
9 br3-a3110s1.wswdc.ip.att.net (12.127.1.18) 16 msec 20 msec 20 msec
10 br1-h11.cgcil.ip.att.net (12.127.15.193) 96 msec 88 msec 104 msec
11 br2-a350s2.cgcil.ip.att.net (12.127.14.234) 104 msec 92 msec 92 msec
12 br1-h11.sl9mo.ip.att.net (12.127.15.210) 108 msec 112 msec 112 msec
13 dc2-h100.bgtmo.ip.att.net (12.127.15.33) 108 msec 104 msec 104 msec

- ------------------------------------------------------------------
Get your *FREE* Parked Domain account at http://www.EZ-Hosting.Com
- ------------------------------------------------------------------
John Fraizer | __ _ |
The System Administrator | / / (_)__ __ ____ __ | The choice
mailto:John.Fraizer@EnterZone.Net | / /__/ / _ \/ // /\ \/ / | of a GNU
http://www.EnterZone.Net/ | /____/_/_//_/\_,_/ /_/\_\ | Generation
PGP Key fingerprint = 7DB6 1CA2 DAA6 43DA 3AAF 44CD 258C 3D7E B425 81A8

Does anybody know if verio actually has ANY backbone whatsoever? It
appears that they just resell, and they dont even resell just 1 network,
they resell a different backbone provider depending on what town they are
in?

I do know that Verio came in and bought out Spacelab.net, which was our
provider. The color & designs of the invoices changed. All the staff stayed
the same, none of my favorite engineers were lost. Verio might be building
their "network" on top of existing local networks and then hauling this
traffic to their peering providers. Take a look at:
http://www.verio.com/netops/netmap.html.