[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
INTERNET namespace: who needs it?
Date: Wed, 11 Oct 89 15:00 PDT
From: TYSON@warbucks.ai.sri.com (Mabry Tyson)
The reason I said "if that is what you want" is that, with the domain system,
the NIC host table is somewhat a relic. I believe there are several sites
that are running without the large INTERNET namespace that it generates.
If we get to beta-test Genera 8.0, I'm going to try to build our worlds without
the INTERNET namespace and see if everything still works.
If there are sites that are running without the large INTERNET namespace,
I'd like to hear what they've had to do to get it to work well.
MIT has been running without an internet namespace for at least a year, maybe
more. We have found that things work a lot better without it because
1. We don't have to bother keeping it up to date.
2. Our namespace servers boot a lot faster and have a lot smaller world
loads.
3. We don't have as many namespace related problems without it.
There are three bugs in the system that make this inconvenient.
1. NETI:ENSURE-DOMAIN-SERVICE-ENTRIES mistakenly tries to update the
INTERNET namespace for uninterned hosts with names like INTERNET|10.2.0.6
If you have an internet namespace, you just get these trash hosts in
your namespace, which is why this bug hadn't been discovered.
If you don't have an internet namespace, the function just blows out.
2. Old world loads that believe in the existence of the INTERNET namespace
do not react in a civilized manner when the namespace server tells them
there isn't an Internet namespace. The just immediately ask the
namespace server about the internet namespace again. Just a couple of
hosts behaving this way can bog down the namespace server with their
continuous requests. We worked around this with a real kludge. In
(flavor:method :server-top-level neti:namespace server) we sleep for
10 seconds every time we get an object-not-found-in-search-list error
for the namespace "INTERNET". This keeps the namespace server from
bogging down.
3. The Mailer doesn't handle domain host names in the mail-lists file very
well. If the mailer can't get domain information at the time it loads
the the mailboxes.text file, it decides the host doesn't exist and the
mailboxes.text file is broken. It's real annoying when the mailer fails
to boot whenever we boot our mailer machine. So annoying we
are considering flushing any Symbolics mailers and exclusively use UNIX
mailers if it doesn't get fixed in 8.0.
I have heard this is hard to fix. We work around it by forwarding
addresses to domain-style hosts through other machines, so there is no
domain-style host name after an @. For example, if Mabry were on a
mailing list on our mailer, he would be listed as
Tyson%warbucks.ai.sri.com@mintaka where mintaka were some other host (it
can even be another Symbolics mailer), instead of just
Tyson@warbucks.ai.sri.com.
I can forward you our patches for (1) and (2) if you want them.
Incidentally, even though we have many large namespaces (though none as large
as the internet namespace is), we don't use changes files because they
invariably screw up. Instead, we just let the namespace servers reload entire
namespaces whenever they boot and get updates. This has solved many of our
namespace inconsistency problems. This probably cant be done if you have an
INTERNET namespace, because of its size.