Re: Network Abuse by Netscape? -- Was: Mosaic replacements, etc...

Rich Brandwein (rhb@hotair.att.com)
Mon, 24 Oct 94 00:20:13 EDT


On Oct 22, 8:36pm, Brian Behlendorf wrote:
>
> Just to clarify - the number of simultaneous TCP connections is
> configurable at run time in Mozilla, the default being 5 I believe.

And this is one of the fundamental problems with this package. We're
getting people here setting this to 12 or 14 or more (or 100 for all I
know - I wonder what the limit is!). It's definitely effecting our
largest server, and I'm in the process of going through analysis of how
bad it actually is coming out to be. Even at *2* it seems to be
pretty devastating to our CERN proxy server, but we'll get some better
stats soon. If anyone out there already has some definitive stats one way
or the other on this issue, I'd also really appreciate it.

We're also not happy about the fact that people's user id's are being
grabbed off the host and inserted into the headers (and in fact their
entire email address will be sent if the email "from" is properly
filled out) FROM field, which is passed in *every* transaction with
no way to turn this off other than put "-" in the from address (or
"robert.e.allen@att.com" in my case - but I guess you can map from that
now...). What happened to all the privacy discussions or have they
moved elsewhere (in our case, and I suspect many other large orgs,
this is also a violation of our security policies)?

-- 
Rich Brandwein				http://www.research.att.com/~rhb/
AT&T Bell Laboratories			Internet:   rich.brandwein@att.com
Room 2G-523 Crawfords Corner Road       Phone:      +1 908 949 2135
Holmdel, NJ  07733-3030                 Facsimile:  +1 908 949 3210

[-- fortune added automagically below (i.e., relationships to the message is magic beyond my control...) --] Meeting, n.: An assembly of people coming together to decide what person or department not represented in the room must solve a problem.