May 2000 - Test - lists.nanog.org

Re: Optical Crossconnects and IP
by Vadim Antonov 09 May '00

09 May '00

> Hop-by-hop routing is on its way out....and not soon enough. Worked fine for the last 20 years. Can you substantiate your assertion? So far all alternatives were shown to bring more problems than improvements. > As for reinventing the virtual circuit, I don't think we can > get much lower than this. I hate to reiterate the obvious, but VCs and global dynamically routed networks don't mix well. --vadim

3 2

Anybody patching Sendmail to filter _attachments_?
by Bill Woodcock 09 May '00

09 May '00

...as opposed to just by subject line? We're getting a lot of requests from customers that we MX for to just throw away either all attachments or any .vbs visual basic attachments. Obviously the Sendmail-Inc.-approved fix is subject-line based, which doesn't take care of that. Anybody else working on this already? -Bill

12 15

Re: Optical Crossconnects and IP
by smd＠clock.org 08 May '00

08 May '00

| Hop-by-hop routing is on its way out....and not soon enough. Mmmm, those drugs they give out in MPLS-land sure are good, aren't they? Sean.

3 2

Japan's NTT Com To Buy Verio
by David Lesher 08 May '00

08 May '00

TOKYO, JAPAN, 2000 MAY 8 (NB) -- By Adam Creed, Newsbytes. NTT Communications Corp., the long distance and international operator of Nippon Telegraph and Telephone Corp. (NTT) [TOKYO:9432], says it has signed a definitive agreement to purchase all the shares of US-based Web site and e-commerce hosting company, Verio Inc. {extracted} $5.5 billion over 525,000 domain names the world's largest operator of Web sites for businesses. {/extracted} So, what will it mean operationally? Spamwise? -- A host is a host from coast to coast.................wb8foz(a)nrk.com & no one will talk to a host that's close........[v].(301) 56-LINUX Unless the host (that isn't close).........................pob 1433 is busy, hung or dead....................................20915-1433

7 8

Re: Optical Crossconnects and IP
by Vadim Antonov 08 May '00

08 May '00

> ** Disclaimer; I work for a vendor: PLURIS, but this question concerns a > personal curiosity that I have about OXCs. ** Disclaimer: I do not work for a vendor: PLURIS > I also don't > see much NSP participation in the relevant standards bodies about signaling > in optical networks. Probably because the said standards bodies (and on-demand fiber capacity allocation in general) have little relevance to the NSPs and Internet backbones in general? --vadim PS I keep wondering why people keep reinventing virtual circuits.

2 1

Optical Crossconnects and IP
by Bora Akyol 08 May '00

08 May '00

** Disclaimer; I work for a vendor: PLURIS, but this question concerns a personal curiosity that I have about OXCs. ** ------------------------------------------------ There has been a lot of work going on about how to utilize OXCs in an internetwork at both OIF and ODSI (and some IDs in IETF). A couple of themes have been emerging: 1. Use the OXCs to set up paths akin to MPLS LSPs. This is great for providers that have MPLS, but what about pure IP-based backbones. 2. Decide which routers connect with what other routers (within the same domain) using configuration. Then use signaling to create circuits between the routers to connect them. I would like to hear about what people are thinking about OXCs, how they are intending on utilizing them, what functionality to they expect from their routers. Let me put in my two cents worth: --------------------------------------------- It is hard not to think of the OXC-based circuit model as another overlay network, but it brings distinct provisioning advantages in terms of speed and flexibility. There apparently are some cost advantages as well. On the other hand, the OXC based optical network is a change in some manner from a converged IP over optical network architecture where IP routers plug directly into DWDMs and have enough ports to handle all of the traffic. For NSPs that have already an MPLS or ATM solution, an OXC-based circuit switched model is not a big leap, but for all IP shops it is. I also don't see much NSP participation in the relevant standards bodies about signaling in optical networks. Thanks in advance Bora Akyol

1 0

GTE DNS?
by Zaid Ali 08 May '00

08 May '00

Is there a known outage on GTE DNS? Zaid ------------------------------------------ Zaid Ali Network Architect WebEx 110 Rose Orchard Way San Jose, CA 95134 Tel: 408-435-7251 fax: 408-435-7004 www.webex.com ----------------------------------------

1 0

Australia OPTUS outage investigated (was Re: DACS blamed for MCI / CSX train problems last week)
by Sean Donelan 08 May '00

08 May '00

On Mon, 08 May 2000, Mohamed Hirse wrote: > A start would be if carriers architecture their network to be redundant > and self healing. Some do, compare Worldcom's reliability to Sprint's reliability. Of course, after the merger things may change (FUD, FUD, FUD). But even then you can have more outages than you can handle. For example, Cable&Wireless OPTUS network in Australia experienced dual fiber outages within a couple of hours bringing much of their network down. http://www.newsbytes.com/pubNews/00/148623.html The most reliable network I know about, FEDWIRE, only has 3-5 way diversity. How does the fed assure the carriers actually route the circuits they way they were ordered? I don't know.

1 0

DACS blamed for MCI / CSX train problems last week
by Sean Donelan 08 May '00

08 May '00

According to Computerworld, the MCI outage affecting CSX rail lines last week was caused by DACS problems. http://www.computerworld.com/home/print.nsf/all/000505DCCE This isn't the first DACS problem nor is MCI the only carrier affected by DACS problems. Bell Atlantic had a multi-day DACS problems, I've experienced 18+ hour MCI DACS problems in the past. What is it about DACS systems which seem to lead to such catastrophic problems? How can you put an entire train system out of operation due to a single piece of equipment?

2 1

Re: Alert timeline for Love Bug
by Neil Long 07 May '00

07 May '00

Just to add that I got an alert from Sophos (another anti-virus vendor) with their detection module at 11:24 (UTC+1) here in the UK They had classified it as Name: VBS/LoveLet-A Type: Visual Basic Script worm Neil

1 0