- NANOG - lists.nanog.org

router monitor/configuration tool (fwd)
by Dave Scrivens 09 Jun '97

09 Jun '97

I was wondering about this myself, has anyone seen the Linmor NMS system. This system has been chosen by MCI after extensive evaluation of existing Network Management Systems to manage there ATM Hyperstream Network. This Network Management Provides several other interesting functions such as Object Oriented Class for an object. It also has a Web Page which can list all objects & the traffic statistics for each link. As well it can do the below. From it a QOS of service report can be generated indicating the health of of your Network. Reports on the Uptine of your BGP routing process as well as OSPF & EIGRP can be generated. It also has the ability to Monitor Unix machines & " Mission " critical processes such as NAMED or mail & News processes. You can actually get an indication that the memory on the Unix box is low & that is the problem. & Unlike HP Openview, Autodiscovery works very nicely on a large Network. Instead of crashing your Network, you get a pretty nice map infront of you. It is also a distributed system architecture where you can have multiple Data Collectors spread out through your Network talking to each other, exchaning information about the Health of the Network. & you can function in Hot Standby mode where if one of your Data Collectors goes down, another will take up the load. I have seen a SPARC IPC manage over 1500 Network Router Objects including including Cisco 7000's & Cisco 7513's. So even though this is alot for a little SPARC Server, it worked pretty good. It is not highly recommeneded to run your Network like this but as a test .... You can also run reports on your Network & figure out where your facilities such as the Memory on Core & Border Routers need to be upgraded or where you are running out of bandwidth & need to add more. As well, remote site customer management where the customer can be given an account to actually monitor his connection with his provider, from his provider, works good. A check for the customer to ensure that he is getting good service from his provider. & since you can get remote site access, you can actually get a Network Management Station from another site. This is a Really neat product, atleast from what I have seen. Has anyone else encountered it ? Anyway check it out ..... WWW.LINMOR.COM They have some really neat stuff.. thanks, Dave Scrivens, iSTAR Internet. > > based on a recent email about bgp peer monitoring and others over the > last few months, we thought this would be a good time to release a tool > called rtrmon (router monitor) that we've been working on with vixie > enterprises. > > please see: www.vix.com/rtrmon > > vixie enterprises provided the development expertise and based the design > on the needs of genuity initially. it's fairly generic and extensible. > currently it: > > - monitors bgp sessions and reports failures > - monitors cpu usage and reports failures > - monitors ospf neighbors > - downloads, diffs, and archives cisco configs > > it's all cisco oriented as that's our primary platform. we'll start a > mailing list soon for discussion (next week) but in the short interrim > please send mail to myself or danny mcpherson (danny(a)genuity.net) for > questions or problems. we'll have this hosted on our own web site(s) > sometime next week as well. enjoy. > > -brett >

1 0

Re: NSP ... New Information (really)
by Larry Vaden 09 Jun '97

09 Jun '97

At 11:39 AM 6/9/97 PDT, Yakov Rekhter wrote: >Load on the routing system could be reduced if you would take a block >of addresses from UUNET, and a block of addresses from ACSI, number >some of your customers out of the UUNET block, some of your customers >out of the ACSI block, and then use something along the lines of "auto >injection" (see draft-bates-multihoming-00.txt) to handle fallback >connectivity. > >Yakov. Yakov, thanks. I'll take your suggestion and ask Kim Hubbard if this meets the "no favored connection" guidelines. At 02:24 PM 6/9/97 CDT, Stan Barber wrote: >ftp://ftp.sesqui.net/pub/internet-drafts/draft-bates-multihoming-00.txt.Z >should get you a copy of this draft. SESQUINET maintains a mirror of >all Internet Drafts. >-- >Stan | Academ Consulting Services |internet: sob(a)academ.com >Olan | For more info on academ, see this |uucp: {mcsun|amdahl}!academ!sob >Barber | URL- http://www.academ.com/academ |Opinions expressed are only mine. Stan, thanks for the URL, as well as for the sesqui.net news feed also. Regards, Larry Vaden, founder and CEO help-desk 903-813-4500 Internet Texoma, Inc. <http://www.texoma.net> direct 903-870-0365 bringing the real Internet to rural Texomaland fax 903-868-8551 Member ISP/C, TISPA and USIPA pager 903-867-6571

1 0

Re: NSP ... New Information
by Larry Vaden 09 Jun '97

09 Jun '97

At 01:55 PM 6/9/97 -0500, Jeremy Porter wrote: > >In message <3.0.1.32.19970608231642.00d49e1c(a)texoma.net>, Larry Vaden writes: >>At 10:54 PM 6/8/97 -0500, Jeremy Porter wrote: >>>This means that you should use renumbering technology as much as >>>possible, so that when you can justify a /19 you can get one, >>>until then, find an ISP that is multi-homes and can provide you >>>the IP addresses you need, and consider two connections to them, >>>giving you the benefit of being multi-homed without the cost of >>>the routers needed for full BGP, or needing to justify a /19. >> >>A quick question: >> >>What is the immediate utilization rate required for an ISP's first /19? >> >>Please provide a basis in RFC2050 et al relevant documents for your answer. > >The /19 is required to be utilized in 6 months, as part of slow start. >(Under Internic guidelines, under ARIN this could be changed.) >Check out www.arin.net, www.ripe.net, and www.apnic.net for the >indivdual registries polices. > >--- >Jeremy Porter, Freeside Communications, Inc. jerry(a)fc.net >PO BOX 80315 Austin, Tx 78708 | 1-800-968-8750 | 512-458-9810 >http://www.fc.net Jeremy, that's the way I read it also, but Kim Hubbard of InterNIC says an ISP applying for their first /19 must show 100% immediate utilization of a /19 to obtain one from InterNIC. I don't see the part of RFC2050 which supports Kim's interpretation, especially for a currently multi-homed ISP. Do you or others? Dr. Jon Postel has also been silent in regard to pointing out the authority for the 100% immediate utilization rate. Thanks, Larry Vaden, founder and CEO help-desk 903-813-4500 Internet Texoma, Inc. <http://www.texoma.net> direct 903-870-0365 bringing the real Internet to rural Texomaland fax 903-868-8551 Member ISP/C, TISPA and USIPA pager 903-867-6571

2 1

Re: NSP ... New Information
by Larry Vaden 09 Jun '97

09 Jun '97

At 11:39 AM 6/9/97 PDT, Yakov Rekhter wrote: >Larry, > >> Are you saying that a /19 wouldn't put less load on the _system_ than the >> current situation + another block? >> >> Our /23 and /21 are from UUNET, our primary provider. ACSI is secondary >> provider. >> >> Teach me (read: offer your constructive criticism). I'm willing to learn. >> > >Load on the routing system could be reduced if you would take a block >of addresses from UUNET, and a block of addresses from ACSI, number >some of your customers out of the UUNET block, some of your customers >out of the ACSI block, and then use something along the lines of "auto >injection" (see draft-bates-multihoming-00.txt) to handle fallback >connectivity. > >Yakov. Yakov, Thanks for the constructive input. I'll try to locate the full URL and we're interested in evaluating and following your suggestion. Will this meet the InterNIC guidelines of "no favored connection"? Regards, Larry Vaden, founder and CEO help-desk 903-813-4500 Internet Texoma, Inc. <http://www.texoma.net> direct 903-870-0365 bringing the real Internet to rural Texomaland fax 903-868-8551 Member ISP/C, TISPA and USIPA pager 903-867-6571

1 0

RE: the original thirteen NSFNET regionals
by Tony Hain 09 Jun '97

09 Jun '97

Actually to complete the story, MFEnet preceded HEPnet & SPAN as DEC was still developing the DECnet protocol at the time. When Sid put together the proposal for SDSC, it was a clone of the operating and successful MFE Computer Center (down to the floor plans of the building). The CTSS operating system for the Cray at that time included the MFEnet protocol. In that same time frame the MFEnet itself was running out of address space and worked through a transition to TCP by running MFEnet over IP. ---------- From: Hans-Werner Braun [SMTP:hwb@rwac.ucsd.edu] Sent: Monday, June 09, 1997 10:19 AM To: Kent W. England Cc: nanog(a)merit.edu Subject: Re: the original thirteen NSFNET regionals > It's also worth noting that DECNET was a leading contender for the NSFNET. > Sid Karin, director of SDSC, tells the story that he was leaning toward > DECNET until he met a taciturn Teutonic network engineer from Ann Arbor who > convinced Sid among others that a TCP/IP backbone would work. You actually do not quite have your story right. Sid used MFENET (Magnetic Fusion Energy) protocols for the SDSCnet (SDSC consortium) satellite network initially. I was also the networking person at the University of Michigan responsible for the UMichigan-SDSCnet connection (with MFENET protocols) at that time (besides other responsibilities). It did work, and SDSC continued to use it despite the NSFNET using IP. Over time SDSC phased MFENET out and IP in, but besides the reponsibility of the taciturn Teutonic network engineer from Ann Arbor for the NSFNET, I don't think he had much to do with the phaseout of the MFENET usage at SDSC. Sid still claims that he is proud of both his MFENET decision as well as the decision to phase it out, as time moved on.

1 0

re: the original thirteen NSFNET regionals
by Rob Gutierrez 09 Jun '97

09 Jun '97

> >Does anyone remember the NSF-sponsored regional networks? Can you fix > >this table--this is off the top of my head so I am missing a few networks. Does NSI (NASA Science Internet) qualify as one of those? Yeah, it was 1989 when it started (Hi Milo!), but ... :) > Name Region NOC was or is at Now > > BARRnet Bay Area Stanford? Berkeley? BBN WR > ^SU then E. Bayshore Rd., Palo Alto The "NOC" (cough) was in Pine Hall at Stanford next to long-timer Ron Roberts office. After the BBN buyout, NOC duties were officially given to BBN Cambridge, somewhat bypassing the E. Bayshore/Palo Alto NOC office. The western region NOC was supposed to open up (under the capable hands of Chris Tsuboi, now with @Home), but time factors never gave Chris the chance before his departure from BBN-WR. All calls went to Cambridge and were referred to Chris as needed. The NOC office is/was still there as a showcase to visitors in the large conference room. And ... > Network Contact in 1990, where are they now? > > BARRNET Paul Baer > > NEARNET John Rugo Paul Baer was part of the original BARRNet admin staff when William "Bill" Yundt got the ball rolling. Paul came over to BBN after the buyout on a consultant basis, but I don't know if he has stayed on at this present day. Bill Yundt also came over to the BBN side, then went onward with Web TV Networks. John Rugo was running NEARNet when he was asked to run BARRNet after the buyout. After a 2-3 year stint in that position, he went on to be on the senior management team at Cisco in their then new ISP group. The only people left from the original BARRNet at BBN-WR is Vince Fuller and Bill Yundt's admin, Nora Lundin. rob.

9 9

Loss of BGP peering monitoring via SNMP
by Curtis Generous 09 Jun '97

09 Jun '97

We have been caught several times lately with some of our border routers loss of BGP peering going undetected for sometime. Is there a way to enable BGP peering loss monitoring via SNMP from a CISCO router? I've looked at both the public and private MIBS but haven't found anything. I am aware of the 'nocol' package but really was hoping for an SNMP based approach to tie into our net-mgmt packages (sunnet and hpov). TIA --curtis -- Curtis Generous generous(a)uucom.com Phone: (703) 461-1350 UUcom Inc., Suite 250, 4875 Eisenhower Avenue, Alexandria, VA 22304-0797

3 2

Re: Information re: Cyberpromo
by Mark E Larson 09 Jun '97

09 Jun '97

Hmmmm. Let's see. Shutoff peering with AGIS. Tell all their customers to leave. Hoping for some business :) At 11:30 PM 6/2/97 -0700, you wrote: > >AGIS has some 175 or so reseller customers, some of whom are largeish. >A number of those are dual-homed or peer with "nearly everyone" at >one or more exchange points, though. In the last few weeks, I've seen >over 10 midsized ISPs announce they were leaving AGIS, many more are talking >about doing so. Both because of ongoing spam support and their low >quality of service (some inherent, some apparently people attacking >Cyberpromo and/or AGIS after being spammed, unfortunately) I now >recommend to all the ISPs I do or have consulted with that are >AGIS customers that they leave. > >I've never actually seen a large network kill itself before. >It looks like they've got circa 6 months to live at this rate. >This is an interesting process... > >-george william herbert >gherbert(a)crl.com >Disclaimer: I speak for nobody, not even my cats. > > Mark E Larson Chief Engineer Senior Network Architect RUSTnet Inc. ----------------------------------------------- R U S T N E T I N C. A member of the Verio Group http://www.rust.net -----------------------------------------------

1 0

Re: NSP ... New Information
by Bill Sommerfeld 09 Jun '97

09 Jun '97

Date: Mon, 9 Jun 1997 00:34:03 -0400 (EDT) From: Edward Fang <edfang(a)visi.net> This is all great and dandy, but then why does it appear that anybody with a cable modem this side of the sun are using static IP's. Huh? When I switched from dialup ISDN to cable-modem service (MediaOne) I went from a static address to a DHCP address. I've been forcably renumbered at least once since then due to a topology change.

1 0

Re: Top 100 Nameservers (Serving the most Domains)
by Havard.Eidnes＠runit.sintef.no 09 Jun '97

09 Jun '97

Hi, nice report! It's not entirely clear what your data source is; I guess it's the zone files for the domains currently under InterNIC's administration, yes? To prevent anyone from taking this report and relabel it as "The Internet's top 100 name servers (by domains served)" (even in their own minds...), I just want to point out that there exists other TLDs on the Internet than those maintained by the InterNIC, and I know for a fact that some name servers in other areas would be included in your list if you had analyzed the data. Regards, - Håvard

3 2