Re: [j-nsp] MX10003 rack size

[Anderson, Charles R]

1000mm deep.  APC AR3100 racks are 600mm x 1070mm.  APC also makes 1200mm deep 
ones, and 750mm wide ones, and both together.

On Wed, Aug 07, 2019 at 04:12:26PM +, Richard McGovern wrote:
> Pete "1000 deep rack"??  Is that fathoms __
> 
> Richard McGovern
> Sr Sales Engineer, Juniper Networks 
> 978-618-3342
>  
> I’d rather be lucky than good, as I know I am not good
> I don’t make the news, I just report it
>  
> 
> On 8/7/19, 6:20 AM, "Pete Webb"  wrote:
> 
> No mate,
> I made the same mistake.
> Minimum you can get away with is 1000 deep racks, and even then you have 
> to leave the front air filter off.
> 
> Pete
> 
> -Original Message-
> From: juniper-nsp  On Behalf Of 
> Sander Steffann
> Sent: 30 July 2019 13:32
> To: nanog ; Juniper List 
> Subject: [j-nsp] MX10003 rack size
> 
> Hi,
> 
> Has anyone ever managed to fit a Juniper MX10003 in a 90cm deep rack? 
> Without applying power tools to either the rack or the router ;)
> 
> Cheers,
> Sander

Re: few big monolithic PEs vs many small PEs

[Anderson, Charles R]

On Fri, Jun 21, 2019 at 09:01:38AM -0500, Aaron Gould wrote:
> I was reading this and thought, planet earth is a single point of failure.
> 
> ...but, I guess we build and design and connect as much redundancy (logic, 
> hw, sw, power) as the customer requires and pays for and that we can 
> truly accomplish.

Fate sharing is also an important concept in system design.

Re: BGP prefix filter list

[Anderson, Charles R]

What about these ones?

https://teamarin.net/2019/05/13/taking-a-hard-line-on-fraud/

On Wed, May 15, 2019 at 01:43:30PM +0200, Baldur Norddahl wrote:
> Hello
> 
> This morning we apparently had a problem with our routers not handling 
> the full table. So I am looking into culling the least useful prefixes 
> from our tables. I can hardly be the first one to take on that kind of 
> project, and I am wondering if there is a ready made prefix list or similar?
> 
> Or maybe we have a list of worst offenders? I am looking for ASN that 
> announces a lot of unnecessary /24 prefixes and which happens to be far 
> away from us? I would filter those to something like /20 and then just 
> have a default route to catch all.
> 
> Thanks,
> 
> Baldur

Re: a quick survey about LLDP and similar

[Anderson, Charles R]

On Thu, Feb 28, 2019 at 10:00:55AM +0100, Pierfrancesco Caci wrote:
> 
> Hello,
> having a bit of a debate in my team about turning on LLDP and/or CDP.
> I would appreciate if you could spend a minute answering this
> survey so I have some numbers to back up my reasoning, or to accept
> defeat.
> 
> https://www.surveymonkey.com/r/TH3WCWP
> 
> Feel free to cross-post to other relevant lists. 
> 
> Thank you
> 
> Pf

We require LLDP/LLDP-MED to configure our VOIP phones.

For trunk links, it is extremely helpful to verify correct topology.

For datacenters, it is EXTREMELY helpful to verify hypervisor connectivity.

Re: fs.com dwdm equipment

[Anderson, Charles R]

I concur.  I have also used CWDM and DWDM optics and they are fine.  I have had 
one QSFP+ optic go bad.

On Mon, Feb 18, 2019 at 07:47:10PM +, Brian R wrote:
> Samir,
> 
> I have purchased over a thousand SFPs from Fiber Store.  I can recall less 
> than 5 having problems when we received them (not all even DOA) and I know 
> less than 10 dead even after deployment.  Some we have ad running for 4+ 
> years.  We have done very little with their SFP+ equipment, really only 
> testing and a few lower priority links.
> The only downside to the SFPs that we found was the variance of power.  Say 
> an Adtran, Cisco, Juniper, HP SFP is rated from -3dB to -8 db (all units I 
> have used them with), the equivalent FS direct SFP we have seen as hot as 3dB 
> and as weak as -15dB.  These extremes are fairly rare but we have still seen 
> them.
> Distribution (approximate):
> 80% SM single fiber SFPs (mostly 10km - 40km, some 60km & 80km)
> 7% 1Gb Copper
> 5% MM SFPs
> 5% SM dual fiber SFPs
> 3% others (SFP+, GPON, testing, etc)
> 
> I have not used them for any DWDM applications and only used them a few times 
> on an older CWDM link that used standard SFPs into a MUX.  This was not over 
> great distances (less than 40 miles).  With WDM SFP power consistency was 
> important so we did not play much with it, granted most of the SFPs I am 
> purchasing are in the $10-$25 range so take the extremes with a grain of salt.
> 
> Their sales has always been very responsive and helpful.  The 
> support/engineering, the few times I worked with them, were helpful but the 
> language barrier was harder here.
> 
> Brian
> 
> 
> From: NANOG  on behalf of Samir Rana 
> 
> Sent: Sunday, February 17, 2019 12:42 PM
> To: nanog@nanog.org
> Subject: fs.com dwdm equipment
> 
> Hello All,
> 
> Does anybody have experience with fs.com their production environment? Are you they working without any issue? How's 
> their warranty support if the issue arises?
> 
> Thanks in advance for all the answers and help.

Re: A few GPON questions...

[Anderson, Charles R]

On Tue, Dec 11, 2018 at 07:07:49PM +0100, Baldur Norddahl wrote:
> >
> >
> > And WDM gear if necessary...heck even passive CWDM if you have a riser
> > space issue.
> >
> 
> WDM is much more expensive than GPON.
> 
> I am still waiting for one of the 10G PON variants to become available. We
> want to deliver 10G to customers as >1G is becoming common on CPE Wi-Fi
> routers. But doing it with WDM is too expensive and p2p uses more fiber
> than we have.

Passive CWDM is cheap and supports 10gig.

Re: A few GPON questions...

[Anderson, Charles R]

On Tue, Dec 11, 2018 at 05:36:47PM +, Aled Morris via NANOG wrote:
> On Tue, 11 Dec 2018 at 17:30, Jason Lixfeld  wrote:
> > There’s only so much space in conduits, risers and ducts.  At some point, 
> > scale would press this up against physical infrastructure realities 
> > depending on how far the active gear at the head end is from the subscriber.
> 
> A point made earlier was that typically in a campus environment, most
> every riser cupboard has access to power so you can easily build a
> regular Ethernet LAN with a switch on every floor/corridor/hub.
> Basically, everywhere that you'd put a GPON splitter.

And WDM gear if necessary...heck even passive CWDM if you have a riser space 
issue.

Re: Cogent charging 50/mo for BGP (not IPs, the service)

[Anderson, Charles R]

I was told they only charge it if you have bigger than a /29 from them.

On Wed, Oct 17, 2018 at 04:12:01PM +, David Hubbard wrote:
> They charge it even if you’re using your own address space.  It’s a fee 
> simply for establishing BGP with them on a given circuit.  I believe if you 
> used static routes and their space, you would not have to pay it.
> 
> From: NANOG  on behalf of Josh Luthman 
> 
> Date: Wednesday, October 17, 2018 at 12:10 PM
> To: Brielle Bruns 
> Cc: NANOG list 
> Subject: Re: Cogent charging 50/mo for BGP (not IPs, the service)
> 
> I view Cogent IP space as a way to lock customers to their service, ie make 
> them sticky.
> Josh Luthman
> Office: 937-552-2340
> Direct: 937-552-2343
> 1100 Wayne St
> Suite 1337
> Troy, OH 45373
> 
> On Wed, Oct 17, 2018, 12:03 PM Brielle Bruns 
> mailto:br...@2mbit.com>> wrote:
> On 10/17/2018 9:47 AM, Josh Luthman wrote:
> > Has anyone else dealt with this mess?  Even my Cogent rep admits it's
> > unique to their business.
> 
> That sounds like the BS the first company I worked for tried to pull.
> 
> One would think they'd welcome customers bringing their own IP space
> since it saves them money by not using up precious Cogent IPv4 address
> space.
> 
> Hell, I even have BGP for v4 and v6 over my CenturyLink biz fiber, and
> its available as part of the enhanced package they offer with no extra fees.

Re: ARIN RPKI TAL deployment issues

[Anderson, Charles R]

On Wed, Sep 26, 2018 at 02:18:43PM -0700, Mark Milhollan wrote:
> On Tue, 25 Sep 2018, Job Snijders wrote:
> 
> >We really need to bring it back down to "apt install rpki-cache-validator"
> 
> You say this as if no packager has a way to display and perhaps require 
> approval of the license nor any way to fetch something remote as part of 
> the installation process, e.g., the Microsoft "freely" supplied TTF 
> files ...
>   [...]
> 
> I bet apt, dnf, pacman, pkg_add, yum, etc., do as well -- actually I 
> know some of those do.  Perhaps fetching as part of installing is less 
> desireable than already present at the outset, but it might appease ARIN 
> and be workable (or superior) for many.

rpm/yum/dnf do NOT have a way to allow the installation of packages to
interact with the user.  They specifically block this functionality
since it goes against their design of allowing non-interactive
installs.