- NANOG - lists.nanog.org

Any one here from CyrusOne ?
by Bob Evans 09 May '17

09 May '17

Hi, Looking for off-line CyrusOne NOC assistance to help our mutual customers reach each others servers. I do not think the issue is CyrusOne's , but it is most likely a CyrusOne customer's that has no network people that comprehend routing issues. 2 days now , I need a little insight. My work around is via a transit provider that does not go through a Cyrusone hop. Whenever Cyrusone and gramtel.net hop appears customer packets drop at gramtel.net hop. On GTT from Amsterdam to ihotelier.com IPv4 traceroute to 199.167.220.52 HOST: cr2-ams1-re1 Loss% Snt Last Avg Best Wrst StDev 1. lag-12.ear3.Amsterdam1.Level 0.0% 5 601.2 121.0 0.8 601.2 268.4 2. ??? 100.0 5 0.0 0.0 0.0 0.0 0.0 3. CYRUSONE-LL.ear2.Chicago2.Le 0.0% 5 94.3 94.4 94.2 94.7 0.2 4. 169.64.242.209.gt001.gramtel 0.0% 5 95.6 94.8 94.4 95.6 0.5 5. ??? 100.0 5 0.0 0.0 0.0 0.0 0.0 >From Chicago ... IPv4 traceroute to 199.167.220.52 HOST: cr1-chi1-re1 Loss% Snt Last Avg Best Wrst StDev 1. as3356.chi11.ip4.gtt.net 20.0% 5 8.4 3.0 1.0 8.4 3.6 2. ??? 100.0 5 0.0 0.0 0.0 0.0 0.0 3. CYRUSONE-LL.ear2.Chicago2.Le 0.0% 5 1.9 1.9 1.9 2.0 0.0 4. 169.64.242.209.gt001.gramtel 0.0% 5 2.1 2.4 2.1 3.1 0.5 5. ??? 100.0 5 0.0 0.0 0.0 0.0 0.0 On Hurricain Electric from Fremont to ihotelier.com hits and stops at gramtel.com core1.fmt1.he.net> traceroute 199.167.220.57 source-ip 216.218.252.161 numeric Target 199.167.220.57 1 1 ms <1 ms <1 ms 10ge7-3.core1.sjc2.he.net (72.52.92.110) 2 <1 ms <1 ms 14 ms asn-qwest-us-as209.10gigabitethernet10-10.core1.sjc2.he.net (216.218.230.250) 3 51 ms 89 ms 61 ms cer-edge-19.inet.qwest.net (67.14.122.141) 4 132 ms 48 ms 59 ms 65.123.102.162 5 63 ms 48 ms 52 ms 209.242.80.97 6 48 ms 49 ms 50 ms 169.64.242.209.gt001.gramtel.net (209.242.64.169) 7 * * * ? 8 * * Thank You Bob Evans CTO

1 0

Anyone here from ihotelier.com or travelclcik.com or gramtel.net
by Bob Evans 09 May '17

09 May '17

Hello, I have 3 customers experiencing routing issues all day to admin.ihotelier.com When the problem occurs the trace stops and drops at a gramtel.net router or server. That traces through GTT then Zayo and halts at gramtel.net. When I put in a temp static around it via another transit it hops through PNAP.net and works fine. I would like to get rid of my temp route for the admin.ihotelier.com /24 range. Thanks Bob Evans CTO

1 0

AusNOG 2017 Call for Presentations
by Mark Prior 09 May '17

09 May '17

The next AusNOG conference is only four months away, scheduled to be held at The Langham, Melbourne, on 7 & 8 September 2017, so we are making our first call for presentations of interest to the Internet operations community. Previous feedback has suggested that attendees would like more content from the community so we'd like to encourage members of the community to submit more proposals, especially if you haven't previously presented! Internet operations is a broad topic but some areas of interest are: * Improving the redundancy/resiliency/sustainability of your network * Using automation to improve service reliability * Taking the network operations BCP to the next level * Using 'offline' communications tools to keep the network working * Evolution in network architectures and scaling issues * Analysis of relevant major operational events of the year Naturally other topics that are of interest to the network operator community are also welcome but presentations of a purely marketing nature are not welcome at this technical event. Notes to presenters =-=-=-=-=-=-=-=-=-= Preference will be given to presentations that result in actual operational outcomes. Session speakers should be prepared to present for 30 minutes. Keynote speakers will be expected to present for 45 minutes to 1 hour. Short presentations of strong operational content are also acceptable. Please note any timing constraints when submitting a presentation. By submitting a presentation for consideration in the AusNOG 2017 programme, and if selected the presenter will allow AusNOG to: * Take photographs of the presentation and presenter * Record and rebroadcast video and audio of the presentation and presenter * Redistribute the presentation slides, audio, video, and photographs electronically, on the AusNOG website, or otherwise but leaving all intellectual property in the hands of presenter or rightful property holder[1]. Deadlines =-=-=-=-= 12 June: Submission of presentation title, presentation description (300 words), and presenter biography (200-400 words) 10 July: Presenters notified of their acceptance status as an AusNOG 2017 presentation. 30 August: Submission of final presentation slides as Portable Document Format (pdf), PowerPoint, or Keynote. Provision of a recent digital photograph (<500k) of the presenter. All submissions must be sent by email to organisers AT ausnog DOT net. A separate call for lightning talks will be made closer to the AusNOG 2017 event. As in previous years you must be registered to attend AusNOG 2017 in order to be considered for a lightning talk. [1]: AusNOG accepts that some speakers are unable to allow us to archive their presentation due to company or corporate policy, and if the situation arises AusNOG will delete all copies in its possession after the presentation.

1 0

outage
by Randy Bush 08 May '17

08 May '17

cf brian nisbet's talk this (cest) afternoon: a fair number of researchers are obsessed with outage detection. they have a problem; outages where one can know the 'ground truth' (operator confirmation of what actually happened down to the links and times) is very hard to get. so, it would be helpful if some core networks would either report the details of an outage every week or so, or create a nice variety of planned outages and descrive the details. randy

1 0

Interconnection Track
by Mehmet Akcin 08 May '17

08 May '17

Hello, As promised few months ago publically I have volunteered to bring together content to have Peering Track back to agenda. Now called "Interconnection Track" I would like to ask those who will attend, have attended in person in the past or those who have organized similar events to chime in and help suggest topics to cover in this 90 min session. I must say, Interconnection Track has been a major part if NANOG for many years. We have watched those who we consider as legends to discuss very important topics there. Please try to make your suggestion in order of importance for you as well as from community. I can try to do my best with help of few folks to bring this track back but you can help make it even better so please take a moment and send me your suggestions. Thanks in advance! Mehmet

6 6

Socal Frontier static IP cutover? (+OT fee bitch)
by Paul B. Henson 06 May '17

06 May '17

I was wondering if anybody has been contacted yet about cutting over their static IP addresses for Frontier business FIOS? Last year my understanding was that they were lent Verizon IP space for one year and everyone needed to be cutover by 4/2017; here it is 5/2017 and I've still heard nothing of it. Did they end up permanently aquiring the IP space from Verizon? Did the cutover get pushed back? I tried calling customer service and didn't really get an answer. On another Frontier note, I got a paper bill recently, which was surprising as I've been on paperless billing forever. I noticed this gem on the bill, which is presumably why it warranted being sent physically: "Effective May 15, 2017, a Business High Speed Internet Fee of $1.99 will be added to your bill." Fucking sleezy telcos. It's the only industry that can get away with basically charging you a fee for the privilege of providing you the service you're already paying them for 8-/. Could you imagine going to McDonalds and paying $1.99 for your happy meal + a $.50 Fast Food Service Fee? Getting your house painted for $500 + a $75 Paintbrush Bristle Motion Fee? I called to ask about it and the excuse was that it's to cover "maintainance of the network". Really? So what's the other $130 a month I'm already paying you guys cover? Raise you prices if you got to raise your prices to cover costs, but again, telcos are pretty much the only industry that tags on random extra "fees" to cover basic costs of doing business. Meh, maybe it's time to see how comparatively evil the cable company is nowadays <sigh>. Maybe if that $1.99 a month would actually result in IPv6 this decade...

1 0

SD-WAN for enlightened
by Kasper Adel 06 May '17

06 May '17

Hi, I'm not sure if the buzzword SD-WAN is used to compensate for another buzzword that got over-utilized (SDN) or it is a true 'new and improved' way of doing things that has some innovation into it. I heard different explanation from different vendors: 1) appliances (+ controller) placed in-line to put traffic in tunnels based on policy, with some DPI and traffic tagging...(to do performance/policy based routing) over an expensive link (MPLS) and a cheap one (broadband) with some 'firewall-like' filtering capabilities. 2) same as above, with a flavor of 'machine learning' to find a pattern for traffic to optimize utilization. 3) a controller that instantiates and tears down tunnels from 'classic routers' based on external policies and Network based features to do performance based routing over an expensive link (MPLS) and a cheap one (broadband) with encryption. Is the above a decent high-level summary? Has anyone tried any of these solutions, any general feedback ? Cheers, Kim

4 5

Need recommendation on an affordable internet edge router
by c b 06 May '17

06 May '17

We have a number of internet edge routers across several data centers approaching EOL/EOS, and are budgeting for replacements. Like most enterprises, we have been Cisco-centric in our routing/switching platforms. The ASR1Ks are too small for our needs and the ASR9Ks are prohibitively expensive and probably overkill. That being said, our IT staff is willing to look at other vendors if they are the right fit. Requirements: * Can handle full internet tables, both v4 and v6 with room for reasonable growth over the next 5 years. * VRF capability. * About 12-ish 10Gb ports and 10-ish 1Gb ports (24-ish total if they are 1Gb/10Gb select-rate ports.) * Full-Feature BGP (address-families, communities, peer-groups, etc...) * Used by carriers or large enterprises in a production role for at least a year (and not causing ulcers) * Affordable. I know that's subjective, but we need a solution that is as close as possible to commodity-pricing if this modernization effort balloons to include all of our data centers. We are open to named vendors and even so-called brite-box solutions. A little nervous about fringe solutions like pure whitebox with Quagga, but if the savings are there and people can vouch for it, we will consider it. In other words, if you've used it and stand by it, we value that input and will put it on the initial list. Also, if you chose solution-X after comparing it to solution-Y it would be very helpful to detail what you tested and why you chose. Thanks in advance.

10 14

Ingress filtering from an external cloud service to the internal network
by Torres, Matt 05 May '17

05 May '17

NANOG, We have a hybrid cloud model that includes an external cloud service that needs to reach back into our internal network. The application documentation states that this connection cannot go through a proxy server. I am not in a position to redesign this solution or change the parameters. My question to NANOG is how to manage (filter/secure) the ingress traffic from the external cloud service. Past network guy managed inbound firewall rules based on the cloud-providers source IP address, but this wasn't sustainable and led to multiple outages as the external (source) IP has changed from time to time. I can define the destination ports well enough, but not the source IP addresses. Any ideas on how I can filter this type of inbound traffic from an internet-based service? Thanks Matt

4 7

Question about experiences with BGP remote-AS
by LF OD 05 May '17

05 May '17

We have a number of small routers in co-lo sites that peer with B2B partners. As more of our partners move to cloud, we are considering a consolidation effort and putting all of our peering routers in a cloud exchange site on a single HA pair of routers. Now, each existing B2B peering router uses a unique private ASN to EBGP peer with partners and they, in turn, EBGP peer with our extranet perimeter ASNs for security vetting and other stuff. We looked for a medium-density router (or L3-switch) that can replace multiple small routers (b2b-only, no internet), but we need to retain all of our existing ASNs and peerings. As it turns out, there are many routers that can do VRFs but you cannot put a unique ASN on each VRF so replicating the old environment isn't quite that straightforward. The BGP remote-as looks to be a possible alternative solution, but we've never used it in production and we are unsure of the caveats. Taken at face value, it looks like we can mimic the multi-router/unique-ASN environment we have today on a single platform. However, networking is rarely as smooth as that so I'm asking some of the BGP gurus... what are the pros/cons of doing using remote-as? If anyone here uses it extensively, we could really use some feedback if you run into challenges or hidden surprises that we wouldn't normally think of beforehand. Thanks in advance! LFOD

4 3