Blog Posts in September 2014
During one of my ExpertExpress engagements I got an interesting question: “could we replace a pair of central firewalls with iptables on the Linux server?”
Short answer: Maybe (depending on your security policy), but I’d still love to see some baseline scrubbing before the traffic hits the server – after all, if someone pwns your server, he’ll quickly turn off iptables.
During the last round of polishing of my Designing Infrastructure for Private Clouds Interop New York session (also available in webinar format) I wondered whether one could use the recently-launched UCS Mini to build my sample private cloud.
I love reading blog posts on Plexxi blog (you SHOULD add them to your RSS reader) and the “It’s the Application, Stupid” series from Mat Matthews is no exception. What pleasantly surprised me was that a large enterprise came to the same conclusions I’m preaching for the last few years.
I hope you know TCP provides a reliable stream service not reliable packet delivery, but you might not have realized all the implications – I found an old post by Robert Graham explaining how things really work and how you can use them to bypass quick-and-dirty IDS that rely on signatures instead of doing proper protocol decodes.
Jeremy Schulman was the driving force behind the Puppet agent that Juniper implemented on some Junos switches (one of the first fully supported Puppet-on-a-switch implementations). In the meantime, he quit Juniper and started his own company focused on a network automation product – more than enough reasons to chat with him on Software Gone Wild.
I’m running or participating in five workshops or sessions during next week’s Interop New York. Three of them build on each other, so you might want to attend all of them in sequence:
Designing Infrastructure for Private Clouds starts with requirements gathering phase and focuses on physical infrastructure design decisions covering compute, storage, physical and virtual networking, and network services. If you plan to build a private (or a reasonable small public) cloud, start here.
- Why is everyone claiming that the network is so slow to change?
- Is that really the case? Why?
- Why is the manual configuration culture so widespread in networking?
- How does the holistic thinking in the design phase dissolve into the box mentality of CLI commands?
- How does the box mentality limit the scalability of network deployments?
How would one interface with external Internet in this scenario? I totally get the virtual network assets mantra, but even a virtual BGP router would need to get a physical interconnect one way or another.
As always, there are plenty of solutions depending on your security needs.
Are you lucky enough to be one of the 87% of North American enterprises that plan to have SDN in production by 2016 or one of the 53% of the companies that plan to have SDN deployed in the near future? Even though we all know how inflated these claims are, you might have to start considering the deployment aspects of a solution a $vendor will persuade your CIO to buy.
If you’ve been reading my blog in the last few months, you might have noticed that I started a new podcast focused on software-defined everything (hence the name: Software Gone Wild – thanks to Jason Edelman).
The latest episodes are always available on this page; you can also subscribe to the podcast feed in RSS, Atom or iTunes format… and if you wonder why we need yet-another podcast, read the About Software Gone Wild document.
If you mention open-source cloud orchestration tools these days, everyone immediately thinks about OpenStack (including the people who spent months or years trying to make it ready for production use). In the meantime, there are at least two other comparable open-source products (CloudStack and Eucalyptus) that nobody talks about. Obviously having a working product is not as sexy as having 50+ vendors and analysts producing press releases.
A while ago Cisco added dynamic FCoE support to Nexus 5000 switches. It sounded interesting and I wanted to talk about it in my Data Center Fabrics update session, but I couldn’t find any documentation at that time.
In the meantime, the Configuring Dynamic FCoE Using FabricPath configuration guide appeared on Cisco’s web site and J Metz wrote a lengthly blog post explaining how it all works, triggering a severe attack of déjà vu.
When we were discussing my autumn travel plans, my lovely wife asked me “What are you going to talk about in Bern?” She has a technical background, but I didn’t feel like going into the intricacies of SDN, SDDC and NetOps, so I told her the essence of my keynote speech:
After the initial onslaught of SDN washing, four distinct approaches to SDN have started to emerge, from centralized control plane architectures to smart reuse of existing protocols.
As always, each approach has its benefits and drawbacks, and there’s no universally best solution. You just got four more (somewhat immature) tools in your toolbox. And now for the details.
I spent last Tuesday in Bern attending the SIGS DC Day Event, and came back home extremely pleasantly surprised. The conference was nice and cozy, giving everyone plenty of opportunities to chat about data center technical challenges (thanks for all the wonderful conversations we had – you know who you are!).
Having the opportunity to meet fellow networking engineers and compare notes is great, but it’s even better to combine that with new knowledge, and that’s where the event really excelled.
Seamus Gilchrist sent me a fantastic list of MPLS- and MPLS-TE-related questions. Instead of starting an email exchange we agreed on something that should benefit a wider community: a lengthy whiteboard session discussing the basics of MPLS, MPLS-TE, load balancing and QoS in MPLS networks…
The first part of our conversation is already online: The Essence of MPLS.
A while ago Rick Parker told me about his amazing project: he started a meetup group that will build a reference private/hybrid cloud heavily relying on virtualized network services, and publish all documentation related to their effort, from high-level architecture to device and software configurations, and wiring plans.
In Episode 8 of Software Gone Wild Rick told us more about his project, and we simply couldn’t avoid a long list of topics including:
A few days ago Garrett Wollman published his exasperating experience running IPv6 on large L2 subnets with Juniper Ex4200 switches, concluding that “… much in IPv6 design and implementation has been botched by protocol designers and vendors …” (some of us would forcefully agree) making IPv6 “…simply unsafe to run on a production network…”
The resulting debate on Hacker News is quite interesting (and Andrew Yourtchenko is trying hard to keep it close to facts) and definitely worth reading… but is ND/MLD really as broken as some people claim it is?
VMware announced several vMotion enhancements in vSphere 6, ranging from “finally” to “interesting”.
vMotion across virtual switches. Finally. The tricks you had to use previous were absolutely bizarre.
Some OpenFlow-focused startups are desperately trying to tell you how redundant their architecture is. Unfortunately all the whitepapers (and the prancing unicorns) cannot change a simple fact: an SDN controller (OpenFlow-based or otherwise) is in some aspects a single failure domain.
One of my readers sent me an interesting challenge: they’re deploying a new DMVPN WAN, and as they cannot expect all locations to have native (non-NAT) IPv4 access, they plan to build the new DMVPN over IPv6. He was wondering whether it would work.
Apart from “you’re definitely going in the right direction” all I could tell him was “looking at the documentation I couldn’t see why it wouldn’t work” Has anyone deployed DMVPN over IPv6 in a production network? Any hiccups? Please share your experience in the comments. Thank you!
TL;DR: I'll be in Bern on September 9th. If you'd like to drop by and discuss network design or automation challenges, read on…
The latest release of Cisco Nexus 1000V for vSphere can handle twice as many vSphere hosts as the previous one (250 instead of 128). Cisco probably did a lot of code polishing to improve Nexus 1000V scalability, but I’m positive most of the improvement comes from interesting architectural changes.
The pilot episode of Software Gone Wild podcast featuring Snabb Switch created plenty of additional queries (and thousands of downloads) – it was obviously time for another deep dive episode discussing the intricate innards of this interesting virtual switch.
If you’re a regular reader of my blog, you know that I spent a lot of time during the last three years debunking SDN myths, explaining the limitations of OpenFlow and pointing out other technologies one could use to program the network.
During the summer of 2014 I organized my SDN- and OpenFlow-related blog posts into a digital book. I want to make this information as useful and as widely distributed as possible – for a limited time you can download the PDF free of charge.
A while ago I wrote about the idea of treating network infrastructure (and all other infrastructure) as code, and using the same processes application developers are using to write, test and deploy code to design and implement networks.
That approach clearly works well if you can virtualize (and clone ad infinitum) everything. We can virtualize appliances or even routers, but installed equipment and high-speed physical infrastructure remain somewhat resistant to that idea. We need a different paradigm, and the best analogy I could come up with is a database.