fredrik's thread

fredrikjj

proof read this please: IPsec Virtual Tunnel Interface | Fredrik's Networking Blog

I feel like I probably haven't done as much research as I should have.

fredrikjj

Wrote a pretty long post on mtu and stuff Fragmentation, MTU, MSS Clamping, Tunnels | Fredrik's Networking Blog

I feel like it should be mostly correct, but any input is appreciated.

lrb

I read your last two posts while eating my dinner and they were really good! Based on a point in your VTI post I have updated one of my evernote notes to reflect something I had written by mistake Keep up the good work mate.

fredrikjj

lrb wrote: »

I read your last two posts while eating my dinner and they were really good! Based on a point in your VTI post I have updated one of my evernote notes to reflect something I had written by mistake Keep up the good work mate.

That's great. Was there anything you disagreed with?

fredrikjj

wrote about phase 1 and phase 2 dmvpn. packet captured phase 2 dmvpn to figure out how the request/reply process works.
mGRE And NHRP – Static, Phase 1, Phase 2 | Fredrik's Networking Blog

fredrikjj

spent the night researching phase 3 dmvpn. hopefully it's correct, but it's not exactly well documented.

DMVPN Phase 3 Spoke-to-Spoke Tunnel Discovery | Fredrik's Networking Blog

Tomorrow I finally get to configure the IPsec and DMVPN stuff in INE's workbook! Exciting new, isn't it.

tomtom1

You're making some nice progress there! If you don't mind me asking, how long have you worked with the Cisco gear to build up some experience paired with the IE. You're unemployed now, but with your numbers in hand you'll surely have some better chances in a working environment.

fredrikjj

tomtom1 wrote: »

You're making some nice progress there! If you don't mind me asking, how long have you worked with the Cisco gear to build up some experience paired with the IE. You're unemployed now, but with your numbers in hand you'll surely have some better chances in a working environment.

I started studying for the CCNA in early 2013 and took the exam in May I think. That was the first time I touched Cisco stuff, but before then I had worked with Extreme gear for like two years in a very junior role. Also, I'm not really trying to get a ccie to get a job because that would be seriously dumb considering the passing rate on this thing. I'd like to think that I'm smart than that... barely.

powmia

fredrikjj wrote: »

I started studying for the CCNA in early 2013 and took the exam in May I think. That was the first time I touched Cisco stuff, but before then I had worked with Extreme gear for like two years in a very junior role. Also, I'm not really trying to get a ccie to get a job because that would be seriously dumb considering the passing rate on this thing. I'd like to think that I'm smart than that... barely.

But if you do pursue the CCIE in hopes of getting a job out of it, there is a slight possibility that you have some moments of clarity in which you step back and actually think about how you would apply the knowledge from this curriculum in the real world. If by some chance you do pass the CCIE, but all you have gained is the muscle memory of rapidly configuring the technologies from the blueprint... then there is a slight possibility that you will be a very useless CCIE.

Basically, I'm only trying to motivate and say that if you are going to dedicate yourself to something... don't settle for anything less than the outcome you desire.... because, come on... nobody goes after the CCIE hoping that they get a CCNP level job.

Even if you never pass, the proper perspective on the material you are learning may provide you with a more valuable understanding than that of some people that have passed.

fredrikjj

powmia wrote: »

But if you do pursue the CCIE in hopes of getting a job out of it, there is a slight possibility that you have some moments of clarity in which you step back and actually think about how you would apply the knowledge from this curriculum in the real world. If by some chance you do pass the CCIE, but all you have gained is the muscle memory of rapidly configuring the technologies from the blueprint... then there is a slight possibility that you will be a very useless CCIE.

Basically, I'm only trying to motivate and say that if you are going to dedicate yourself to something... don't settle for anything less than the outcome you desire.... because, come on... nobody goes after the CCIE hoping that they get a CCNP level job.

Even if you never pass, the proper perspective on the material you are learning may provide you with a more valuable understanding than that of some people that have passed.

I'm not sure I get exactly what you are trying to say, but I do try to understand what I'm doing and not just configure stuff. That seems like a no-brainer. Of course, I'm going to have certain blind spots due to lack of real world experience, but on the other hand, that could lead to me being more careful and not making so many assumptions.

powmia

I'm saying, don't sell yourself short.

gorebrush

Why stop at one CCIE? I'm not going stop at 1

fredrikjj

I started studying multicast today. I get the general idea and I read the chapter in Comer a few months ago, but I'm not exactly going into it with a lot of prior knowledge. Given that, it'll be a while until I'll even know what topics I want to blog about. Once I'm good at multicast it should be pretty straight forward to do the CCNP:SP Adv. Route exam since it's mainly about BGP and multicast if I remember correctly, but I haven't decided if I want to take that detour. I probably do, but my current epic workstation with 4 GB of RAM is not going to run many instances of XRv so that's something I would have to address in that case.

Why stop at one CCIE? I'm not going stop at 1

I have faith in you.

fredrikjj

Reading Developing IP Multicast Networks Vol1.

PIM General Rule 1
PIM General Rule 2
PIM General Rule 3
PIM General Rule 4
PIM General Rule 5
PIM General Rule 6

PIM-DM Rule 1
PIM-DM Rule 2
PIM-DM Rule 3

PIM-SM Rule 1
PIM-SM Rule 2
PIM-SM Rule 3
PIM-SM Rule 4
PIM-SM Rule 5
PIM-SM Rule 6
PIM-SM Rule 7
PIM-SM Rule 8

Proxy-Join Timer Rule 1
Proxy-Join Timer Rule 2
Proxy-Join Timer Rule 3


You guys know these by heart, yes?

lrb

Honestly I did not like that book. I found I couldn't stay awake through some of it and much preferred this one.

http://www.amazon.com/Interdomain-Multicast-Routing-Practical-Solutions/dp/0201746123

Might be a good supplement for you

JustFred

Fred,

There are tons of networking jobs in Sweden and Norway (don't go to Denmark lol) do whatever you can to get a job. Everything you are learning could end up being wasted. Get the job man. Sweden is just like Holland, you can afford to stay home for a while, but don't let it get the best of you. Interview interview interview

fredrikjj

lrb wrote: »

Honestly I did not like that book. I found I couldn't stay awake through some of it and much preferred this one.

Interdomain Multicast Routing: Practical Juniper Networks and Cisco Systems Solutions: Practical Juniper Networks and Cisco Systems Solutions: 9780201746129: Computer Science Books @ Amazon.com

Might be a good supplement for you

I don't find Williamson's book nearly as boring as you did and I'm already 75% through it, including notes. Perhaps it's because I'm still in my multicast honeymoon period I might check out that other book as well at some point since I seem to have a high tolerance for networking books.

fredrikjj

JustFred wrote: »

Fred,

There are tons of networking jobs in Sweden and Norway (don't go to Denmark lol) do whatever you can to get a job. Everything you are learning could end up being wasted. Get the job man. Sweden is just like Holland, you can afford to stay home for a while, but don't let it get the best of you. Interview interview interview

I usually don't like discussing these things because it quickly becomes very speculative and whiny, but since you bring it up. You've made the kind of assumptions that are so easy to make when you are an insider in a particular industry; that there are a lot of jobs and that you just have to interview for them. It's common to use human dating as an analogy for the job market, and you're essentially the cool guy who can't understand why someone can't get a girlfriend - just ask a girl out, dummy. I'm sure that you mean well though, and I'm not actually offended

Generally speaking, to get a job you need to pass some initial screening where your credentials are checked against other people's credentials, with the winners being interviewed. At this point they don't care about your knowledge of the OSPF LSDB or MPLS VPNs or any other piece of technology for that matter, only that you have degree X, worked at job Y and have certificate Z. I'm obviously being too reductionist here to make a point, but if you don't pass the initial screening, it's much harder to get a job. My resume is simply worse than many other resumes out there applying for similar jobs. I also don't live where 90% of the jobs are, and in Sweden that's a major barrier to entry because it's (probably correctly) assumed that it will be very difficult for you to move due to the severe shortage of rental apartments.

That said, if you or anyone else think that people are making a mistake by not hiring me, I'd be willing to move anywhere and work for you.

JustFred

Hehehehehe point taken. I still believe you can find a job, so keep looking and be aggressive about it. Are there any work shops or networking events you could attend? It could lead to opportunities

fredrikjj

I haven't been posting here or written any blog post lately but that's because I've been studying multicast. I've read Developing IP Multicast Networks Vol 1 and I'm about to finish the multicast section in INE's workbook. I've also found Denise Fishburne's Cisco Live presentation on multicast troubleshooting helpful. Hopefully I'll have time to get back to the blogging once I'm done with this initial work because it's something I enjoy quite a bit.

fredrikjj

I like forums more than blogs since it's much easier to have a two way conversation, but I feel like a blog is taken more seriously by other people, and it's easier for an outside observer to get an overall feel for where you're at in your studies since they don't have to sift through a bunch of forum posts. That said, I don't want to just abandon this thread and make it another ccie burn out thread, because I'm not burned out. I took a break from reading IPv6 Fundamentals to write this post because I thought to myself "wow, I can't believe I'm this excited about reading a book about IPv6, it's almost silly". I probably feel that way because I have some prior knowledge on IPv6 and the book is at just the right level of difficulty (not very hard).

I think that this is underestimated sometimes, reading books in the correct order. It seems like you could just pick up the hardest books right away and as long as you try hard enough, you could get through them, and learn something. You probably could, and it would speed up your learning process, but it's much harder. You risk giving up, and it seems like there's a strong survivor bias going on here, with people only looking at what the brightest people are doing. They forget about all the people who bought a bunch of advanced books, but never finished them.

Like most people I probably underestimated the scope of the CCIE RS. If you already have strong fundamentals and just need to read the two volume cert guide to get up to speed on things that you haven't worked on for a while, you are in a completely different place than me. I feel like I need to catch up on 20 years of networking, and mostly by reading a ton of books. That I'm already more than 50% done with the INE workbook, and probably could get some basic connectivity going in a full scale lab, feels like a false metric to go by.

My actual progress is probably more like 20%, but that feels fine. The next step for me is the CCNP SP exams due to their overlap with CCIE RS, and the fact that I do want to work for an ISP once I graduate. The reading and labbing I've done on MPLS and BGP and L3VPNs is probably my favorite stuff so for. It seems like many (most?) people who actually pass CCIE RS or CCIE SP have either that one or the predecessor, CCIP. In a way it's a test run because CCNP SP also requires you to pick your own literature and to follow an exam topics list. If I can't pass that one, I'm not CCIE calibre anyway and there's no point in going for it. It also allows me to focus on the topics that some people think are the harder RS topics like multicast, bgp, qos, mpls, etc.

Anyway, I'm going back to my book.

gorebrush

Yeah I regret not doing CCIP - I was going to do it but didn't bother/too busy/other excuses here.

creamy_stew

Hey Fredrik,

Good progress as always.

I just have to ask:
*Should you get hired, in what position+tier do you expect to get hired?

*How long would you expect to be in your initial position before unicorns and rainbows engulf your being and carry you away to the position you deserve?

fredrikjj

creamy_stew wrote: »

Hey Fredrik,

Good progress as always.

I just have to ask:
*Should you get hired, in what position+tier do you expect to get hired?

If someone hires me, I would expect to be hired in the lowest tier, and I don't have a problem with that. My skill set is fairly limited even if I know certain things fairly well. Besides, it's probably seen as a liability to know too much in an entry level position anyway. Hypothetically, if I pass CCNP SP I don't expect it to make me more marketable since I wouldn't need to have advanced BGP/MPLS knowledge to perform my duties in an entry level position. It may even be detrimental since I could be seen as some kind of loose cannon and/or brain **** cheater. I hope that someone will care, but I don't expect it.

*How long would you expect to be in your initial position before unicorns and rainbows engulf your being and carry you away to the position you deserve?

I know you are being facetious, but I don't really feel that kind of entitlement and I have no idea how things like that work in the real world.

fredrikjj

I've almost finished IPv6 Fundamentals. Some chapters were easy and some were a bit harder, but most of it was actually things I had seen before which felt good. One exception is NAT64, the final chapter, which will be brand new to me tomorrow. I've taken some notes, but not more than around 30 handwritten pages; it's not a dense book and many sections I just read and moved on. In particular it has 3 chapters (of 10 in total) on basic IPv6 configuration, routing and IPv6 routing protocols that shouldn't be news to anyone with a CCNP.

I "get" the multipoint IPv6 over IPv4 tunnels in a way that I didn't really do when I briefly looked at them during CCNP Route, probably as a result of looking at DMVPN a few months ago. It's interesting to see patterns in protocols and how they solve roughly the same problem in different ways. ARP solves the same problem that embedding IPv4 addresses in the IPv6 prefix in a 6to4 tunnel does. I also amused myself by turning ISATAP into a ghetto static DMVPN with OSPFv3 running multipoint non-broadcast; probably not the intended use case.

There's a new book coming out that I will probably pick up. It's called IP Routing on Cisco IOS, IOS XE, and IOS XR: An Essential Guide to Understanding and Implementing IP Routing Protocols and it could force everyone to update their reading lists if it's actually good. As someone who thinks he will attempt CCNP:SP it looks almost to good to be true because the description says:

"Content was created in alignment with Learning@Cisco, responding to more efficient self-study content for the Cisco Career Certifications within CCNA/CCNP/CCIE Service Provider, Cisco IOS XR Specialist Certification, and the routing components of several other Cisco Certifications (such as CCNA/CCNP/CCIE Routing and Switching)."

lrb

I'm hoping that book can be sort of an updated TCP/IP Volume 1. Not that these protocols have changed hugely over the years but their use cases have and there are things like flowspec and LISP that might be included in the book.

fredrikjj

Yeah that would be great since Routing TCP/IP vol 1 and 2 has way too much legacy stuff at this point.

PS.
I finished IPv6 Fundamentals today. I conceptually understood the section on NAT64, but failed to implement it in the topology I had running (yeah I just have routers running.. nerd). The configuration details of this feature aren't terrible important to me right now so I'm moving on to the next thing I have planned which is writing a few blog posts on IPv6 based on the CCNP SP Adv Route blueprint. First up is DHCPv6.

fredrikjj

nevermind. formatting completely broke and I can't rewrite the post atm.

fredrikjj

I wrote a post on DHCPv6. Half of it is technical information from IPv6 Fundamentals and the other half is me pontificating on why stateful DHCPv6 is kind of dumb and that you might as well run stateless. If something sounds wrong and you point that out, I won't get mad.

https://fredrikjj.wordpress.com/2014/12/04/dhcp-in-ipv6/

There's a great debate on stateless vs. stateful and I might read up on that and write my next post on what the arguments are for each method. Or I'll just move on to looking at the various IPv6 over IPv4 tunnels. I'll probably do that because I'm not super interested in DHCP.

fredrikjj

I'm working on IPv6 tunneling and I must say that I really enjoy this topic. Whenever multiple layers are introduced it suddenly becomes more interesting. Hacked together BGP over 6to4 yesterday because I just assumed that it would work even though IPv6 Fundamentals said that you couldn't do dynamic routing protocols (the IGPs wont work though). A very enjoyable lab experience I must say, but perhaps not super practical, I don't know. But basically, if you use BGP over 6to4 you gain a lot of freedom in how you address each IPv6 island since it each prefix you advertise will get the 6to4 prefix as BGP next hop. Every destination advertised will therefore automatically resolve to the 6to4 tunnel.

I'm now reading a 6RD white paper because they only covered 6to4 and isatap in the book. The impression I get so far is that 6rd is supposed to be a more general version of 6to4 that supports public global unicast addresses. But, I also think that you could use BGP with 6to4 to achieve the same goal.

PS.
I'm also having a minor issue with XRv. It seems like I can't ping link local ipv6 addresses on a gre over ipv4 tunnel. I can ping the global unicast. This works by default in IOS and I'm not sure if it's some kind of bug in XRv or if I need to enable something. Do you guys know what's going on there?