Networking

Few days back a friend of mine (who works for an ISP) congratulated me for joining HE. Along with wishes he told me that our bgp.he.net doesn’t works well and the reason he fealt so is because he couldn’t see all peers for his ASN in our tool.

This is not a problem and to be more broader - same applies on all popular tools other then bgp.he.net like RIPE Stats, Robtex AS analysis etc. The reason many of these tools do not and cannot show all peers is because they show what they see from the point of collection. E.g right now I am on BSNL (AS9829).

Just was looking around at EDNS support by Google. To find how it supports and how packet looks like I created a test NS records for dnstest.anuragbhatia.com pointing to one of test server (178.238.225.247). I wasn’t running any DNS server on the server. Just ran quick tcpdump.  

At server end:

sudo tcpdump 'port 53 and dst 178.238.225.247' -nn -vvv -w sample.pcap

Then I forcefully triggered DNS queries via Google’s recursor using:**

Few months ago I moved away from Google Chrome to Opera Mobile on my Android device. Google Chrome is pretty loaded and overall slow.   Recently I noticed browsing was pretty slow. I noticed that “Off-Road mode” was enabled.  

I disabled it and performance was much better. I did heard of it in past and clearly it’s a proxy mode where packets between Opera instance running on cell phone and destination server are routed via an Opera server which uses some special compression technologies and helps in making browsing faster. Carrying with my obsession for looking at ASNs and IP address, I enabled it again and visited bgp.he.net and was surprised to see the result.

I have been to quite a few countries but I must say Japan just stands out in internet connectivity. Overall connectivity is just amazing out here. As I landed on airport in Fukuoka, I noticed open free wifi (just one signup online form to accept TOS and it was up), later I noticed Fukuoka City Wifi project and it’s really visible across streets and very much works. As

I got to hotel, I was given SSID for wifi and it was just up! No crazy proxy, no crazy use of hotel room numbers/last name combinations. I was getting 20Mbps speed on wifi. This was a clear sign that transit was not bottleneck and likely wifi/end point connectivity was the one which was putting it on to 20Mbps (802.11n on a good quality router with 5Ghz). As I connected my laptop on wired LAN, I noticed (which I did expected by now) - connection synced at 100Mbps LAN and that was pretty much internet speed I was getting.

Just thought to put a quick blog post. Right now I am sitting at Shanghai airport (China) and looking at local Internet routing table of this region. As well know - routing between India & China very much sucks and most of packets go via US.

Here’s a trace from China to India to one of servers I manage:

Anurags-MacBook-Pro:~ anurag$ mtr -wrc 1 lamp1
HOST: Anurags-MacBook-Pro.local Loss% Snt Last Avg Best Wrst StDev
1.|– 172.21.183.254 0.0% 1 1.7 1.7 1.7 1.7 0.0
2.|– 192.168.88.2 0.0% 1 4.2 4.2 4.2 4.2 0.0
3.|– 180.168.176.97 0.0% 1 3.4 3.4 3.4 3.4 0.0
4.|– 124.74.100.181 0.0% 1 3.4 3.4 3.4 3.4 0.0
5.|– 218.1.2.69 0.0% 1 7.4 7.4 7.4 7.4 0.0
6.|– 124.74.215.65 0.0% 1 4.2 4.2 4.2 4.2 0.0
7.|– 61.152.86.198 0.0% 1 7.4 7.4 7.4 7.4 0.0
8.|– ??? 100.0 1 0.0 0.0 0.0 0.0 0.0
9.|– 202.97.35.26 0.0% 1 8.5 8.5 8.5 8.5 0.0
10.|– 202.97.52.30 0.0% 1 226.8 226.8 226.8 226.8 0.0
11.|– ldngw1.arcor-ip.net 0.0% 1 209.2 209.2 209.2 209.2 0.0
12.|– 145.253.33.238 0.0% 1 208.8 208.8 208.8 208.8 0.0
13.|– ??? 100.0 1 0.0 0.0 0.0 0.0 0.0
14.|– 182.19.15.1 0.0% 1 358.9 358.9 358.9 358.9 0.0
15.|– 203.122.61.148.reverse.spectranet.in 0.0% 1 362.1 362.1 362.1 362.1 0.0
16.|– ??? 100.0 1 0.0 0.0 0.0 0.0 0.0
17.|– 203.92.40.206.reverse.spectranet.in 0.0% 1 378.6 378.6 378.6 378.6 0.0
Anurags-MacBook-Pro:~ anurag$

And return path (India > China)

Like last three APRICOTs, I would be attending APRICOT 2015 this time. It’s in Japan and it would be fun as always to meet networking community. :)

Buzz me if any of readers here are attending. Would be great to meet & greet!

Yesterday I was driving and radio was pretty boring. Next, I connected cell phone to car’s stereo (I use a PT-750 to wirelessly connected my devices to car’s audio system). Next I tuned into Gaana.com app and experience was overall good. The way whole setup was working itself is a wonder - wireless profiles keeping layer 3 link (IP address of device) consistent and handovers happening on layer 1. On top of that a while world of backbone routing across AS9498 backbone the hosting provider’s network of the app. Now an interesting thing in this setup was the IP allocations. I that IP allocated by Airtel was 100.92.215.253.

And my post on 1st April. Don’t take it as April fool post ;)

Multiple times NIXI’s AS24029 has been reported as acting like transit ASN for multiple networks. I have analysed it in past and this is very much because of route leaks by few specific networks. I have explained difference in peering Vs transit routes and their handling previously on my blog.

In short: A network is supposed to re-announce it’s peering and transit routes only to customer and not to any other peer or upstream. Whenever NIXI’s ASN appears in global routing table, its always the case where one or more networks are re-announcing routes learnt via NIXI to their upstream transits. 

And I am back from Malaysia after attending APRICOT 2014. It was a slightly slow event this time as less people came up due to change of location from Thailand to Malaysia. But I kind of enjoy the APRICOT in start of year. :)

It has been quite sometime when I blogged. After getting into Spectranet I got relatively more busy along with bit of travelling to Delhi NCR which has been taking lot of time. I wish to blog more over time. 

I thought about this long back - “Who pays to whom in case of internet bandwidth?” I have been working in this domain from sometime and so far I have learnt that it’s really complex. I will try to put a series of blog post to give some thoughts on this subject. Firstly we have to understand that when we talk about “bandwidth price” it’s often layer 3 bandwidth which you buy in form of capacity over ethernet GigE, Ten-GigE and so on (or STMs if you are in India). As we know from back school class in networking - layer 3 works over layer 2 and so to deliver “bandwidth” on layer 3, one needs layer 2 physical circuit. Price paid by companies on layer 2 Vs layer 3 varies significantly based on their location, type of business, their target goal etc. E.g a content heavy company like Google pays hell lot of money on layer 2 circuits while it is strongly believed among networking community that Google is a tier 1 network and hence a “transit free” zone and they do not pay any amount on layer 3. In general the trend is pretty much as big networks have larger network footprint and connected “PoPs” over layer 2 (leading to a higher layer 2 bill) while relatively lower layer 3 bill while small networks depend significantly just on transit bandwidth (in form of layer3) and have very low layer 2 footprint.