Posts

Quick traceroute between Reliance and Neotel (South African subsidiary of Tata Communications)

Using Reliance Mumbai node: 

You requested a Traceroute from Mumbai (62.216.135.130) to neotel.co.za (41.168.1.110) 

   1 ge-0-2-0.0.pjr02.mmb004.flagtel.com (85.95.25.70) 80.252 ms 79.822 ms 79.737 ms  
   MPLS Label=351552 CoS=0 TTL=1 S=1  
   2 so-1-3-0.0.pjr02.hkg005.flagtel.com (85.95.25.118) 79.688 ms 79.717 ms 79.779 ms  
   MPLS Label=301328 CoS=0 TTL=1 S=1  
   3 so-5-0-0.0.cjr04.hkg003.flagtel.com (85.95.25.214) 79.790 ms 81.271 ms 84.868 ms  
   MPLS Label=300272 CoS=0 TTL=1 S=1  
   4 xe-1-2-0.0.cji01.hkg003.flagtel.com (62.216.128.66) 79.504 ms 79.587 ms 79.452 ms  
   5 62.216.145.194 (62.216.145.194) 85.660 ms 79.769 ms *  
   6 if-2-0-0-1123.core3.HK2-HongKong.as6453.net (216.6.95.137) [AS 6453] 79.930 ms 79.884 ms 79.775 ms  
   7 180.87.15.81 (180.87.15.81) [AS 6453] 295.413 ms 295.675 ms 295.695 ms  
   MPLS Label=300800 CoS=0 TTL=1 S=1  
   8 if-5-2.tcore2.CXR-Chennai.as6453.net (180.87.15.70) [AS 6453] 296.944 ms if-2-2.tcore1.SVW-Singapore.as6453.net (180.87.12.1) [AS 6453] 311.951 ms 311.784 ms  
   MPLS Label=299968 CoS=0 TTL=1 S=1  
   9 if-3-3.tcore1.CXR-Chennai.as6453.net (180.87.36.5) [AS 6453] 311.636 ms if-5-2.tcore1.CXR-Chennai.as6453.net (180.87.12.54) [AS 6453] 312.689 ms 312.041 ms  
   MPLS Label=300688 CoS=0 TTL=1 S=1  
  10 if-4-0-0.core1.CFO-Chennai.as6453.net (180.87.36.18) [AS 6453] 144.922 ms if-15-1-0.core1.CFO-Chennai.as6453.net (180.87.36.14) [AS 6453] 145.223 ms if-4-0-0.core1.CFO-Chennai.as6453.net (180.87.36.18) [AS 6453] 144.627 ms  
   MPLS Label=1578 CoS=0 TTL=1 S=1  
  11 116.0.86.1 (116.0.86.1) [AS 6453] 155.796 ms 155.857 ms 154.918 ms  
   MPLS Label=3107 CoS=0 TTL=1 S=1  
  12 116.0.86.30 (116.0.86.30) [AS 6453] 341.019 ms 312.304 ms 322.348 ms  
   MPLS Label=306816 CoS=0 TTL=1 S=1  
  13 if-2-2.tcore1.JSO-Johannesburg.as6453.net (216.6.55.85) [AS 6453] 312.133 ms 311.860 ms 311.718 ms  
  14 216.6.55.38 (216.6.55.38) [AS 6453] 324.450 ms 295.062 ms 296.033 ms  
  15 41.160.0.243 (41.160.0.243) [AS 36937] 312.371 ms 313.074 ms 311.998 ms  
  16 * * *  
  17 * * *  
  18 * * *

So it’s like Mumbai-India-Reliance | Hong Kong-Reliance | Hong Kong - Tata | Chennai -India-Tata | Somewhere in middle-Europe | Johannesburg - Tata

I tried looking for script for database dumps but most of them are way too complex then a simple solution one can implement. 

Thus, I wrote a quick simple 1 page (infact few lines) bash script for database dumps.

Here’s the code:  

#!/bin/bash

#Script for MySQL automated dumps

dbhost=DB-HOSTNAME-HERE  
dbusername=DB-USERNAME-HERE  
dbpassword=DB-PASSWORD-HERE

# Enter database names here  
dbname=( database1 database2 database3 )

for i in "${dbname[@]}"

do  
timestamp=backup_`date +%F.%T`  
output=$i+$timestamp  
mysqldump -h $dbhost -u $dbusername -p$dbpassword -C $i | bzip2 -c > /backups/MySQL-automated/$output  
done

To use this script, simply add names of database inside brackets in front of “dbname” like
dbname=( database1 database2 database3 ) and that’s all. You will get output in form of compressed .sql file.

I usually find that Gmail phone performs lot better then Skype for calls to PSTN. Usually latency on Skype is high as compared to Gmail phone in India (in US it was almost no difference).

I looked around today and here’s bit of test data I collected by calling couple of numbers from Skype & Gmail phone & monitored UDP traffic to collect information about destination servers which is very likely IP-PSTN gateway.

Just now realized that I have over few thousand files in “Latest IP” directory. This is result of poor workaround I created sometimes back to keep track of IP address.

To fix it, I wrote a quick shell script and good enough - it works! :)

Prime purpose of this script is to write latest IP address along with date/time stamp. Thus, one can simply call this script from cron job to keep track of IP address usage.

Yesterday I called MTS Data Card support but their IVRS system was failing in giving me my balance details. Eventually I decided to email their support and glad to say support email was also easily available on their website.

Today I saw acknowledgement mail in spam. No big deal but I usually dig around genuine mails which go in spam to find exact cause. In this case I found mail was sent to me from  customercare.del@mtsindia.in and the server which relayed this mail was:

Today was our Summer Project viva day and overall it went good for me. Although I was suppose to give viva on Sat, but that just didn’t happened as I was kicked out  and insulted from my respective teachers for “not wearing formal clothes”. That’s very  sad. I am not taking it personally but I am just pointing out the level of education and overall mentality in state college education system. I am not saying quality is low or teachers are poor or something like that. I personally feel we have very good teachers overall but it’s just whole system which is screwed up badly and causing problems for just everyone including teachers too.

DNS Lesson time!

Today, I was just looking into website - ideaisp.net

This site is overall pretty interesting as it’s showing various upstream links of IDEA Cellular ISP backbone. Seems like they are just using Cable & Wireless for most of International bandwidth although IDEA ISP is not functional yet. Anyways, this post is not about their geeky upstream peers & routing tables but how bad one can setup DNS! :)

Today, I was reading New Telecom Policy from Dept. of Telecom. Must say I am disappointed. Everyday I hear a new story on 3G & LTE in India. About wireless we all know that due to super limited spectrum, it’s good only smartphones. Hard to call even LTE as an alternate even to DSL. LTE has yet to come, but still it will hardly compete with DSL in tier 3 cities and rural India. For tier 1 cities like Delhi, Chennai, Bangalore and tier 2 cities like Gurgaon, Jaipur - broadband still suffers badly and we all know but just not accepting that wireless broadband is not way out to that. I am not against wireless broadband. I totally agree to fact that for mass deployment wireless is way to go but I strongly feel that another serious effort is needed in wired broadband connectivity. I am happy to get 2Mbps connection via 3G on my Idea cellular phone, and I don’t really complain for it’s cost because of spectrum crunch and all but I feel super surprised on fact that I get 512Kbps capped broadband on DSL when technically it can go over 16Mbps easily. It’s hard to comment on how well fiber connections to Gram Panchayats will perform. All we can say it’s good and nice initiative given they don’t create parallel infrastructure. But why Govt. is missing out demand in big cities where wired infrastructure is “decent” or can be made decent (based on demand)? I don’t see any good efforts being made by Govt. for improving broadband speeds or connectivity by making maximum use of existing copper infrastructure. Working professionals in cities like Gurgaon/Chennai still suffer badly for “decent” broadband while most of them could have given broadband - demand & technology - both of things are there. Just missing willingness on side of Govt. What’s point in FTTH now which “can” give 1Gbps speed given one is ready to pay ~$1500 a month for that sort of speeds?  

Seems like BSNL-NIB (AS9829) routing with Softlayer’s Singapore datacenter (AS36351) is messed up. Example route between BSNL consumer connection (from Haryana, India) to Softlayer hosted site dot19.com

1 router2 (192.168.1.100) [AS8151/AS28513] 3.136 ms 3.874 ms 4.675 ms
2 117.200.48.1 (117.200.48.1) [AS9829] 29.088 ms 32.395 ms 35.773 ms
3 218.248.173.42 (218.248.173.42) [AS9829] 42.411 ms 44.746 ms 47.176 ms
4 218.248.255.70 (218.248.255.70) [AS9829] 92.977 ms 95.322 ms 97.758 ms
5 64.213.76.69 (64.213.76.69) [AS3043] 323.921 ms 324.594 ms 327.478 ms
6 ix-20-0.tcore2.LVW-LosAngeles.as6453.net (209.58.53.9) [AS6453] 333.339 ms 302.673 ms 303.992 ms
7 if-8-2-1-0.tcore1.PDI-PaloAlto.as6453.net (64.86.252.114) [AS6453] 419.641 ms 423.086 ms 424.411 ms
8 if-2-2.tcore2.PDI-PaloAlto.as6453.net (66.198.127.2) [AS6453] 425.475 ms 426.956 ms 430.239 ms
9 if-9-2.tcore1.TV2-Tokyo.as6453.net (180.87.180.18) [*] 436.026 ms 433.594 ms 434.911 ms
10 180.87.180.50 (180.87.180.50) [*] 439.543 ms 438.330 ms 441.001 ms
11 ae7.bbr01.eq01.tok01.networklayer.com (50.97.18.162) [AS36351] 471.290 ms 446.401 ms 472.233 ms
12 ae1.bbr01.eq01.sng02.networklayer.com (50.97.18.165) [AS36351] 477.914 ms 480.721 ms 485.605 ms
13 ae5.dar01.sr03.sng01.networklayer.com (50.97.18.197) [AS36351] 499.273 ms 502.262 ms 501.499 ms
14 po1.fcr01.sr03.sng01.networklayer.com (174.133.118.131) [AS36351] 489.251 ms po2.fcr01.sr03.sng01.networklayer.com (174.133.118.133) [AS36351] 496.491 ms 497.322 ms
15 * * *
16 * * *
17 * * *
18 * * *

Messed up routing tables of BSNL or lack of co-ordination between Tata Comm & Softlayer?

Today I was showing a good friend - how root servers are working and how they are connected. I was explaining him anycasting and gave example of k-root node which is hosted in Delhi by niXi i.e National Internet Exchange of India. I was totally stunned to find traceroute result to anycasting based block of k-root-node hosted in Delhi from my BSNL connection. Checkout traceroute here.

It seems like BSNL has gone totally wild in listening to announcements for 193.0.14.0/23 via ASN 25152 which is of  RIPE-NCC-K-ROOT nodes.