Enough documentation is available about this on the internet, but for some reason I find that they are slightly fragmented and none of them give an end-to-end view. So here is an order in which you can go through these steps smoothly –
1) Follow the steps mentioned on this page – https://pages.github.com/ — This will activate your <github_username>.github.io page
2) For your custom domain to show/point to this URL, now follow the steps mentioned here – https://help.github.com/articles/adding-a-cname-file-to-your-repository/ — While configuring the DNS settings, pay special attention to whether your custom domain URL is going to be a TLD(Top Level Domain) URL or a sub domain URL. You need to only one of the two steps mentioned in that article, i.e. you setup an A record, or a CNAME record. Unless you have some special use-case, do not setup both(I cant imagine when that can make sense). A friend of mine ended up setting up both records and got a 400 Nginx Bad Request when attempting to access his website URL on the browser!
In this context, this Stack Overflow reply – http://stackoverflow.com/a/22374542/183902 – helped in troubleshooting the CNAME and A record mix up for my friend.
3) This should get you up and running with a Hello World website. Now install jekyll from Ruby Gems and structure and style your pages! If you are looking for some quick setup, or are better at editing CSS than writing from scratch(like me :D), then you can take a look at these two themes that are good starting points – https://github.com/poole/hyde and https://github.com/poole/lanyon . Just fork and clone these repos and use the files for your site, with or without customizations.
There your Custom Domain website, powered by Github Pages is ready!
Thanks to my friend’s attempt to setup his site, I ended up reviving http://vatsala.co.in from the dead. Only a landing page there right now, need to bring myself to set up the full site.
So far in my work life, I have been more of a generalist who knows a little of a lot of things, but none in great depth.
I happened to chance upon the careers page of a promising startup and realised I could probably never apply to any of the roles mentioned there due to the above fact. The prospect of becoming ‘out of market’ freaks me out in a way.
I am taking this 4 week course about the fundamentals for R programming via Coursera
Here is a summary of what I learnt as part of Week 1
- The origins of R
- The basic units of data that one can deal with in R
- Reading From CSV files and working with the data
- Conditional filtering of thus imported data and basic mathematical aggregation such as mean, max,min etc
So far the Quiz and the assignment has been not too tough, not too easy at the same time.
Even though there is enough documentation about this topic on the internet, often the topic of resolving merge conflicts becomes a white elephant that everyone talks about but does not how to do well. So here is a video that explains exactly how it is done —
I came to know about this via FB, shared by good friend Rajmahendra Hegde – https://www.linkedin.com/in/rajmahendra
On Saturday, I attended vodQA 2014 – an annual event run by Thoughtworks. vodQA is an event focussed around testing and quality assurance targetted towards professionals in the software testing field.
There were multiple tracks in the event, sharing significant insights below from the ones I that I attended –
- Responsive web testing
- The talk was about the Speakers’ experience with using http://galenframework.com/ – a framework that enables one to do automated testing of the look and feel of your responsive websites
- Based on Selenium, the tests written with this framework will help you do Purely UI and view level acceptance testing.
- While there are language specific libraries (Java,JS), knowledge of any of these languages is not necessary to get started with writing tests using this framework as the basic syntax of the spec files are based on Markdown
- Tests written with this framework are not to replace the functional tests otherwise written with Selenium, instead to do cross-form factor, cross-browser rendering of UI elements of your responsive web applications
- All tests generate clean HTML reports that can be observed later
- Crowd Sourced Testing – Apparently, there are websites and web applications where the testing of web applications can be crowd sourced where, as an organisation, you get a better variety of people(with diverse perspectives) to test, use your application and give feedback, and as a tester, an opportunity to widen your horizons in terms of capability of testing multiple application. These sites come with multiple bells and whistles to safeguard the interests of all the stakeholders involved in a transaction. One such service – https://www.passbrains.com/
- Common Tools used for automating load testing – I did a dipstick survey of what tools are most commonly used for load testing – LoadRunner, JMeter, Apache Bench
- How to break your application using Gatlling
- This was a short demonstration of Gatling – an opensource tool based on Scala,Akka and Netty used to perform load testing
- Top advantages observed –
- Simple DSL to specify the scenarios
- Capability to do both Spike load testing, Endurance(Sustained load) testing
- Other tools in the category are Apache Bench, Siege, Locust, and JMeter
- While the possibilities are endless, the metrics to really care about while testing one’s application are response times to the 90th, 95th and 99th percentile
- This tool generates clean HTML reports and graphs from the tests run
While this blog has been suffering neglect, I have been experimenting with a bunch of things in the last few months. Sharing what I learnt today – When we modify the last_value of a table sequence in postgres, the changes do not reflect immediately, i.e. the next value of that column(which is populated by the sequence) does not start with the new value right away.
To quote the documentation –
“ALTER SEQUENCE will not immediately affect
nextval results in backends, other than the current one, that have preallocated (cached) sequence values. They will use up all cached values prior to noticing the changed sequence generation parameters. The current backend will be affected immediately.” – Notes section of http://www.postgresql.org/docs/9.2/static/sql-altersequence.html
If you have the later versions of sqlite, then using the command line client sqlite3 to open a .db file is easy, and is available on the 1st page of a web search, which is that you get into the commandline with sqlite3 command and then type “.open </path/to/filename>” and connect to a specific exisiting database.
But in version 3.6.20 – when you do a .help – you will notice that there is no .open command. And when you type “.open </path/to/filename>” , sqlite3 will politely respond with the message – ‘Error: unknown command or invalid arguments: “open”. Enter “.help” for help’
The solution is to pass the .db file name as a command line argument – i.e. you do a ‘sqlite3 </path/to/filename.db>’ to connect to an existing database.
This is a version specific post. Other environment details – CentOS 6.4 (final) and SQLite was the version bundled with Python 2.6.6 .