• Startup employees don’t make as much cash as I thought it would be. And that shouldn’t be the persp­ective.
• An exit or acquisition does not mean happiness. You have to enjoy the journey. I spent time with a founder with a ‘successful’ exit and he wasn’t as happy as I thought he’d be.
• Most companies are ‘sold’ and not ‘bought’.
• Aquihires are starting to become a non-prestige term.
• Many try to exit the scene with big bank balances. Much like the Hollywood and California gold rush.
• People are obsessed with failing fast.
• Ultra local services like Uber, Lyft, Instacart, ZeroCater, Chewse, etc. are raising more cash than world domination disruptive ideas.
• Series A crunch is REAL. But Angel investing continues. Contrary to Sam Altman’s upside risk, I met a person doing $ 5000*200. And that makes everyone an Angel investor. JOBS act will help.
• Palo Alto/University Ave is no more the place to be. You have to be in San Franciso. Caltrain/BART rides are perceived too long.
• South Bay is getting more expensive. Rentals at $2500 for 2 bedrooms in Mountain View and Palo Alto. East PA and Fremont continue to be cheap.
• HYPE is now a subject of discussion. MailBox as an example.

Do not look at this with judgmental eyes. These are only observances.

The post The little things I learned from the valley – April 2013 edition appeared first on gigpeppers.

This to me is the best part of building a company. The serendipitous hit of the most unlikely scenarios adding up as experiences and owning up to them. Nothing beats this.

Last Friday evening was rather amusing.

Two new Cucumbertown users Courtney & Samantha pinged us and complained their recipes images were upside down.We probed. The images were indeed upside down.

What was more surprising was that the users said the images looked fine on their laptops.
I pinged Courtney & Samantha to send us the original images.

Arun inspected it and took a look at the exif data. Both of them came from iPhones.

A quick google search and cnet had the answer.

tl;dr version:
“Before iOS 5, the only way to snap a photo with your iPhone was to tap the onscreen shutter button–always a somewhat awkward maneuver. But with iOS 5, Apple transformed the volume-up button into a shutter release, thereby making iPhone photography feel a bit more natural.
Just one problem: when you flip your phone around so the volume-up button is facing, well, up, you end up with upside-down photos and videos.”

There goes!

With a bit of inspection we saw hope. Iphone was preserving the orientation in the Exif data. This is good news. At least we can take a shot at attempting to rotate.

Cucumbertown is a Python shop and we use the PIL library to do image manipulation.

Investigating more and as usual somebody has already tried to figure out a solution on stackoverflow.

Pretty simple and the solution went out in the next 10 mins.

Except for a minor bug We forgot that images could uploaded be without exif data. There goes the fix and to production.

Its pretty interesting how a minor change in technology can have a tremendous impact on something that relies on it.

Code Credits: Arun

The post My images are all upside down! appeared first on gigpeppers.

Naturally we have quite a few alerts if it hits upper thresholds of 3 seconds plus.

Couple of days before Chris Zacharias wrote about how page weight matters and how YouTube dealt with this. So when a mails pop up in your inbox with this alert from Google Analytics with over 20 seconds in page load, you stop everything to find what’s happening.

Usually a delay in page load is often picked up immediately after a production push mostly by random testing or from our highly engaged users. But this didn’t happen and the alert came a day later.

That freaks you out. Unknown problems for which you haven’t yet figured out root cause IMO is greater danger than a big bug where you know the issue.

We started probing and that’s when we saw results like this.

And of course Google Analytics was averaging the time and the average was skewing the results.

Correlating this with other results and the jig saw started fell into place. Cucumbertown was picked up by a food channel in Nigeria and a prominent blogger in Thailand and this was bringing in the crowds. But the page load in these countries as you can see is ridiculously low.

Cucumbertown is an asset heavy website and there is a significant cost associated in loading the basic scripts even thought we delay everything through requireJS and dynamically load JavaScript based on need. Even the basic DOM load is taking time.

Corroborate this with the load time in the US at 2.5 seconds, the speed of light and latency at 43ms for DSL devices across the globe and it’s time to start thinking about CDN’ing assets.

At Zynga we had initially relied on Akamai but later switched to LimeLight and they look like a prime candidate to rely on.

But the recent activity by CloudFlare on HackerNews and the umbrella features they offer seemed tempting to explore. And I took a dip in the water to test CloudFlare.

Right now this blog is served via CloudFlare off a free plan that includes CDNing the assets. The experience is good when there is a rush of requests but if the site is not accessed consistently then it seems like the cache gets evicted. The page load is worse when there is a first time hit after inactivity. But then the follow up requests page loads at 1.5-2 seconds.

I have always thought of CDN’s as an enterprisy expensive word. But here we are; a startup serving the globe and this solution looks imperative.

Do you have any advice? How’s your experience been?

The post Page speed really does matter appeared first on gigpeppers.

• The community was showing positive emotions, constructive criticism. HN community’s remarks and constructive criticism are pearls of wisdom.
• I did performance engineering at Zynga for a while. This shouldn’t have happened.

It’s somewhat interesting to understand what happened underneath and how we went on to fix it.
Here’s how the drama unfolded.
At about 8:30 AM PST I posted the news to HN and waited for someone to up vote. A bump and since it was a “Show HN”, it started moving fast and pretty soon was on its way to the front page.

The connections started increasing and things looked pretty good.

The traffic coming in brought with it people who loved food & had so much positive emotions.
Love was in the air.

Juggling between the intense stream of answering mails from new members & questions on HN I found myself singing Backstreet Boys for an hour…

Until one of our users Trey and an HN user shot me an IM.

We started probing the graphs and it looked like the world was coming to an end.

User United857 struck the nail in the coffin with this message and we were officially down by 11:30 PM IST. It was 404 hell.

The worst part, ssh did not work anymore. All we could wait was to get the terminal buffers to allow us to type a character at a time.
Eventually we got the monit alerts.
Continuous 404’s ensured we went out of the front page and pretty soon down the drain.

Even though HN traffic is not our direct market, posting on HN, taking feedback and learning has always been my dream. And it was…
Without considering that we were pretty down & drained already and that it’s nearly 1 AM in Bangalore we decided to get to the root of this.

What became of this investigation is a lesson in carelessness. One I hope I’ll remember for sometime.
So back to the train of events…
Once the domino effect started all we could do was to wait and watch httpd & java processes fail via monit mails.
I’ll save the detailed investigation (its too verbose) and go directly to inference.
Digging through the trails later we saw that
Free memory had decreased, IO wait increased drastically & immediately after that Solr hung. Apache piled the requests and stopped responding. This led Nginx to return 404’s. This continued for 20-30 minutes as a black box.
Eventually Solr crashed releasing some memory. This helped us restart the services. But the damage was done by then.
Digging through logs one by one and eventually in dmesg the culprit showed up

So the root cause analysis narrowed down to two points. Why did free memory come down so low and why did IOWait go up so high.
Luckily for us everything was graphed and soon enough we saw that people came in directly and started searching heavily. Solr usage went very high and it started evicting. We did not anticipate this. Pretty bad!
To make matters worse the high usage with memory consumption sent the OS to swap. Even then this should have been handled. The dmesg OOM led us to the details and that’s when we realized Linode offers only a max of 512 MB swap by default. We had gone with it and since memory + swap overflowed pretty quickly the system was OOMing.

Lessons:

• Mockup high traffic scenarios & test (Although its unlikely it would have hit us to test search so much)
• Run Solr separately from your web nodes. It’s an all, in memory process (We are a lean startup and sometimes running process together is a call we have to take. Making it all separate machines is too costly)
• Run sufficient swap.

But between failures we noticed a pattern that shouldn’t have happened.
Apache requests were in sync with Nginx.

Cucumbertown is a site that’s heavily cached. Anonymous requests are all http cached at the load balancer level & only the logged in ones hit the backend web.

Digging further and we started seeing this (below). For every Nginx hit there is equivalent apache hit. But this was only for the homepage.

Since the homepage is query heavy this had a heavy impact on the underlying infrastructure. In fact you could very well conclude that the homepage hits were the reason we went down (The explore items that you see on homepage – they are all in fact a manifestation of Solr searches)

This was surprising. We had our pages cached as follows.

location = / {
 proxy_pass http://localhost:82;
 proxy_cache cache;
 proxy_cache_key $cache_key; 
 proxy_cache_valid 200 302 2m;
 proxy_cache_use_stale updating;
 }

Things should have been fine. The show recipe pages were.
A curl request exposed the blunder.

Cherian-Mac: cherianthomas$ curl -I http://www.cucumbertown.com/
HTTP/1.1 200 OK
Server: ngx_openresty
Content-Type: text/html; charset=utf-8
Connection: keep-alive
Keep-Alive: timeout=60
Vary: Accept-Encoding
Vary: Cookie
Set-Cookie: csrftoken=Dl4mvy4Rky7sfZwqek27hFrCXzWCi9As; expires=Fri, 18-Oct-2013 02:15:32
 GMT; Max-Age=31449600; Path=/
X-Cache-Status: EXPIRED

I’ll leave it to the reader why X-Cache-Status always returned expired.

And now if you are like us running a machine with very little swap, create a secondary one with sufficient space and a lower priority. This will decrease your chances for OOM drastically.
Here’s how to do it.

cd /
sudo dd if=/dev/zero of=swapfile bs=1024 count=2097152 
sudo mkswap swapfile

Change count accordingly
Add this to the line lower than the swap file line in /etc/fstab

/swapfile swap swap defaults 0 0
sudo swapon -a 
swapon -s

Additionally add the following two lines to your /etc/sysctl.conf

vm.panic_on_oom=1 kernel.panic=15 
vm.panic_on_oom=1 line enables panic on OOM;

The kernel.panic=15 line tells the kernel to reboot fifteen seconds after panicking.
In the worst-case scenario you can reboot than be hung for a longer duration

Graphs courtesy DataDog.

The post Post mortem of a failed HackerNews launch appeared first on gigpeppers.

The post Barcamp Flash Tech Talk appeared first on gigpeppers.

Another concern was that most of the the ncover tutorials on the web showed how to test a single unit test assembly even though ncover comes with a nunitproject nant task to pick out multiple test files(which is the highway case).

Without much further ado, here’s what I did

Download version 1.5.8 of NCover, v 1.4 of NCover Explorer and v 1.4 of NCover Extras

NCover Extras encapsulates the functionalities provided by NCover as a set of nant tasks. This needs to be loaded.Also CoverLib.dll that comes with NCover needs to be registered.

<loadtasks assembly="${ncover.nant.task.dir}NCoverExplorer.NAntTasks.dll"/>

<exec program="regsvr32" workingdir="${ncover.dir}" commandline="/s coverlib.dll"/>

Add these lines before the target is called.

Now for the actual task…

<target name="coverage">

  <nunitproject project="${output.dir}tests.nunit">

    <fileset basedir="${output.dir}">

        <include name="*.*Test*.dll" />

      </fileset>

  </nunitproject>

  <ncover

    program="${ncover.dir}NCover.Console.exe"

    commandLineExe="${nunitconsoleexe.dir}nunit-console-x86.exe"

    logLevel="Verbose"

    commandLineArgs="${output.dir}tests.nunit /nodots /nologo"

    verbose="true"

    logFile="${coverage.reports.dir}/coverage.log"

    workingDirectory="${output.dir}"

    coverageFile="${coverage.reports.dir}Unit.Test.Coverage.xml">

    <assemblies basedir="${output.dir}">

      <include name="CompanyName*.dll" />

      <exclude name="*.*Test*.dll" />

    </assemblies>

  </ncover>

  <ncoverexplorer

    program="${ncoverexplorer.dir}NCoverExplorer.Console.exe"

    projectName="ProjectName"

    reportType="ModuleClassFunctionSummary"

    outputDir="${coverage.reports.dir}"

    xmlReportName="TestCoverage.xml"

    sort="FunctionCoverageDescending"

    htmlReportName="TestCoverage.html"

    showExcluded="True"

    minimumCoverage="80"

    satisfactoryCoverage="80" >

    <fileset>

      <include name="${coverage.reports.dir}Unit.Test.Coverage.xml" />

    </fileset>

    <exclusions>

      <exclusion type="Assembly" pattern="*.Tests" />

      <exclusion type="Namespace" pattern="*.Tests*" />

    </exclusions>

  </ncoverexplorer>

</target>

The node with the nunitproject picks all the test dll’s from the output directory and builds a tests.nunit project file. This will be the input to Nunit which NCover will profile.

Digging into the NCover node you’ll see how NCover is being instructed to take pick the assemblies. If you are copying the code verbatim then change the assemblies node to pick your project dll’s and avoid test dll’s(By convention test assemblies are named ProjectName.Tests.dll). The output of this profiling task will be an xml file(Unit.Test.Coverage.xml in this case) which will be the input to NCoverExplorer, to get those fancy html outputs.

Here’s a typical NCover output xml..

The next node ncoverexplorer takes the xml files and builds a html output. Something similar to what’s shown below. Additionally the attribute failMinimum(not shown in this case) will fail the build if coverage doesn’t meet the coverage percentage specified.

Here’s a sample output.

One little quirk I encountered here was using nunit-console.exe instead of nunit-console-x86.exe. My 64 bit Server 2008 build machine rejected the first executable outright since it ran as 64 bit.

What pissed me of in this exercise was the time it took to integrate a seemingly simple open source project. Pretty much the same experience I had dealing with DD-WRT firmware sources. Once the author/company decides to to release a higher version of the project as a paid version or to make money out of the project, they then obfuscate source paths or put in obstacles that deter consumers from downloading those open source versions.

In case of the DD-WRT the source was hidden inside GB’s of data and in this case the official NCover website removed all the forums and added nag screens asking users to signup(given below) before downloading..

The post Integrating NCover 1.5 with Nant builds appeared first on gigpeppers.

The post Targeted Ad’s appeared first on gigpeppers.

You can’t expect me to write blogengine plugins for Disqus, super cache, cl forms, Akismet etc. Its way too much work from my focus area.

Being a lot of guys who understand technology terms like scripting and html I chose to do a source control install rather than those “one click installs”. Primary reason for this is that WordPress comes with a lot of security issues(they being the most popular) and upgrading through svn is a breeze.

Here’s how I did it.

Choose a good hosting provider. Mine’s DreamHost. They will give you ssh access apart from the usual ftp and cpanel stuff.

Create a folder in your root to host your blog (use software like putty to get into your remote hosting machine).

Assuming that your hosting provider has svn installed (which most of them do), run svn co http://svn.automattic.com/wordpress/tags/2.6.1 . (last dot included). Then go through the standard installation procedures.

Next time you need to upgrade do a svn up (as always backup everything) and you are done.

Update: This post has some missing sentences at the top.

The post Migrating to WordPress appeared first on gigpeppers.

buildings

Originally uploaded by alainlm

A picture like this can take you a long way..

The post Buildings appeared first on gigpeppers.

enum Status {

  Not_Available,

  Available_For_Game,

  Available_For_Discussion

}

The string conversion here for the first one would yield Not_Available. This is not desirable in some cases where you don’t want the “_”. A better implementation would be to create attributes to this enum and then extract this.

  public enum Status {

    [Status(Description = "Not Available")]

    Not_Available = 1,

    [Status(Description = "Available For Game")]

    Available_For_Game = 2,

    [Status(Description = "Available For Discussion")]

    Available_For_Discussion = 3,

  }

  public class StatusEnumInfo {

    private static StatusAttribute[] edesc;

    public static String GetDescription(object e)

    {

      System.Reflection.FieldInfo f = e.GetType().GetField(e.ToString());

      StatusEnumInfo.edesc = f.GetCustomAttributes(typeof(StatusAttribute), false) as StatusAttribute[];

      if (StatusEnumInfo.edesc != null && StatusEnumInfo.edesc.Length == 1)

        return StatusEnumInfo.edesc[0].Description;

      else

        return String.Empty;

    }

    public static object GetEnumFromDesc(Type t, string desc)

    {

      Array x = Enum.GetValues(t);

      foreach (object o in x) {

        if (GetDescription(o).Equals(desc)) {

          return o;

        }

      } return String.Empty;

    }

  }

  public class StatusAttribute : Attribute {

    public String Description { get; set; }

  }

  public class Implemenation {

    public void Run()

    {

      Status statusEnum = (Status)StatusEnumInfo.GetEnumFromDesc(typeof(Status), "Not Available");

      String statusString = StatusEnumInfo.GetDescription(Status.Available_For_Discussion);

    }

  }

You are of course free to generalize this. Correct me if this is not a correct implementation.

The post Enum with attributes appeared first on gigpeppers.