<?xml version="1.0" encoding="UTF-8"?>
<rss version="2.0" xmlns:atom="http://www.w3.org/2005/Atom" xmlns:dc="http://purl.org/dc/elements/1.1/">
  <channel>
    <title>DEV Community: Sandun Siwantha</title>
    <description>The latest articles on DEV Community by Sandun Siwantha (@sandun_siwantha).</description>
    <link>https://dev.to/sandun_siwantha</link>
    <image>
      <url>https://media2.dev.to/dynamic/image/width=90,height=90,fit=cover,gravity=auto,format=auto/https:%2F%2Fdev-to-uploads.s3.amazonaws.com%2Fuploads%2Fuser%2Fprofile_image%2F3850448%2F506e6aa7-4e6b-4643-9699-77d13b41db8f.jpg</url>
      <title>DEV Community: Sandun Siwantha</title>
      <link>https://dev.to/sandun_siwantha</link>
    </image>
    <atom:link rel="self" type="application/rss+xml" href="https://dev.to/feed/sandun_siwantha"/>
    <language>en</language>
    <item>
      <title>What I’m Building: IoT Pipelines, Webhook Retries, and Real-World Debugging</title>
      <dc:creator>Sandun Siwantha</dc:creator>
      <pubDate>Mon, 30 Mar 2026 04:30:11 +0000</pubDate>
      <link>https://dev.to/sandun_siwantha/what-im-building-iot-pipelines-webhook-retries-and-real-world-debugging-1h7h</link>
      <guid>https://dev.to/sandun_siwantha/what-im-building-iot-pipelines-webhook-retries-and-real-world-debugging-1h7h</guid>
      <description>&lt;p&gt;Lately I’ve been working on a few things:&lt;/p&gt;

&lt;ul&gt;
&lt;li&gt;IoT telemetry pipelines&lt;/li&gt;
&lt;li&gt;Webhook retry systems (handling failed deliveries)&lt;/li&gt;
&lt;li&gt;Debugging real-world data flow issues from devices&lt;/li&gt;
&lt;/ul&gt;

&lt;p&gt;Stack:&lt;/p&gt;

&lt;ul&gt;
&lt;li&gt;Linux + Docker&lt;/li&gt;
&lt;li&gt;MQTT&lt;/li&gt;
&lt;li&gt;Custom backend services&lt;/li&gt;
&lt;/ul&gt;

&lt;p&gt;Biggest lesson so far:&lt;br&gt;
Most systems fail not because of logic, but because of &lt;strong&gt;network and reliability issues&lt;/strong&gt;.&lt;/p&gt;

&lt;p&gt;Currently exploring:&lt;/p&gt;

&lt;ul&gt;
&lt;li&gt;AI-assisted coding workflows&lt;/li&gt;
&lt;li&gt;Running local models for dev productivity&lt;/li&gt;
&lt;/ul&gt;

&lt;p&gt;If you’re working on similar systems, let’s connect.&lt;/p&gt;

</description>
      <category>ai</category>
      <category>architecture</category>
      <category>backend</category>
      <category>iot</category>
    </item>
    <item>
      <title>Building a Reliable IoT Telemetry Pipeline (Lessons from Real Systems)</title>
      <dc:creator>Sandun Siwantha</dc:creator>
      <pubDate>Mon, 30 Mar 2026 04:29:13 +0000</pubDate>
      <link>https://dev.to/sandun_siwantha/building-a-reliable-iot-telemetry-pipeline-lessons-from-real-systems-20lf</link>
      <guid>https://dev.to/sandun_siwantha/building-a-reliable-iot-telemetry-pipeline-lessons-from-real-systems-20lf</guid>
      <description>&lt;h3&gt;
  
  
  🚀 Context
&lt;/h3&gt;

&lt;p&gt;Most IoT tutorials stop at “send sensor data to the cloud.”&lt;br&gt;
In reality, things break — networks drop, endpoints fail, and data gets lost.&lt;/p&gt;

&lt;p&gt;I’ve been working on real-world IoT pipelines, and here’s a practical breakdown of what actually matters.&lt;/p&gt;




&lt;h3&gt;
  
  
  🧱 Core Architecture
&lt;/h3&gt;

&lt;p&gt;Typical flow:&lt;/p&gt;

&lt;p&gt;Device → Gateway → Message Broker → Processing → Storage/API&lt;/p&gt;

&lt;p&gt;Key components I used:&lt;/p&gt;

&lt;ul&gt;
&lt;li&gt;MQTT for device communication&lt;/li&gt;
&lt;li&gt;Dockerized services for isolation&lt;/li&gt;
&lt;li&gt;Linux-based gateways&lt;/li&gt;
&lt;/ul&gt;




&lt;h3&gt;
  
  
  ⚠️ Real Problems You Will Hit
&lt;/h3&gt;

&lt;ul&gt;
&lt;li&gt;Intermittent connectivity (especially in edge environments)&lt;/li&gt;
&lt;li&gt;Data loss during API failures&lt;/li&gt;
&lt;li&gt;Duplicate events&lt;/li&gt;
&lt;li&gt;Backpressure when ingestion spikes&lt;/li&gt;
&lt;/ul&gt;




&lt;h3&gt;
  
  
  🛠️ What Actually Works
&lt;/h3&gt;

&lt;h4&gt;
  
  
  1. Retry Mechanism (Critical)
&lt;/h4&gt;

&lt;p&gt;Implement webhook/API retry with:&lt;/p&gt;

&lt;ul&gt;
&lt;li&gt;exponential backoff&lt;/li&gt;
&lt;li&gt;retry queues&lt;/li&gt;
&lt;li&gt;dead-letter handling&lt;/li&gt;
&lt;/ul&gt;

&lt;h4&gt;
  
  
  2. Local Buffering
&lt;/h4&gt;

&lt;p&gt;Gateways should:&lt;/p&gt;

&lt;ul&gt;
&lt;li&gt;store data locally (disk/queue)&lt;/li&gt;
&lt;li&gt;flush when connection restores&lt;/li&gt;
&lt;/ul&gt;

&lt;h4&gt;
  
  
  3. Idempotency
&lt;/h4&gt;

&lt;p&gt;Design APIs to safely handle duplicates:&lt;/p&gt;

&lt;ul&gt;
&lt;li&gt;use unique event IDs&lt;/li&gt;
&lt;li&gt;avoid side-effect duplication&lt;/li&gt;
&lt;/ul&gt;

&lt;h4&gt;
  
  
  4. Observability
&lt;/h4&gt;

&lt;p&gt;Add:&lt;/p&gt;

&lt;ul&gt;
&lt;li&gt;structured logs&lt;/li&gt;
&lt;li&gt;basic metrics (success/fail rates)&lt;/li&gt;
&lt;li&gt;alerting hooks&lt;/li&gt;
&lt;/ul&gt;




&lt;h3&gt;
  
  
  🧪 Example: Simple Retry Logic (Pseudo)
&lt;/h3&gt;



&lt;div class="highlight js-code-highlight"&gt;
&lt;pre class="highlight shell"&gt;&lt;code&gt;&lt;span class="k"&gt;if &lt;/span&gt;request_failed:
  retry_count++
  &lt;span class="nb"&gt;wait&lt;/span&gt; &lt;span class="o"&gt;=&lt;/span&gt; base_delay &lt;span class="k"&gt;*&lt;/span&gt; 2^retry_count
  requeue&lt;span class="o"&gt;(&lt;/span&gt;event, &lt;span class="nb"&gt;wait&lt;/span&gt;&lt;span class="o"&gt;)&lt;/span&gt;
&lt;/code&gt;&lt;/pre&gt;

&lt;/div&gt;






&lt;h3&gt;
  
  
  🧠 Key Takeaway
&lt;/h3&gt;

&lt;p&gt;IoT is not about sending data.&lt;br&gt;
It’s about &lt;strong&gt;guaranteeing delivery under failure&lt;/strong&gt;.&lt;/p&gt;




&lt;p&gt;If you’re building something similar, I’m happy to exchange ideas 👍&lt;/p&gt;

</description>
      <category>iot</category>
      <category>docker</category>
      <category>linux</category>
      <category>api</category>
    </item>
  </channel>
</rss>
