<?xml version="1.0" encoding="UTF-8"?>
<rss version="2.0" xmlns:atom="http://www.w3.org/2005/Atom">
  <channel>
    <title>Blogs — Chintamani Gawade</title>
    <link>https://3dwag98.github.io/blog/</link>
    <description>Blogs on building software at scale.</description>
    <language>en</language>
    <atom:link href="https://3dwag98.github.io/feed.xml" rel="self" type="application/rss+xml"/>
    <item>
      <title>Seven hours of GitHub — why fixing the cause did not end the outage</title>
      <link>https://3dwag98.github.io/blog/post.html?p=github-outage-august-2026</link>
      <guid isPermaLink="true">https://3dwag98.github.io/blog/post.html?p=github-outage-august-2026</guid>
      <pubDate>Tue, 25 Aug 2026 09:00:00 GMT</pubDate>
      <description>On 17 August 2026 GitHub was degraded for 7h 47m. One misconfigured autoscaling rule started it; retry logic — GitHub's own, plus a latent bug in VS Code — kept it going for four and a half hours after the cause was fixed. Every component and term explained from the ground up, then set against why Git is unusually hard to host at all.</description>
      <category>incident-review</category>
      <category>distributed-systems</category>
      <category>kubernetes</category>
      <category>git</category>
    </item>
    <item>
      <title>Built, shipped, still running</title>
      <link>https://3dwag98.github.io/blog/post.html?p=four-banks</link>
      <guid isPermaLink="true">https://3dwag98.github.io/blog/post.html?p=four-banks</guid>
      <pubDate>Wed, 19 Aug 2026 09:00:00 GMT</pubDate>
      <description>The portfolio gives each role one line. This is the detail behind those lines — seven years across four banks, what the systems actually were, and what each one taught me.</description>
      <category>platform-engineering</category>
      <category>full-stack</category>
      <category>spring-boot</category>
      <category>kubernetes</category>
    </item>
    <item>
      <title>Where a slow page actually comes from</title>
      <link>https://3dwag98.github.io/blog/post.html?p=tracing-a-slow-page</link>
      <guid isPermaLink="true">https://3dwag98.github.io/blog/post.html?p=tracing-a-slow-page</guid>
      <pubDate>Fri, 14 Aug 2026 09:00:00 GMT</pubDate>
      <description>A page that takes four seconds is rarely slow for the reason the first person to look at it thinks. How to find the real one, from the browser down to the query plan.</description>
      <category>full-stack</category>
      <category>performance</category>
      <category>postgres</category>
      <category>observability</category>
    </item>
    <item>
      <title>Designing for failure — failure modes, degradation and what resilience costs</title>
      <link>https://3dwag98.github.io/blog/post.html?p=designing-for-failure</link>
      <guid isPermaLink="true">https://3dwag98.github.io/blog/post.html?p=designing-for-failure</guid>
      <pubDate>Sun, 02 Aug 2026 09:00:00 GMT</pubDate>
      <description>Failure is not one thing, and neither is the answer to it. A working method for deciding what a service does when the truth is unavailable — worked through the authorization hot path, and honest about what each mechanism costs.</description>
      <category>distributed-systems</category>
      <category>resilience</category>
      <category>redis</category>
      <category>spring-boot</category>
    </item>
    <item>
      <title>Four papers that explain caching</title>
      <link>https://3dwag98.github.io/blog/post.html?p=caching-papers</link>
      <guid isPermaLink="true">https://3dwag98.github.io/blog/post.html?p=caching-papers</guid>
      <pubDate>Sun, 12 Jul 2026 09:00:00 GMT</pubDate>
      <description>What ARC, TinyLFU, Facebook's memcache paper and S3-FIFO actually say — the mechanisms drawn out, and why each one exists.</description>
      <category>caching</category>
      <category>system-design</category>
      <category>distributed-systems</category>
    </item>
  </channel>
</rss>
