<?xml version="1.0" encoding="UTF-8"?><rss xmlns:dc="http://purl.org/dc/elements/1.1/" xmlns:content="http://purl.org/rss/1.0/modules/content/" xmlns:atom="http://www.w3.org/2005/Atom" version="2.0" xmlns:cc="http://cyber.law.harvard.edu/rss/creativeCommonsRssModule.html">
    <channel>
        <title><![CDATA[Stories by Vadigireddygariswetha on Medium]]></title>
        <description><![CDATA[Stories by Vadigireddygariswetha on Medium]]></description>
        <link>https://medium.com/@vadigireddygariswetha?source=rss-a37bccd54c8e------2</link>
        <image>
            <url>https://cdn-images-1.medium.com/fit/c/150/150/1*rF_cnge_kOIqB6Z0VqVOjg.jpeg</url>
            <title>Stories by Vadigireddygariswetha on Medium</title>
            <link>https://medium.com/@vadigireddygariswetha?source=rss-a37bccd54c8e------2</link>
        </image>
        <generator>Medium</generator>
        <lastBuildDate>Thu, 19 Nov 2026 11:11:05 GMT</lastBuildDate>
        <atom:link href="https://medium.com/@vadigireddygariswetha/feed" rel="self" type="application/rss+xml"/>
        <webMaster><![CDATA[yourfriends@medium.com]]></webMaster>
        <atom:link href="http://medium.superfeedr.com/" rel="hub"/>
        <item>
            <title><![CDATA[Understanding Rollback in MuleSoft: Why Your Database Updates Might Not “Undo” Themselves]]></title>
            <link>https://medium.com/@vadigireddygariswetha/understanding-rollback-in-mulesoft-why-your-database-updates-might-not-undo-themselves-a1968621d9ba?source=rss-a37bccd54c8e------2</link>
            <guid isPermaLink="false">https://medium.com/p/a1968621d9ba</guid>
            <category><![CDATA[programming]]></category>
            <category><![CDATA[api-development]]></category>
            <category><![CDATA[mulesoft]]></category>
            <category><![CDATA[integration]]></category>
            <category><![CDATA[softwarengineering]]></category>
            <dc:creator><![CDATA[Vadigireddygariswetha]]></dc:creator>
            <pubDate>Wed, 08 Jul 2026 07:36:21 GMT</pubDate>
            <atom:updated>2026-07-08T07:36:21.670Z</atom:updated>
            <content:encoded><![CDATA[<figure><img alt="" src="https://cdn-images-1.medium.com/max/787/1*G65GKeK5rOGXeZSxTGxyeA.png" /></figure><p><strong>A Scenario That Trips Up Every MuleSoft Developer</strong></p><p>Picture this integration flow. It looks simple on paper:</p><ul><li>Update the ORDERS table → mark an order as PROCESSING</li><li>Insert all line items into ORDER_ITEMS</li><li>Generate a payment transaction reference number</li><li>Call an external Payment Gateway API to charge the customer</li><li>Update ORDER_ITEMS with the payment transaction reference</li></ul><p>Now ask yourself: <strong>if step 5 fails, do steps 1 and 2 automatically get undone?</strong></p><p>If your gut answer is “yes, because it’s all one flow”- that’s exactly the misconception this article exists to fix. The honest answer is <strong>no, not unless you explicitly design it that way </strong>and understanding why is one of the most important lessons in building reliable MuleSoft integrations.</p><p><strong>What “Rollback” Actually Means</strong></p><p>In genric if something goes wrong while saving data, the database should “undo” everything it tries to save like pressing Ctrl+Z so we don’t end up with half-saved, broken data.</p><p>Now think of a database transaction like editing a Word document with Track Changes on you can type paragraph after paragraph, but nothing is “final” until you click <strong>Save</strong>. If your laptop crashes halfway through, Word can discard the unsaved changes and reopen the last saved version clean, with no half-written sentence stuck in the middle.</p><p>A database transaction works the same way:</p><ul><li><strong>Commit</strong> = Save. The changes become permanent.</li><li><strong>Rollback</strong> = Ctrl+Z. Everything since the last commit is discarded, as if it never happened.</li></ul><p>The entire point of a transaction is to guarantee <strong>“all or nothing.”</strong> Either every step in the transaction succeeds and is saved together, or if any step fails, all of them are undone together. There’s no in-between state where some rows are saved and others aren’t.</p><p>That’s the theory. MuleSoft doesn’t give you this behavior automatically, you have to ask for it.</p><h3>The Default Behavior in Mule: Auto-Commit, Step by Step</h3><p>Here’s the part that surprises a lot of developers: <strong>by default, every Database connector operation in Mule commits immediately, on its own, the moment it finishes.</strong> There is no invisible “flow-level transaction” wrapping your whole process unless you build one yourself.</p><p>So walking back through the scenario:</p><ul><li><strong>Step 1</strong> (Update ORDERS to “PROCESSING”) → runs, commits instantly</li><li><strong>Step 2</strong> (Insert into ORDER_ITEMS) → runs, commits instantly</li><li><strong>Step 3</strong> (Generate payment reference) → just in-memory logic, nothing to commit</li><li><strong>Step 4</strong> (Call Payment Gateway API) → customer gets charged, and that can’t be silently taken back</li><li><strong>Step 5</strong> (Update ORDER_ITEMS with the payment reference) → fails</li></ul><p>By the time step 5 even runs, steps 1 and 2 are already permanently saved in the database. There’s nothing left to “undo,” because they were never part of a shared transaction with step 5 in the first place. Each Database connector call is its own isolated unit of work.</p><p>So the honest answer to the original question is: <strong>no, steps 1 and 2 will not be rolled back when step 5 fails</strong>, unless you explicitly wrap steps 1, 2, and 5 inside the same transaction. And even then, there’s a catch involving the Payment Gateway that we’ll get to next.</p><h3>How to Actually Group Steps Into One Transaction</h3><p>To make multiple database operations succeed or fail together, you wrap them in a <strong>Try Scope</strong> configured to be transactional, and mark each Database operation inside it to join that transaction.</p><pre>&lt;try transactionalAction=&quot;ALWAYS_BEGIN&quot;&gt;<br><br>    &lt;db:update config-ref=&quot;Database_Config&quot; transactionalAction=&quot;ALWAYS_JOIN&quot;&gt;<br>        &lt;db:sql&gt;<br>            UPDATE ORDERS<br>            SET STATUS = &#39;PROCESSING&#39;<br>            WHERE ORDER_ID = :orderId<br>        &lt;/db:sql&gt;<br>        &lt;db:input-parameters&gt;&lt;![CDATA[#[{orderId: vars.orderId}]]]&gt;&lt;/db:input-parameters&gt;<br>    &lt;/db:update&gt;<br><br>    &lt;db:bulk-insert config-ref=&quot;Database_Config&quot; transactionalAction=&quot;ALWAYS_JOIN&quot;&gt;<br>        &lt;db:sql&gt;<br>            INSERT INTO ORDER_ITEMS (ORDER_ID, PRODUCT_SKU, QUANTITY, PRICE)<br>            VALUES (:orderId, :productSku, :quantity, :price)<br>        &lt;/db:sql&gt;<br>        &lt;db:input-parameters&gt;&lt;![CDATA[#[payload]]]&gt;&lt;/db:input-parameters&gt;<br>    &lt;/db:bulk-insert&gt;<br><br>    &lt;error-handler&gt;<br>        &lt;on-error-propagate type=&quot;ANY&quot;&gt;<br>            &lt;logger level=&quot;ERROR&quot;<br>                    message=&quot;Transaction rolled back for order #[vars.orderId]: #[error.description]&quot; /&gt;<br>        &lt;/on-error-propagate&gt;<br>    &lt;/error-handler&gt;<br><br>&lt;/try&gt;</pre><p>What changed here:</p><ul><li>transactionalAction=&quot;ALWAYS_BEGIN&quot; on the &lt;try&gt; scope tells Mule to start a new local transaction</li><li>transactionalAction=&quot;ALWAYS_JOIN&quot; on each Database operation tells Mule not to commit on its own, and to join the transaction that’s already open</li><li>If the insert fails, Mule automatically rolls back the update too, because they share the same underlying database connection</li><li>Only when the &lt;try&gt; block finishes successfully does Mule actually commit</li></ul><p>Steps 3, 4, and 5 sit outside this &lt;try&gt; block on purpose. That’s the key insight of this whole article.</p><p>One detail trips people up here: transactionalAction=&quot;ALWAYS_BEGIN&quot; on the &lt;try&gt; scope only works if the underlying Database Config is actually set up to support a transaction. In the connector&#39;s Advanced settings (or on the connection element in XML), the <strong>Transaction Type</strong> needs to be set to LOCAL — not NONE. LOCAL tells the driver &quot;these statements can share a single connection and be committed or rolled back together.&quot; ALWAYS_BEGIN is what actually opens that transaction at runtime, and ALWAYS_JOIN is what attaches each statement to it. All three pieces have to line up, or you&#39;re back to auto-commit behavior without realizing it.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/855/1*8hFWzZ3GU65yfNLrE2zhOg.png" /></figure><p>When the payment-reference update fails, an alert like this typically lands in an inbox or a monitoring channel:</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/827/1*2l8N-4a704oO21IQY2E3qw.png" /></figure><h3>Key Takeaways</h3><ul><li>Mule doesn’t auto-rollback across separate operations, each Database call commits on its own unless you explicitly group them in a transactional &lt;try&gt; scope</li><li>ALWAYS_BEGIN + ALWAYS_JOIN is the standard pattern for making multiple DB statements succeed or fail as one unit</li><li>External systems like payment gateways or third-party APIs aren’t transactional resources, once something’s sent, it’s sent</li><li>Distributed XA transactions exist but are rarely worth the complexity for mixed DB + API flows</li><li>Compensating actions are the pragmatic, production-proven approach group what can be rolled back, flag what can’t</li><li>Design for visible failure, not silent inconsistency</li></ul><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=a1968621d9ba" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[Introduction to Cloud Platforms in MuleSoft]]></title>
            <link>https://medium.com/@vadigireddygariswetha/introduction-to-cloud-platforms-in-mulesoft-fce2db14003c?source=rss-a37bccd54c8e------2</link>
            <guid isPermaLink="false">https://medium.com/p/fce2db14003c</guid>
            <dc:creator><![CDATA[Vadigireddygariswetha]]></dc:creator>
            <pubDate>Tue, 16 Sep 2025 12:00:29 GMT</pubDate>
            <atom:updated>2025-09-16T12:02:12.875Z</atom:updated>
            <content:encoded><![CDATA[<p>Cloud platforms play a crucial role in modern integration architecture. MuleSoft Anypoint Platform itself is cloud-native, and it supports integrations with leading cloud providers like <strong>AWS (Amazon Web Services), Microsoft Azure, and Google Cloud Platform (GCP)</strong>.</p><p>Using cloud platforms, developers can build scalable, secure, and highly available integrations, deploy Mule apps, and connect cloud-native services with on-premise applications.</p><ol><li><strong>MuleSoft CloudHub</strong></li></ol><blockquote>CloudHub is MuleSoft’s fully managed <strong>Integration Platform as a Service (iPaaS)</strong>.<br>It allows you to deploy Mule applications in the cloud without worrying about infrastructure management.<br>Provides auto-scaling, high availability, and easy monitoring</blockquote><p><strong>Real-Time Scenario:</strong></p><p>A company wants to expose a REST API to their partners, with data coming from an on-premises ERP system.</p><p><strong>Solution:</strong></p><p>Design the Mule app in Anypoint Studio.<br>Deploy the application to CloudHub.<br>Use CloudHub’s VPC and VPN capabilities to securely connect to on-prem systems.<br>API is accessible globally via a public domain.</p><p><strong>Benefits:</strong></p><p>Zero server management.<br>Easy scalability.<br>Centralized monitoring with Anypoint Monitoring.</p><p><strong>2. AWS (Amazon Web Services)</strong></p><blockquote>By using the <strong>AWS Connectors</strong> in MuleSoft Anypoint Studio, developers can connect to various AWS services such as <strong>Amazon S3, SQS, SNS, DynamoDB, Lambda, and RDS</strong> without writing complex custom code or managing low-level configurations.</blockquote><p><strong>Real-Time Scenario:</strong></p><p>Upload customer files to S3 for archival, triggered by an HTTP request.</p><p><strong>Solution:</strong></p><p>Configure HTTP Listener Connector to accept file upload requests.<br>Use the AWS S3 Connector to upload files to a specific S3 bucket.<br>Respond back to the caller with a success message.</p><p><strong>Example DataWeave Transformation:</strong></p><p>%dw 2.0<br>output application/json<br> — -<br>{<br> bucket: “customer-files”,<br> fileKey: “uploads/” ++ vars.fileName,<br> fileSize: vars.fileSize<br>}</p><p><strong>3. Microsoft Azure</strong></p><blockquote>Supports connectors for services such as Azure Blob Storage, Azure SQL Database, Event Hubs, and Azure Active Directory.</blockquote><p><strong>Real-Time Scenario:</strong></p><p>Process orders submitted via an HTTP API and store order data in Azure SQL Database.</p><p><strong>Solution:</strong></p><p>HTTP Listener accepts order data.<br>DataWeave maps the order JSON payload to the relational table structure.<br>Use Azure SQL Connector to insert data into the database.<br>This provides seamless integration between web APIs and cloud databases.</p><p><strong>4. Google Cloud Platform (GCP)</strong></p><blockquote>Connect to Google Cloud Storage, BigQuery, Pub/Sub, Firebase, etc., using Google Cloud Connectors.</blockquote><p><strong>Real-Time Scenario:</strong></p><p>Publish application logs to Google Cloud Pub/Sub for real-time monitoring.</p><p><strong>Solution:</strong></p><p>Collect logs in MuleSoft app.<br>Use Google Pub/Sub Connector to publish logs to a Pub/Sub topic.<br>Downstream consumers subscribe to the topic and process logs in near real-time.</p><p><strong>Conclusion</strong></p><p>Integrating cloud platforms like AWS, Azure, GCP, and CloudHub with MuleSoft enables scalable, secure, and efficient integrations. With powerful connectors and API-led architecture, MuleSoft simplifies connecting systems across cloud and on-premise environments, helping businesses accelerate digital transformation and improve agility.</p><p>This article provided an overview of cloud platforms in MuleSoft. In the upcoming articles, we will go deeper into each platform — starting with AWS — covering hands-on examples, real-time use cases, best practices, and step-by-step integration flows.</p><p>Reference: <br><a href="https://www.mulesoft.com/integration/aws?utm_source=chatgpt.com">https://www.mulesoft.com/integration/aws?utm_source=chatgpt.com</a></p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=fce2db14003c" width="1" height="1" alt="">]]></content:encoded>
        </item>
    </channel>
</rss>