<?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: Pravinraj Vincent</title>
    <description>The latest articles on DEV Community by Pravinraj Vincent (@pravinraj).</description>
    <link>https://dev.to/pravinraj</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.us-east-2.amazonaws.com%2Fuploads%2Fuser%2Fprofile_image%2F4191253%2F76aee8af-be2b-4657-8018-a04e544eb3ea.jpg</url>
      <title>DEV Community: Pravinraj Vincent</title>
      <link>https://dev.to/pravinraj</link>
    </image>
    <atom:link rel="self" type="application/rss+xml" href="https://dev.to/feed/pravinraj"/>
    <language>en</language>
    <item>
      <title>UiPath Pitfall #5: Expecting Try Catch to Run Before the Global Exception Handler</title>
      <dc:creator>Pravinraj Vincent</dc:creator>
      <pubDate>Fri, 06 Nov 2026 23:30:35 +0000</pubDate>
      <link>https://dev.to/pravinraj/uipath-pitfall-5-expecting-try-catch-to-run-before-the-global-exception-handler-4cjd</link>
      <guid>https://dev.to/pravinraj/uipath-pitfall-5-expecting-try-catch-to-run-before-the-global-exception-handler-4cjd</guid>
      <description>&lt;p&gt;You put the risky activity inside &lt;code&gt;Try Catch&lt;/code&gt;, add a matching exception type, and write your recovery. Then you add a Global Exception Handler. Suddenly the robot retries the activity and stops without reaching your local &lt;code&gt;Catch&lt;/code&gt;.&lt;/p&gt;

&lt;p&gt;It's tempting to rebuild the catch or assume Studio has lost track of the scope. Check the global handler first.&lt;/p&gt;

&lt;p&gt;This resurfaced in &lt;a href="https://forum.uipath.com/t/exception-not-getting-caught/5788543" rel="noopener noreferrer"&gt;Exception not getting caught&lt;/a&gt;. Earlier reports describe &lt;a href="https://forum.uipath.com/t/local-trycatch-bypassed-by-global-exception-handler-in-uipath-studio-25-0-157/2806500" rel="noopener noreferrer"&gt;local catches apparently being bypassed&lt;/a&gt; and &lt;a href="https://forum.uipath.com/t/uipath-global-exception-handler-is-not-working-with-try-catch/2833327" rel="noopener noreferrer"&gt;a download process retrying and exiting instead of recovering locally&lt;/a&gt;. They establish a recurring source of confusion, not a confirmed diagnosis for every affected project.&lt;/p&gt;

&lt;h2&gt;
  
  
  The name doesn't tell the whole story
&lt;/h2&gt;

&lt;p&gt;The &lt;a href="https://docs.uipath.com/studio/standalone/latest/user-guide/global-exception-handler" rel="noopener noreferrer"&gt;Global Exception Handler docs&lt;/a&gt; explain that, with nested activities, the handler executes for each activity in the call stack. Activities directly inside &lt;code&gt;Try Catch&lt;/code&gt; are treated differently from activities nested inside another container.&lt;/p&gt;

&lt;p&gt;These layouts therefore aren't interchangeable when investigating handler execution:&lt;br&gt;
&lt;/p&gt;

&lt;div class="highlight js-code-highlight"&gt;
&lt;pre class="highlight plaintext"&gt;&lt;code&gt;Try Catch
  Try
    Failing activity
&lt;/code&gt;&lt;/pre&gt;

&lt;/div&gt;





&lt;div class="highlight js-code-highlight"&gt;
&lt;pre class="highlight plaintext"&gt;&lt;code&gt;Try Catch
  Try
    Sequence
      Failing activity
&lt;/code&gt;&lt;/pre&gt;

&lt;/div&gt;



&lt;p&gt;These are layout sketches, not executable workflow code. UiPath recommends a &lt;code&gt;Sequence&lt;/code&gt; inside &lt;code&gt;Try&lt;/code&gt; when using the global handler with &lt;code&gt;Try Catch&lt;/code&gt; so the global handler executes. Adding that container isn't a blanket fix to make your local catch run first.&lt;/p&gt;

&lt;h2&gt;
  
  
  Read the result literally
&lt;/h2&gt;

&lt;p&gt;The handler's &lt;code&gt;result&lt;/code&gt; argument controls what happens next:&lt;/p&gt;

&lt;ul&gt;
&lt;li&gt;
&lt;code&gt;Continue&lt;/code&gt; rethrows the exception.&lt;/li&gt;
&lt;li&gt;
&lt;code&gt;Ignore&lt;/code&gt; ignores it and continues from the next activity.&lt;/li&gt;
&lt;li&gt;
&lt;code&gt;Retry&lt;/code&gt; retries the activity that failed.&lt;/li&gt;
&lt;li&gt;
&lt;code&gt;Abort&lt;/code&gt; stops execution after the handler runs.&lt;/li&gt;
&lt;/ul&gt;

&lt;p&gt;For a VB project, the rethrow choice can be set with an &lt;code&gt;Assign&lt;/code&gt; activity, targeting &lt;code&gt;result&lt;/code&gt; with the value &lt;code&gt;ErrorAction.Continue&lt;/code&gt;. Written as an assignment:&lt;br&gt;
&lt;/p&gt;

&lt;div class="highlight js-code-highlight"&gt;
&lt;pre class="highlight vb"&gt;&lt;code&gt;&lt;span class="n"&gt;result&lt;/span&gt; &lt;span class="o"&gt;=&lt;/span&gt; &lt;span class="n"&gt;ErrorAction&lt;/span&gt;&lt;span class="p"&gt;.&lt;/span&gt;&lt;span class="n"&gt;Continue&lt;/span&gt;
&lt;/code&gt;&lt;/pre&gt;

&lt;/div&gt;



&lt;p&gt;This is a branch choice, not a complete handler or a universal replacement for your stopping policy.&lt;/p&gt;

&lt;p&gt;The important distinction is &lt;code&gt;Continue&lt;/code&gt; versus &lt;code&gt;Ignore&lt;/code&gt;. “Continue” here doesn't mean successful recovery. It means the error is rethrown. “Ignore” doesn't forward the error to your catch.&lt;/p&gt;

&lt;p&gt;If your design needs an enclosing, matching &lt;code&gt;Catch&lt;/code&gt; to handle the failure, inspect the handler branch that returns &lt;code&gt;Abort&lt;/code&gt;. Test the &lt;code&gt;Continue&lt;/code&gt; rethrow path in an isolated copy before changing the shared handler. Don't replace every abort with continue without deciding which errors should still stop the process.&lt;/p&gt;

&lt;h2&gt;
  
  
  Rule out the separate silent-error setting
&lt;/h2&gt;

&lt;p&gt;The &lt;a href="https://docs.uipath.com/activities/other/latest/workflow/try-catch" rel="noopener noreferrer"&gt;Try Catch activity docs&lt;/a&gt; state that no error is caught if an activity in &lt;code&gt;Try&lt;/code&gt; has &lt;code&gt;ContinueOnError&lt;/code&gt; set to &lt;code&gt;True&lt;/code&gt;. Check it where the failing activity exposes the property. That's different from a global handler returning &lt;code&gt;Ignore&lt;/code&gt;, even if both leave you wondering why your catch didn't run.&lt;/p&gt;

&lt;p&gt;A useful reproduction keeps the container layout unchanged and varies one handler result at a time. Compare the route through the global handler and local catch before restoring the real recovery steps. That gives you evidence instead of another round of catch-type guessing.&lt;/p&gt;

&lt;p&gt;Have you hit this in an existing process? Share the layout and handler result in the &lt;a href="https://forum.uipath.com/t/uipath-pitfall-of-the-week-5-expecting-try-catch-to-run-before-the-global-exception-handler/5788632" rel="noopener noreferrer"&gt;original forum thread&lt;/a&gt;.&lt;/p&gt;

&lt;p&gt;Originally posted on the UiPath Forum: &lt;a href="https://forum.uipath.com/t/uipath-pitfall-of-the-week-5-expecting-try-catch-to-run-before-the-global-exception-handler/5788632" rel="noopener noreferrer"&gt;https://forum.uipath.com/t/uipath-pitfall-of-the-week-5-expecting-try-catch-to-run-before-the-global-exception-handler/5788632&lt;/a&gt;&lt;/p&gt;

&lt;p&gt;I post short UiPath fix videos from real forum questions on my YouTube channel, &lt;a href="https://www.youtube.com/@rpasimplified" rel="noopener noreferrer"&gt;RPA Simplified&lt;/a&gt;.&lt;/p&gt;

</description>
      <category>rpa</category>
      <category>studio</category>
      <category>automation</category>
      <category>uipath</category>
    </item>
    <item>
      <title>Snowflake connector 200-row cap in UiPath Integration Service: how to paginate or go async</title>
      <dc:creator>Pravinraj Vincent</dc:creator>
      <pubDate>Wed, 04 Nov 2026 02:43:33 +0000</pubDate>
      <link>https://dev.to/pravinraj/snowflake-connector-200-row-cap-in-uipath-integration-service-how-to-paginate-or-go-async-1j93</link>
      <guid>https://dev.to/pravinraj/snowflake-connector-200-row-cap-in-uipath-integration-service-how-to-paginate-or-go-async-1j93</guid>
      <description>&lt;p&gt;&lt;em&gt;Pravinraj Vincent, UiPath Solutions Engineer&lt;/em&gt;&lt;/p&gt;

&lt;p&gt;If you've moved a Snowflake query from the old activity package to the Integration Service Snowflake connector and your result suddenly stops at 200 rows, nothing is broken. That's how the connector is designed. This came up on the forum recently (someone needed about 80,000 rows, and I &lt;a href="https://forum.uipath.com/t/how-to-retrieve-large-datasets-80k-rows-with-the-new-snowflake-connector-200-row-limit-issue/5774775/5" rel="noopener noreferrer"&gt;replied in the thread&lt;/a&gt;), so here's the longer version: why it happens, and two patterns that work.&lt;/p&gt;

&lt;h2&gt;
  
  
  Why you only get 200 rows
&lt;/h2&gt;

&lt;p&gt;The docs for &lt;strong&gt;Execute Select Query Synchronously&lt;/strong&gt; are explicit: it retrieves at most 200 records. Even if you set LIMIT to a higher number, you still get a maximum of 200. A bigger number in the SQL won't help, and there's no setting that raises the cap.&lt;/p&gt;

&lt;p&gt;What the docs recommend is to page through the data yourself. Put &lt;code&gt;LIMIT&lt;/code&gt; and &lt;code&gt;OFFSET&lt;/code&gt; directly in the Select Query and run it inside a Do While. The connector's technical references say the same thing: when you expect a large result set, always paginate with &lt;code&gt;LIMIT&lt;/code&gt; and &lt;code&gt;OFFSET&lt;/code&gt;.&lt;/p&gt;

&lt;h2&gt;
  
  
  Pattern A: page the synchronous query
&lt;/h2&gt;

&lt;p&gt;Use this when each page comes back quickly.&lt;/p&gt;

&lt;ol&gt;
&lt;li&gt;Write a query with a stable sort on a unique key, then &lt;code&gt;LIMIT 200 OFFSET n&lt;/code&gt;:
&lt;/li&gt;
&lt;/ol&gt;

&lt;div class="highlight js-code-highlight"&gt;
&lt;pre class="highlight sql"&gt;&lt;code&gt;   &lt;span class="k"&gt;SELECT&lt;/span&gt; &lt;span class="n"&gt;ORDER_ID&lt;/span&gt;&lt;span class="p"&gt;,&lt;/span&gt; &lt;span class="n"&gt;CUSTOMER_ID&lt;/span&gt;&lt;span class="p"&gt;,&lt;/span&gt; &lt;span class="n"&gt;AMOUNT&lt;/span&gt;
   &lt;span class="k"&gt;FROM&lt;/span&gt; &lt;span class="n"&gt;SALES&lt;/span&gt;&lt;span class="p"&gt;.&lt;/span&gt;&lt;span class="n"&gt;ORDERS&lt;/span&gt;
   &lt;span class="k"&gt;ORDER&lt;/span&gt; &lt;span class="k"&gt;BY&lt;/span&gt; &lt;span class="n"&gt;ORDER_ID&lt;/span&gt;
   &lt;span class="k"&gt;LIMIT&lt;/span&gt; &lt;span class="mi"&gt;200&lt;/span&gt; &lt;span class="k"&gt;OFFSET&lt;/span&gt; &lt;span class="mi"&gt;0&lt;/span&gt;
&lt;/code&gt;&lt;/pre&gt;

&lt;/div&gt;



&lt;ol&gt;
&lt;li&gt;Generate the output schema with a literal query like the one above, with no variables in it. Generate output schema only works on a query without workflow variables. Once the schema exists, swap the &lt;code&gt;0&lt;/code&gt; for your offset variable.&lt;/li&gt;
&lt;li&gt;Inside a Do While, run &lt;strong&gt;Execute Select Query Synchronously&lt;/strong&gt;, append that page's records to your own collection, and add 200 to the offset.&lt;/li&gt;
&lt;li&gt;Stop when a page comes back with fewer than 200 rows. At that point you've reached the end.&lt;/li&gt;
&lt;/ol&gt;

&lt;p&gt;Two things will bite you here:&lt;/p&gt;

&lt;ul&gt;
&lt;li&gt;
&lt;strong&gt;Always add ORDER BY.&lt;/strong&gt; Snowflake's own docs say that without ORDER BY, LIMIT results are non-deterministic, because query results aren't in any particular order. If you page without a sort, you can get the same row twice and miss others. Sort on a unique key. Keep the ORDER BY at the same query level as the LIMIT, since Snowflake warns that an ORDER BY in a subquery doesn't carry over to an outer LIMIT.&lt;/li&gt;
&lt;li&gt;
&lt;strong&gt;Keep the data still while you page.&lt;/strong&gt; Each page is a separate query. If rows get inserted or deleted between pages, the offsets shift. Run it against data that isn't changing during the extract, or against a fixed snapshot.&lt;/li&gt;
&lt;/ul&gt;

&lt;p&gt;Also watch for the connector's query restrictions. Comment patterns (&lt;code&gt;--&lt;/code&gt;, &lt;code&gt;#&lt;/code&gt;, &lt;code&gt;/* */&lt;/code&gt;) are blocked anywhere in the query, even inside string values, so don't leave SQL comments in the Select Query. The synchronous activity only accepts SELECT statements. The Snowflake VARIANT data type isn't supported.&lt;/p&gt;

&lt;h2&gt;
  
  
  Pattern B: go async for slow queries
&lt;/h2&gt;

&lt;p&gt;The technical references say to use &lt;strong&gt;Execute Any Query Asynchronously&lt;/strong&gt; for any query that takes more than 30 seconds. That's the path when the query itself is slow, not just the result set big.&lt;/p&gt;

&lt;ol&gt;
&lt;li&gt;
&lt;strong&gt;Execute Any Query Asynchronously&lt;/strong&gt; runs the query and returns a &lt;strong&gt;Query ID&lt;/strong&gt;. This activity also accepts non-SELECT work like insert, update and stored procedures.&lt;/li&gt;
&lt;li&gt;
&lt;strong&gt;Get Asynchronous Query Execution Status&lt;/strong&gt; takes that Query ID and returns a Status. Loop with a short Delay while it's &lt;code&gt;RUNNING&lt;/code&gt;, &lt;code&gt;QUEUED&lt;/code&gt;, &lt;code&gt;BLOCKED&lt;/code&gt; or &lt;code&gt;RESUMING_WAREHOUSE&lt;/code&gt;. Continue on &lt;code&gt;SUCCESS&lt;/code&gt;. Treat &lt;code&gt;FAILED_WITH_ERROR&lt;/code&gt; (a problem in the query) and &lt;code&gt;FAILED_WITH_INCIDENT&lt;/code&gt; (a server-side incident) as failures.&lt;/li&gt;
&lt;li&gt;
&lt;strong&gt;Get Asynchronous Query Execution Results&lt;/strong&gt; returns the rows. It has the same cap: at most 200 records per request. So page it the same way, using its own &lt;strong&gt;Offset&lt;/strong&gt; and &lt;strong&gt;Limit&lt;/strong&gt; properties. Its &lt;strong&gt;Additional conditions&lt;/strong&gt; field takes WHERE and ORDER BY clauses, so you can put &lt;code&gt;ORDER BY ORDER_ID&lt;/code&gt; there and keep the pages in a stable order. The docs example is &lt;code&gt;COLUMN_1='value' ORDER BY COLUMN_2 DESC&lt;/code&gt;.&lt;/li&gt;
&lt;/ol&gt;

&lt;p&gt;To generate the output schema for Get Asynchronous Query Execution Results, use a real Query ID from Execute Any Query Asynchronously. After the schema is generated, you can replace it with a variable. If you change the query itself, regenerate the schema. Changing only variables doesn't need it.&lt;/p&gt;

&lt;p&gt;If you don't want a robot sitting in a polling loop, the connector also has an &lt;strong&gt;Asynchronous Query Status Changed&lt;/strong&gt; trigger. It can start a process when an async query succeeds or hits an error.&lt;/p&gt;

&lt;h2&gt;
  
  
  Which one to pick
&lt;/h2&gt;

&lt;div class="table-wrapper-paragraph"&gt;&lt;table&gt;
&lt;thead&gt;
&lt;tr&gt;
&lt;th&gt;Situation&lt;/th&gt;
&lt;th&gt;Use&lt;/th&gt;
&lt;/tr&gt;
&lt;/thead&gt;
&lt;tbody&gt;
&lt;tr&gt;
&lt;td&gt;Query is fast, you just need more than 200 rows&lt;/td&gt;
&lt;td&gt;Pattern A: Execute Select Query Synchronously with ORDER BY + LIMIT 200 OFFSET n in a Do While&lt;/td&gt;
&lt;/tr&gt;
&lt;tr&gt;
&lt;td&gt;Query takes more than about 30 seconds&lt;/td&gt;
&lt;td&gt;Pattern B: Execute Any Query Asynchronously, poll status, then page the results 200 at a time&lt;/td&gt;
&lt;/tr&gt;
&lt;tr&gt;
&lt;td&gt;You need insert, update or a stored procedure&lt;/td&gt;
&lt;td&gt;Execute Any Query Asynchronously (the synchronous activity is SELECT only)&lt;/td&gt;
&lt;/tr&gt;
&lt;/tbody&gt;
&lt;/table&gt;&lt;/div&gt;

&lt;p&gt;If you're on Studio Desktop, the technical references also point to the Native Search activity in Snowflake package version 5.0.1. Its docs list a default limit of 1,000 records, with LIMIT and OFFSET in the query for paging. The point still stands: you page, you don't pull everything in one call.&lt;/p&gt;

&lt;p&gt;The original forum thread, with my reply, is here: &lt;a href="https://forum.uipath.com/t/how-to-retrieve-large-datasets-80k-rows-with-the-new-snowflake-connector-200-row-limit-issue/5774775/5" rel="noopener noreferrer"&gt;How to retrieve large datasets (~80k rows) with the new Snowflake Connector?&lt;/a&gt;&lt;/p&gt;

&lt;p&gt;More short UiPath fixes on the RPA Simplified YouTube channel: &lt;a href="https://www.youtube.com/@rpasimplified" rel="noopener noreferrer"&gt;https://www.youtube.com/@rpasimplified&lt;/a&gt;&lt;/p&gt;




&lt;h2&gt;
  
  
  Sources (claim to URL, all fetched Nov 3, 2026)
&lt;/h2&gt;

&lt;div class="table-wrapper-paragraph"&gt;&lt;table&gt;
&lt;thead&gt;
&lt;tr&gt;
&lt;th&gt;Claim&lt;/th&gt;
&lt;th&gt;Source&lt;/th&gt;
&lt;/tr&gt;
&lt;/thead&gt;
&lt;tbody&gt;
&lt;tr&gt;
&lt;td&gt;Execute Select Query Synchronously returns max 200 records even with a higher LIMIT; use LIMIT/OFFSET in the Select Query inside a Do While&lt;/td&gt;
&lt;td&gt;&lt;a href="https://docs.uipath.com/activities/other/latest/integration-service/uipath-snowflake-snowflake-execute-select-query-synchronously" rel="noopener noreferrer"&gt;https://docs.uipath.com/activities/other/latest/integration-service/uipath-snowflake-snowflake-execute-select-query-synchronously&lt;/a&gt;&lt;/td&gt;
&lt;/tr&gt;
&lt;tr&gt;
&lt;td&gt;Generate output schema needs a query without variables; variables can be added after&lt;/td&gt;
&lt;td&gt;same page, and &lt;a href="https://docs.uipath.com/activities/other/latest/integration-service/uipath-snowflake-snowflake-technical-references" rel="noopener noreferrer"&gt;https://docs.uipath.com/activities/other/latest/integration-service/uipath-snowflake-snowflake-technical-references&lt;/a&gt;
&lt;/td&gt;
&lt;/tr&gt;
&lt;tr&gt;
&lt;td&gt;Comment patterns blocked even inside strings; sync activity is SELECT only (INSERT/UPDATE/DELETE/MERGE/TRUNCATE blocked); VARIANT not supported&lt;/td&gt;
&lt;td&gt;Execute Select Query Synchronously page (above)&lt;/td&gt;
&lt;/tr&gt;
&lt;tr&gt;
&lt;td&gt;Always paginate large result sets with LIMIT and OFFSET&lt;/td&gt;
&lt;td&gt;Technical references (above)&lt;/td&gt;
&lt;/tr&gt;
&lt;tr&gt;
&lt;td&gt;Use Execute Any Query Asynchronously for queries over 30 seconds; non-SELECT commands only via the async activity&lt;/td&gt;
&lt;td&gt;Technical references (above)&lt;/td&gt;
&lt;/tr&gt;
&lt;tr&gt;
&lt;td&gt;Async activity returns a Query ID; runs select, insert, update and stored procedures; VARIANT not supported&lt;/td&gt;
&lt;td&gt;&lt;a href="https://docs.uipath.com/activities/other/latest/integration-service/uipath-snowflake-snowflake-execute-any-query-asynchronously-" rel="noopener noreferrer"&gt;https://docs.uipath.com/activities/other/latest/integration-service/uipath-snowflake-snowflake-execute-any-query-asynchronously-&lt;/a&gt;&lt;/td&gt;
&lt;/tr&gt;
&lt;tr&gt;
&lt;td&gt;Status values SUCCESS, RUNNING, QUEUED, BLOCKED, RESUMING_WAREHOUSE, FAILED_WITH_ERROR, FAILED_WITH_INCIDENT and their meanings&lt;/td&gt;
&lt;td&gt;&lt;a href="https://docs.uipath.com/activities/other/latest/integration-service/uipath-snowflake-snowflake-get-asynchronous-query-execution-status" rel="noopener noreferrer"&gt;https://docs.uipath.com/activities/other/latest/integration-service/uipath-snowflake-snowflake-get-asynchronous-query-execution-status&lt;/a&gt;&lt;/td&gt;
&lt;/tr&gt;
&lt;tr&gt;
&lt;td&gt;Get Asynchronous Query Execution Results: max 200 records per request; Offset, Limit and Additional conditions (WHERE, ORDER BY); schema from a real Query ID, then a variable; regenerate if the query changes&lt;/td&gt;
&lt;td&gt;&lt;a href="https://docs.uipath.com/activities/other/latest/integration-service/uipath-snowflake-snowflake-get-asynchronous-query-execution-result" rel="noopener noreferrer"&gt;https://docs.uipath.com/activities/other/latest/integration-service/uipath-snowflake-snowflake-get-asynchronous-query-execution-result&lt;/a&gt;&lt;/td&gt;
&lt;/tr&gt;
&lt;tr&gt;
&lt;td&gt;Asynchronous Query Status Changed trigger can start a process on success or error&lt;/td&gt;
&lt;td&gt;Technical references (above)&lt;/td&gt;
&lt;/tr&gt;
&lt;tr&gt;
&lt;td&gt;Native Search on Studio Desktop from package 5.0.1&lt;/td&gt;
&lt;td&gt;Technical references (above)&lt;/td&gt;
&lt;/tr&gt;
&lt;tr&gt;
&lt;td&gt;Native Search default limit 1,000 records, LIMIT/OFFSET in the query to paginate&lt;/td&gt;
&lt;td&gt;&lt;a href="https://docs.uipath.com/activities/other/latest/integration-service/uipath-snowflake-snowflake-native-search" rel="noopener noreferrer"&gt;https://docs.uipath.com/activities/other/latest/integration-service/uipath-snowflake-snowflake-native-search&lt;/a&gt;&lt;/td&gt;
&lt;/tr&gt;
&lt;tr&gt;
&lt;td&gt;Without ORDER BY, LIMIT results are non-deterministic; keep ORDER BY and LIMIT at the same query level&lt;/td&gt;
&lt;td&gt;&lt;a href="https://docs.snowflake.com/en/sql-reference/constructs/limit" rel="noopener noreferrer"&gt;https://docs.snowflake.com/en/sql-reference/constructs/limit&lt;/a&gt;&lt;/td&gt;
&lt;/tr&gt;
&lt;tr&gt;
&lt;td&gt;Forum case: ~80k rows, new connector returns 200&lt;/td&gt;
&lt;td&gt;&lt;a href="https://forum.uipath.com/t/how-to-retrieve-large-datasets-80k-rows-with-the-new-snowflake-connector-200-row-limit-issue/5774775" rel="noopener noreferrer"&gt;https://forum.uipath.com/t/how-to-retrieve-large-datasets-80k-rows-with-the-new-snowflake-connector-200-row-limit-issue/5774775&lt;/a&gt;&lt;/td&gt;
&lt;/tr&gt;
&lt;/tbody&gt;
&lt;/table&gt;&lt;/div&gt;

</description>
      <category>uipath</category>
      <category>rpa</category>
      <category>automation</category>
      <category>snowflake</category>
    </item>
    <item>
      <title>An Operating Model for Reliable UiPath Unattended Automation</title>
      <dc:creator>Pravinraj Vincent</dc:creator>
      <pubDate>Mon, 02 Nov 2026 01:04:17 +0000</pubDate>
      <link>https://dev.to/pravinraj/an-operating-model-for-reliable-uipath-unattended-automation-590k</link>
      <guid>https://dev.to/pravinraj/an-operating-model-for-reliable-uipath-unattended-automation-590k</guid>
      <description>&lt;p&gt;&lt;em&gt;Pravinraj Vincent, UiPath Solutions Engineer&lt;/em&gt;&lt;/p&gt;

&lt;h2&gt;
  
  
  Who this is for
&lt;/h2&gt;

&lt;p&gt;This paper is for platform leads, solution architects and COE owners who already run unattended robots in production. It isn't a build guide. It covers the decisions above individual workflows: machines and folders, runtime capacity, where retries live, what you measure, and how changes reach production.&lt;/p&gt;

&lt;p&gt;Every product behavior below is taken from UiPath documentation, linked inline. Where I give an opinion, I mark it as my recommendation and say which facts it rests on.&lt;/p&gt;

&lt;h2&gt;
  
  
  Thesis
&lt;/h2&gt;

&lt;p&gt;Most unattended failures I see on the forum aren't bugs in the business logic. They come from the platform around it: a session that can't be created after patching, a browser policy pushed by another team, a runtime pool sized for the average day, a retry policy that counts the same invoice twice. Reliability at scale is mostly an operating model problem. The platform team's job is to name the failure domains, give each an owner, and design so a failure in one domain stays there.&lt;/p&gt;

&lt;h2&gt;
  
  
  The failure domains
&lt;/h2&gt;

&lt;div class="table-wrapper-paragraph"&gt;&lt;table&gt;
&lt;thead&gt;
&lt;tr&gt;
&lt;th&gt;Domain&lt;/th&gt;
&lt;th&gt;What breaks&lt;/th&gt;
&lt;th&gt;Documented example&lt;/th&gt;
&lt;/tr&gt;
&lt;/thead&gt;
&lt;tbody&gt;
&lt;tr&gt;
&lt;td&gt;Windows session&lt;/td&gt;
&lt;td&gt;The robot can't log on or can't reach the desktop&lt;/td&gt;
&lt;td&gt;Logon policies such as "Display information about previous logons during user logon" block session creation with error 0x80070520 (&lt;a href="https://docs.uipath.com/robot/standalone/latest/admin-guide/session-troubleshooting" rel="noopener noreferrer"&gt;session troubleshooting&lt;/a&gt;)&lt;/td&gt;
&lt;/tr&gt;
&lt;tr&gt;
&lt;td&gt;Host configuration&lt;/td&gt;
&lt;td&gt;Same package behaves differently per machine&lt;/td&gt;
&lt;td&gt;Resolution and scaling differ between an RDP session and a local one (&lt;a href="https://docs.uipath.com/robot/standalone/latest/admin-guide/execution-troubleshooting" rel="noopener noreferrer"&gt;execution troubleshooting&lt;/a&gt;)&lt;/td&gt;
&lt;/tr&gt;
&lt;tr&gt;
&lt;td&gt;Browser policy&lt;/td&gt;
&lt;td&gt;Extension or selectors stop working after a policy change&lt;/td&gt;
&lt;td&gt;
&lt;code&gt;ProcessIsolationEnabled&lt;/code&gt; affects selectors outside the HTML content; &lt;code&gt;DeveloperToolsAvailability&lt;/code&gt; settings limit the extension (&lt;a href="https://docs.uipath.com/studio/standalone/latest/user-guide/chrome-group-policies" rel="noopener noreferrer"&gt;Chrome group policies&lt;/a&gt;)&lt;/td&gt;
&lt;/tr&gt;
&lt;tr&gt;
&lt;td&gt;Office dependency&lt;/td&gt;
&lt;td&gt;Excel steps fail where Excel isn't installed or set up&lt;/td&gt;
&lt;td&gt;Modern and classic Excel activities need Excel installed; Workbook activities don't (&lt;a href="https://docs.uipath.com/activities/other/latest/productivity/about-the-excel-activities-pack" rel="noopener noreferrer"&gt;Excel package&lt;/a&gt;)&lt;/td&gt;
&lt;/tr&gt;
&lt;tr&gt;
&lt;td&gt;Capacity&lt;/td&gt;
&lt;td&gt;Jobs sit in Pending&lt;/td&gt;
&lt;td&gt;Runtimes are drawn from the tenant pool per connected host (&lt;a href="https://docs.uipath.com/orchestrator/automation-cloud/latest/user-guide/useful-concepts-in-unattended-automation" rel="noopener noreferrer"&gt;unattended concepts&lt;/a&gt;)&lt;/td&gt;
&lt;/tr&gt;
&lt;tr&gt;
&lt;td&gt;Work distribution&lt;/td&gt;
&lt;td&gt;Items processed late, twice, or never&lt;/td&gt;
&lt;td&gt;Queue trigger scheduling can over- or under-schedule depending on a tenant setting (&lt;a href="https://docs.uipath.com/orchestrator/automation-cloud/latest/user-guide/queue-triggers" rel="noopener noreferrer"&gt;queue triggers&lt;/a&gt;)&lt;/td&gt;
&lt;/tr&gt;
&lt;tr&gt;
&lt;td&gt;Change&lt;/td&gt;
&lt;td&gt;A new version breaks production&lt;/td&gt;
&lt;td&gt;Pending jobs pick up an edited process version, running jobs don't (&lt;a href="https://docs.uipath.com/orchestrator/automation-cloud/latest/user-guide/managing-processes" rel="noopener noreferrer"&gt;managing processes&lt;/a&gt;)&lt;/td&gt;
&lt;/tr&gt;
&lt;/tbody&gt;
&lt;/table&gt;&lt;/div&gt;

&lt;p&gt;My recommendation: give each row a named owner. The session and browser rows usually belong to the desktop or VDI team, not the RPA team, and the fastest fix is a shared change calendar with them.&lt;/p&gt;

&lt;h2&gt;
  
  
  Machine and folder strategy
&lt;/h2&gt;

&lt;p&gt;Machine templates are UiPath's recommended machine type for unattended work. One template lets many hosts connect with a single key or client credential pair, regardless of hostname (&lt;a href="https://docs.uipath.com/orchestrator/automation-cloud/latest/user-guide/useful-concepts-in-unattended-automation" rel="noopener noreferrer"&gt;unattended concepts&lt;/a&gt;). The docs attach conditions to that: hosts under one template should share the same applications, installed at the same paths and versions, and every robot account in a folder needs a matching account on every machine in that template, because Orchestrator can start a job on any valid account and machine pair.&lt;/p&gt;

&lt;p&gt;That turns template design into a blast-radius decision.&lt;/p&gt;

&lt;div class="table-wrapper-paragraph"&gt;&lt;table&gt;
&lt;thead&gt;
&lt;tr&gt;
&lt;th&gt;Option&lt;/th&gt;
&lt;th&gt;Choose it when&lt;/th&gt;
&lt;th&gt;Trade-off&lt;/th&gt;
&lt;/tr&gt;
&lt;/thead&gt;
&lt;tbody&gt;
&lt;tr&gt;
&lt;td&gt;One broad template for a pool&lt;/td&gt;
&lt;td&gt;Hosts are built from one image and all processes need the same apps&lt;/td&gt;
&lt;td&gt;Simplest capacity math; one bad image change affects every process on the pool&lt;/td&gt;
&lt;/tr&gt;
&lt;tr&gt;
&lt;td&gt;Templates split by application stack (for example SAP hosts vs browser-only hosts)&lt;/td&gt;
&lt;td&gt;Processes need different software or different access rights&lt;/td&gt;
&lt;td&gt;Smaller blast radius; runtimes are split across pools, so idle capacity in one can't help another&lt;/td&gt;
&lt;/tr&gt;
&lt;tr&gt;
&lt;td&gt;Templates split by environment (non-production vs production)&lt;/td&gt;
&lt;td&gt;You want image changes to land somewhere safe first&lt;/td&gt;
&lt;td&gt;Needs a second set of hosts; gives you a natural canary&lt;/td&gt;
&lt;/tr&gt;
&lt;/tbody&gt;
&lt;/table&gt;&lt;/div&gt;

&lt;p&gt;UiPath also recommends non-persistent VDI to keep host configuration consistent, and installing the Robot in Service Mode for unattended work (&lt;a href="https://docs.uipath.com/orchestrator/automation-cloud/latest/user-guide/how-is-unattended-automation-performed" rel="noopener noreferrer"&gt;how unattended automation is performed&lt;/a&gt;). A User Mode robot can't run unattended automations (&lt;a href="https://docs.uipath.com/robot/standalone/latest/admin-guide/execution-troubleshooting" rel="noopener noreferrer"&gt;execution troubleshooting&lt;/a&gt;). For images, the CLI supports &lt;code&gt;UiRobot.exe connect&lt;/code&gt; with &lt;code&gt;--url&lt;/code&gt; plus either &lt;code&gt;--key&lt;/code&gt; or &lt;code&gt;--clientID&lt;/code&gt; and &lt;code&gt;--clientSecret&lt;/code&gt; (&lt;a href="https://docs.uipath.com/robot/standalone/latest/admin-guide/command-line-interface" rel="noopener noreferrer"&gt;Robot CLI&lt;/a&gt;). My recommendation is to run that at first boot rather than baking a connected robot into the golden image, so each clone registers under its own final hostname.&lt;/p&gt;

&lt;p&gt;Robot accounts assigned to a parent folder inherit access to its subfolders, and inherited roles can't be removed lower down (&lt;a href="https://docs.uipath.com/orchestrator/automation-cloud/latest/user-guide/useful-concepts-in-unattended-automation" rel="noopener noreferrer"&gt;unattended concepts&lt;/a&gt;). I'd design the folder tree around ownership and access boundaries first and around org charts second, because that inheritance rule makes over-broad parent assignments hard to undo.&lt;/p&gt;

&lt;h2&gt;
  
  
  Capacity and licensing
&lt;/h2&gt;

&lt;p&gt;The runtime rules, from &lt;a href="https://docs.uipath.com/orchestrator/automation-cloud/latest/user-guide/useful-concepts-in-unattended-automation" rel="noopener noreferrer"&gt;unattended concepts&lt;/a&gt;:&lt;/p&gt;

&lt;ul&gt;
&lt;li&gt;Runtimes are assigned to a machine object, and that number applies to each host connected to it. Three runtimes on a template with three connected hosts means up to nine concurrent automations.&lt;/li&gt;
&lt;li&gt;Runtimes come from the tenant pool when a host connects and go back when it disconnects.&lt;/li&gt;
&lt;li&gt;A host can run many background processes but only one foreground process at a time. UiPath's own example: 10 background processes plus 1 foreground process need 11 runtimes but only one host. A second simultaneous foreground process needs a second host.&lt;/li&gt;
&lt;/ul&gt;

&lt;p&gt;So capacity has two dimensions: runtimes and foreground slots. My recommendation is to model them separately. Background-heavy portfolios are runtime-bound; UI-heavy portfolios are host-bound, and adding runtimes alone won't clear the queue.&lt;/p&gt;

&lt;p&gt;For Cloud Robot - Serverless, consumption is per minute by machine size (Small 1 RU per minute, Standard 2, Medium 4, Large 10), and if the size is too small the process times out but still consumes robot units for the time it ran (&lt;a href="https://docs.uipath.com/orchestrator/automation-cloud/latest/user-guide/managing-processes" rel="noopener noreferrer"&gt;managing processes&lt;/a&gt;). Undersizing to save units can cost more.&lt;/p&gt;

&lt;h2&gt;
  
  
  Queue and retry design
&lt;/h2&gt;

&lt;p&gt;Queues are where reliability is either designed in or bolted on. A few documented behaviors drive the design:&lt;/p&gt;

&lt;ul&gt;
&lt;li&gt;Auto retry has separate options for Failed items and Abandoned items, with a Max # of retries field that takes a value from 1 to 50. Manual retries don't count toward that maximum (&lt;a href="https://docs.uipath.com/orchestrator/automation-cloud/latest/user-guide/managing-queues-in-orchestrator" rel="noopener noreferrer"&gt;managing queues&lt;/a&gt;).&lt;/li&gt;
&lt;li&gt;Only items failed with an Application exception are retried (&lt;a href="https://docs.uipath.com/orchestrator/automation-cloud/latest/user-guide/automation-best-practices" rel="noopener noreferrer"&gt;automation best practices&lt;/a&gt;). Business exceptions aren't.&lt;/li&gt;
&lt;li&gt;An item that stays In Progress for a long time (approx. 24 hours) without being processed becomes Abandoned (&lt;a href="https://docs.uipath.com/orchestrator/automation-cloud/latest/user-guide/queue-item-statuses" rel="noopener noreferrer"&gt;queue item statuses&lt;/a&gt;).&lt;/li&gt;
&lt;li&gt;Each retry is a new item. The &lt;code&gt;Key&lt;/code&gt; field holds the GUID of the first item in the chain, &lt;code&gt;RetryNumber&lt;/code&gt; counts automatic retries, and manual retries are tracked separately in &lt;code&gt;ManualRetryNumber&lt;/code&gt; (&lt;a href="https://docs.uipath.com/orchestrator/automation-cloud/latest/user-guide/managing-queues-in-orchestrator" rel="noopener noreferrer"&gt;managing queues&lt;/a&gt;).&lt;/li&gt;
&lt;li&gt;Retried and deleted items don't take part in unique reference checks, so UiPath advises against deleting the first item in a retry chain.&lt;/li&gt;
&lt;/ul&gt;

&lt;div class="table-wrapper-paragraph"&gt;&lt;table&gt;
&lt;thead&gt;
&lt;tr&gt;
&lt;th&gt;Retry layer&lt;/th&gt;
&lt;th&gt;Scope&lt;/th&gt;
&lt;th&gt;Use it for&lt;/th&gt;
&lt;/tr&gt;
&lt;/thead&gt;
&lt;tbody&gt;
&lt;tr&gt;
&lt;td&gt;Activity-level retry in a Global Exception Handler (&lt;code&gt;result = Retry&lt;/code&gt;)&lt;/td&gt;
&lt;td&gt;One activity&lt;/td&gt;
&lt;td&gt;Rare, attended or simple linear processes. Not available in libraries (&lt;a href="https://docs.uipath.com/studio/standalone/latest/user-guide/global-exception-handler" rel="noopener noreferrer"&gt;GEH&lt;/a&gt;)&lt;/td&gt;
&lt;/tr&gt;
&lt;tr&gt;
&lt;td&gt;Framework retry inside the job&lt;/td&gt;
&lt;td&gt;One transaction&lt;/td&gt;
&lt;td&gt;Transient system errors you can recover from in-session&lt;/td&gt;
&lt;/tr&gt;
&lt;tr&gt;
&lt;td&gt;Queue auto retry&lt;/td&gt;
&lt;td&gt;One item, new job possible&lt;/td&gt;
&lt;td&gt;Failures that need a fresh session or machine&lt;/td&gt;
&lt;/tr&gt;
&lt;tr&gt;
&lt;td&gt;Manual retry by a reviewer&lt;/td&gt;
&lt;td&gt;One item&lt;/td&gt;
&lt;td&gt;Failures that need a human decision&lt;/td&gt;
&lt;/tr&gt;
&lt;/tbody&gt;
&lt;/table&gt;&lt;/div&gt;

&lt;p&gt;My recommendation: pick one automated layer per failure type and write it down. The layers multiply. A GEH that retries an activity, inside a framework that retries the transaction, on a queue that retries the item, can turn one bad record into many attempts, and each queue retry shows up as a new item in reporting.&lt;/p&gt;

&lt;p&gt;On triggers, the docs say queue triggers are optimized for processes that loop through all available items before exiting, there's one queue trigger per queue, and a periodic check (30 minutes by default, configurable down to 10) restarts work for postponed or unprocessed items. They also warn that this check can tie up resources outside business hours if items are left over (&lt;a href="https://docs.uipath.com/orchestrator/automation-cloud/latest/user-guide/queue-triggers" rel="noopener noreferrer"&gt;queue triggers&lt;/a&gt;). The "Enable pending jobs strategy" tenant setting changes the capacity formula: on, it may over-schedule; off, it may under-schedule, with the periodic check catching up. I'd treat that setting as an architecture decision, not a default, and record why you chose it.&lt;/p&gt;

&lt;h2&gt;
  
  
  Graceful control
&lt;/h2&gt;

&lt;p&gt;Orchestrator's Stop is a soft stop. The job is marked Stopping and the robot is notified, but execution continues. The workflow has to check for it with &lt;a href="https://docs.uipath.com/activities/other/latest/workflow/should-stop" rel="noopener noreferrer"&gt;Should Stop&lt;/a&gt; and exit on its own; if it never acts on the signal, the job runs to the end and finishes as Successful. Kill forcefully stops the job and marks it Stopped (&lt;a href="https://docs.uipath.com/orchestrator/automation-cloud/latest/user-guide/managing-jobs" rel="noopener noreferrer"&gt;managing jobs&lt;/a&gt;). The Robot guide adds that Kill first sends a Cancel request, which lets Finally blocks run cleanup, before it forces termination (&lt;a href="https://docs.uipath.com/robot/standalone/2023.10/admin-guide/stopping-a-process" rel="noopener noreferrer"&gt;stopping a process, Robot 2023.10&lt;/a&gt;). UiPath recommends Stop over Kill (&lt;a href="https://docs.uipath.com/orchestrator/automation-cloud/latest/user-guide/automation-best-practices" rel="noopener noreferrer"&gt;automation best practices&lt;/a&gt;). The same Robot page notes that a stopped or cancelled transaction can be recorded as successful if status is set in the wrong block, and advises setting Successful only in the Try block and Failed only in the Catch block.&lt;/p&gt;

&lt;p&gt;This matters because machine maintenance mode offers three choices: wait for running jobs to finish (the default, which issues no stop command), stop all running jobs and wait, or kill them now. UiPath notes that jobs only shut down gracefully if the automation is built to handle stop requests. Maintenance is set per session (machine plus hostname), and pending jobs keep their status until another machine, or the same one once it's back, picks them up (&lt;a href="https://docs.uipath.com/orchestrator/automation-cloud/latest/user-guide/managing-machines" rel="noopener noreferrer"&gt;managing machines&lt;/a&gt;). If your processes honor Should Stop, "stop and wait" becomes a safe patch-night tool. If they don't, you're left choosing between waiting indefinitely and killing.&lt;/p&gt;

&lt;h2&gt;
  
  
  Observability and SLOs
&lt;/h2&gt;

&lt;p&gt;Orchestrator gives you several building blocks:&lt;/p&gt;

&lt;ul&gt;
&lt;li&gt;
&lt;strong&gt;Queue SLAs.&lt;/strong&gt; SLA predictions per queue, with an optional Risk SLA buffer, up to 90 days (&lt;a href="https://docs.uipath.com/orchestrator/automation-cloud/latest/user-guide/managing-queues-in-orchestrator" rel="noopener noreferrer"&gt;managing queues&lt;/a&gt;).&lt;/li&gt;
&lt;li&gt;
&lt;strong&gt;Notifications.&lt;/strong&gt; Subscribable events include unattended robot unresponsive or disconnected, trigger automatically disabled, job execution failed, jobs pending, transaction failed with application exception, and queue SLA error (&lt;a href="https://docs.uipath.com/orchestrator/automation-cloud/latest/user-guide/notifications" rel="noopener noreferrer"&gt;notifications&lt;/a&gt;).&lt;/li&gt;
&lt;li&gt;
&lt;strong&gt;Job recording.&lt;/strong&gt; Recording of failed jobs, failed queue transactions, or all jobs. The maximum duration depends on the plan: 3 minutes for non-entitled cloud tenants, up to 15 on Enterprise or Advanced (&lt;a href="https://docs.uipath.com/orchestrator/automation-cloud/latest/user-guide/managing-processes" rel="noopener noreferrer"&gt;managing processes&lt;/a&gt;).&lt;/li&gt;
&lt;/ul&gt;

&lt;p&gt;Two documented limits shape how far you can lean on these. Transaction-failure notifications are deduplicated to 10 per queue per 60 minutes, with an hourly summary (&lt;a href="https://docs.uipath.com/orchestrator/automation-cloud/latest/user-guide/notifications" rel="noopener noreferrer"&gt;notifications&lt;/a&gt;). And by default a trigger is disabled automatically after 10 failed launches if it hadn't launched successfully in the last day (&lt;a href="https://docs.uipath.com/orchestrator/automation-cloud/latest/user-guide/managing-triggers" rel="noopener noreferrer"&gt;managing triggers&lt;/a&gt;). That second one is quiet: work just stops arriving.&lt;/p&gt;

&lt;p&gt;My recommendation is to define SLOs on business outcomes, not job status: share of business items finished inside the queue SLA, counted once per &lt;code&gt;Key&lt;/code&gt;, plus time from item creation to final status. Job success rate alone hides the stopped-but-successful case above and double counts retries. Treat "trigger automatically disabled" and "unattended robot disconnected" as page-worthy, since both stop throughput without a failed job to show for it.&lt;/p&gt;

&lt;h2&gt;
  
  
  Governance and change control
&lt;/h2&gt;

&lt;p&gt;Automation Ops policies can restrict which package feeds Studio can use, enforce Workflow Analyzer rules, and require projects to pass analysis before running or publishing. Robot policies add Runtime Analyzer rules that block listed apps, URLs, or email recipients at runtime. Policies deploy at tenant, group, or user level (&lt;a href="https://docs.uipath.com/automation-ops/automation-cloud/latest/user-guide/governance-intro" rel="noopener noreferrer"&gt;governance&lt;/a&gt;).&lt;/p&gt;

&lt;p&gt;For infrastructure change, take the browser. UiPath documents that &lt;code&gt;DeveloperToolsAvailability = 2&lt;/code&gt; breaks the Chromium API input method unless the allowlist and blocklist policies are both set for the UiPath extension ID, and that &lt;code&gt;DisableScreenshots&lt;/code&gt; on Chrome 154 and later breaks Chromium API input and Inject Js Script (&lt;a href="https://docs.uipath.com/studio/standalone/latest/user-guide/chrome-group-policies" rel="noopener noreferrer"&gt;Chrome group policies&lt;/a&gt;). Those are security team decisions. My recommendation is to register the robot pool as a stakeholder in browser and GPO change, and to route those changes through a non-production template first.&lt;/p&gt;

&lt;h2&gt;
  
  
  Release and rollback
&lt;/h2&gt;

&lt;div class="table-wrapper-paragraph"&gt;&lt;table&gt;
&lt;thead&gt;
&lt;tr&gt;
&lt;th&gt;Option&lt;/th&gt;
&lt;th&gt;Choose it when&lt;/th&gt;
&lt;th&gt;Trade-off&lt;/th&gt;
&lt;/tr&gt;
&lt;/thead&gt;
&lt;tbody&gt;
&lt;tr&gt;
&lt;td&gt;Manual publish and process update in Orchestrator&lt;/td&gt;
&lt;td&gt;Small estate, few releases&lt;/td&gt;
&lt;td&gt;No enforced tests or approval&lt;/td&gt;
&lt;/tr&gt;
&lt;tr&gt;
&lt;td&gt;Automation Ops Pipelines&lt;/td&gt;
&lt;td&gt;You want UiPath-native CI/CD from GitHub or Azure DevOps&lt;/td&gt;
&lt;td&gt;Pipeline jobs run on an unattended robot, so it needs a dedicated folder, a robot account, and a machine or serverless robot to run on (&lt;a href="https://docs.uipath.com/automation-ops/automation-cloud/latest/user-guide/about-automation-ops---pipelines" rel="noopener noreferrer"&gt;Pipelines&lt;/a&gt;)&lt;/td&gt;
&lt;/tr&gt;
&lt;tr&gt;
&lt;td&gt;UiPath CLI, Azure DevOps extension, or Jenkins plugin&lt;/td&gt;
&lt;td&gt;You already run an enterprise CI/CD system (&lt;a href="https://docs.uipath.com/automation-ops/automation-cloud/latest/user-guide/about-cicd-pipelines" rel="noopener noreferrer"&gt;CI/CD&lt;/a&gt;)&lt;/td&gt;
&lt;td&gt;You own the pipeline code&lt;/td&gt;
&lt;/tr&gt;
&lt;/tbody&gt;
&lt;/table&gt;&lt;/div&gt;

&lt;p&gt;Built-in pipeline templates include "Update with tests" and "Build and promote with approval," which adds an approval step before promoting to a second environment (&lt;a href="https://docs.uipath.com/automation-ops/automation-cloud/latest/user-guide/about-automation-ops---pipelines" rel="noopener noreferrer"&gt;Pipelines&lt;/a&gt;).&lt;/p&gt;

&lt;p&gt;Rollback is a process setting: Edit Process lets you pick a version or select Rollback to return to the previous one, and UiPath recommends keeping at least one old version for that purpose (&lt;a href="https://docs.uipath.com/orchestrator/automation-cloud/latest/user-guide/managing-processes" rel="noopener noreferrer"&gt;managing processes&lt;/a&gt;, &lt;a href="https://docs.uipath.com/orchestrator/automation-cloud/latest/user-guide/automation-best-practices" rel="noopener noreferrer"&gt;automation best practices&lt;/a&gt;). Running jobs keep the version they started with, while pending jobs and new triggers use the updated one. My recommendation: release when the queue is drained or the trigger is paused, so you know which version touched which items.&lt;/p&gt;

&lt;h2&gt;
  
  
  Cost and risk
&lt;/h2&gt;

&lt;p&gt;The cheapest reliability work is usually removing a dependency rather than hardening it. Two examples from the docs. Workbook activities read and write Excel files without Excel installed (&lt;a href="https://docs.uipath.com/activities/other/latest/productivity/about-the-excel-activities-pack" rel="noopener noreferrer"&gt;Excel package&lt;/a&gt;), which takes Office setup out of the robot's failure domain when you don't need the Excel UI. And the Microsoft 365 Download File activity fetches a OneDrive or SharePoint file through a connection into a local path, with an explicit "if a file already exists" choice (&lt;a href="https://docs.uipath.com/activities/other/latest/productivity/office365-drive-download-file-connections" rel="noopener noreferrer"&gt;Download File&lt;/a&gt;). My recommendation is to stage inputs that way instead of relying on a per-user sync client being caught up when a robot session starts. With &lt;code&gt;Sites.Selected&lt;/code&gt;, browsing lists nothing, so you enter the file URL or ID.&lt;/p&gt;

&lt;p&gt;On risk, the trade-off to state openly is concentration: one big shared pool uses runtimes efficiently and also turns one image or policy change into a tenant-wide incident.&lt;/p&gt;

&lt;h2&gt;
  
  
  Maturity model
&lt;/h2&gt;

&lt;p&gt;This is my own framing, not a UiPath model.&lt;/p&gt;

&lt;div class="table-wrapper-paragraph"&gt;&lt;table&gt;
&lt;thead&gt;
&lt;tr&gt;
&lt;th&gt;Level&lt;/th&gt;
&lt;th&gt;Characteristics&lt;/th&gt;
&lt;/tr&gt;
&lt;/thead&gt;
&lt;tbody&gt;
&lt;tr&gt;
&lt;td&gt;1. Reactive&lt;/td&gt;
&lt;td&gt;Per-VM machines, jobs on schedules, failures found by business users&lt;/td&gt;
&lt;/tr&gt;
&lt;tr&gt;
&lt;td&gt;2. Standardized&lt;/td&gt;
&lt;td&gt;Machine templates on a common image, Service Mode robots, queues with set statuses, one retry layer per failure type&lt;/td&gt;
&lt;/tr&gt;
&lt;tr&gt;
&lt;td&gt;3. Observed&lt;/td&gt;
&lt;td&gt;Queue SLAs, subscribed notifications for robot, trigger, and transaction events, outcome-based SLOs counted per &lt;code&gt;Key&lt;/code&gt;
&lt;/td&gt;
&lt;/tr&gt;
&lt;tr&gt;
&lt;td&gt;4. Governed&lt;/td&gt;
&lt;td&gt;Automation Ops policies, CI/CD with tests and approval, documented rollback, robot pool included in GPO and browser change&lt;/td&gt;
&lt;/tr&gt;
&lt;tr&gt;
&lt;td&gt;5. Engineered&lt;/td&gt;
&lt;td&gt;Non-production template as canary, capacity modeled on runtimes and foreground slots, dependencies removed where possible, maintenance via graceful stop&lt;/td&gt;
&lt;/tr&gt;
&lt;/tbody&gt;
&lt;/table&gt;&lt;/div&gt;

&lt;h2&gt;
  
  
  One-page checklist
&lt;/h2&gt;

&lt;p&gt;&lt;strong&gt;Failure domains&lt;/strong&gt;&lt;/p&gt;

&lt;ul&gt;
&lt;li&gt;[ ] Each domain (session, host, browser, Office, capacity, distribution, change) has a named owner&lt;/li&gt;
&lt;li&gt;[ ] Logon and RDS session policies on robot hosts match UiPath's session troubleshooting guidance&lt;/li&gt;
&lt;/ul&gt;

&lt;p&gt;&lt;strong&gt;Machines and folders&lt;/strong&gt;&lt;/p&gt;

&lt;ul&gt;
&lt;li&gt;[ ] Unattended hosts use machine templates and Service Mode robots&lt;/li&gt;
&lt;li&gt;[ ] Hosts under a template share apps, paths, and versions&lt;/li&gt;
&lt;li&gt;[ ] Every robot account in a folder exists on every host in its templates&lt;/li&gt;
&lt;/ul&gt;

&lt;p&gt;&lt;strong&gt;Capacity&lt;/strong&gt;&lt;/p&gt;

&lt;ul&gt;
&lt;li&gt;[ ] Runtimes sized for peak overlap across all folders on a template&lt;/li&gt;
&lt;li&gt;[ ] Foreground slots (hosts) modeled separately from runtimes&lt;/li&gt;
&lt;/ul&gt;

&lt;p&gt;&lt;strong&gt;Queues&lt;/strong&gt;&lt;/p&gt;

&lt;ul&gt;
&lt;li&gt;[ ] One automated retry layer per failure type, documented&lt;/li&gt;
&lt;li&gt;[ ] Reporting counts business items by &lt;code&gt;Key&lt;/code&gt;, not by queue item&lt;/li&gt;
&lt;li&gt;[ ] First items in retry chains are never deleted&lt;/li&gt;
&lt;li&gt;[ ] Pending jobs strategy chosen deliberately&lt;/li&gt;
&lt;/ul&gt;

&lt;p&gt;&lt;strong&gt;Control and observability&lt;/strong&gt;&lt;/p&gt;

&lt;ul&gt;
&lt;li&gt;[ ] Processes honor Should Stop; status set only in Try and Catch&lt;/li&gt;
&lt;li&gt;[ ] Queue SLAs and Risk SLAs set&lt;/li&gt;
&lt;li&gt;[ ] Alerts on robot disconnected, trigger auto-disabled, job failed, queue SLA error&lt;/li&gt;
&lt;/ul&gt;

&lt;p&gt;&lt;strong&gt;Change&lt;/strong&gt;&lt;/p&gt;

&lt;ul&gt;
&lt;li&gt;[ ] Automation Ops policies for feeds and analyzer rules&lt;/li&gt;
&lt;li&gt;[ ] CI/CD with tests and approval&lt;/li&gt;
&lt;li&gt;[ ] At least one previous process version kept for rollback&lt;/li&gt;
&lt;li&gt;[ ] Browser and GPO changes tested on a non-production template first&lt;/li&gt;
&lt;/ul&gt;




&lt;p&gt;I post short UiPath fixes from real forum questions on my YouTube channel, &lt;a href="https://www.youtube.com/@rpasimplified" rel="noopener noreferrer"&gt;RPA Simplified&lt;/a&gt;.&lt;/p&gt;

</description>
      <category>uipath</category>
      <category>rpa</category>
      <category>automation</category>
      <category>architecture</category>
    </item>
    <item>
      <title>UiPath Pitfall #4: A Variable With the Same Name as Your Argument</title>
      <dc:creator>Pravinraj Vincent</dc:creator>
      <pubDate>Fri, 30 Oct 2026 00:56:01 +0000</pubDate>
      <link>https://dev.to/pravinraj/uipath-pitfall-4-a-variable-with-the-same-name-as-your-argument-3h4h</link>
      <guid>https://dev.to/pravinraj/uipath-pitfall-4-a-variable-with-the-same-name-as-your-argument-3h4h</guid>
      <description>&lt;p&gt;You assign a value to your Out argument, a Message Box right after says it's filled, and a few activities later (or back in the parent workflow) it's &lt;code&gt;Nothing&lt;/code&gt;. No error and no runtime warning. It's just empty.&lt;/p&gt;

&lt;p&gt;This came up again on the UiPath Forum in &lt;a href="https://forum.uipath.com/t/out-param-nothing-when-leaving-do/5786276" rel="noopener noreferrer"&gt;Out param Nothing when leaving Do&lt;/a&gt;, and it's one of those bugs that makes you doubt Studio itself.&lt;/p&gt;

&lt;h2&gt;
  
  
  The cause
&lt;/h2&gt;

&lt;p&gt;Most of the time, a variable and an argument share a name.&lt;/p&gt;

&lt;p&gt;You've got &lt;code&gt;out_Result&lt;/code&gt; in the Arguments panel, and somewhere along the way a variable called &lt;code&gt;out_Result&lt;/code&gt; also got created, often scoped to a Sequence or a Do block. Inside that scope, your Assign writes to the variable. Outside it, you're reading the argument, which never got touched.&lt;/p&gt;

&lt;p&gt;The Managing Arguments docs spell it out: if a variable and an argument have the same name, the variable is always the one used at runtime. The same goes for two variables in different scopes of one file. The inner scope wins.&lt;/p&gt;

&lt;h2&gt;
  
  
  How it sneaks in
&lt;/h2&gt;

&lt;p&gt;For me it's usually &lt;code&gt;Ctrl + K&lt;/code&gt; in the To field of an Assign. That shortcut creates a new &lt;strong&gt;variable&lt;/strong&gt;, not an argument. So if you type the argument's name and hit &lt;code&gt;Ctrl + K&lt;/code&gt;, you've just made a duplicate scoped to wherever you were.&lt;/p&gt;

&lt;p&gt;(&lt;code&gt;Ctrl + M&lt;/code&gt; is the one that creates an In argument.)&lt;/p&gt;

&lt;h2&gt;
  
  
  The fix
&lt;/h2&gt;

&lt;ul&gt;
&lt;li&gt;
&lt;strong&gt;Find the duplicate.&lt;/strong&gt; Select the container (click the Do, Sequence or Try block, not just the root) and check the Variables panel for the argument's name. Delete the variable, keep the argument, then re-pick the argument in the Assign.&lt;/li&gt;
&lt;li&gt;
&lt;strong&gt;Let Workflow Analyzer catch it.&lt;/strong&gt; Run Analyze File or Analyze Project from the Design ribbon. There are rules for exactly this:

&lt;ul&gt;
&lt;li&gt;
&lt;code&gt;ST-NMG-006&lt;/code&gt; Variable Overrides Argument&lt;/li&gt;
&lt;li&gt;
&lt;code&gt;ST-NMG-005&lt;/code&gt; Variable Overrides Variable&lt;/li&gt;
&lt;/ul&gt;
&lt;/li&gt;
&lt;/ul&gt;

&lt;p&gt;If you're on a team, set &lt;code&gt;ST-NMG-006&lt;/code&gt; to Error in the Workflow Analyzer settings. Then it shows up as an error in the Errors panel, which is a lot harder to scroll past.&lt;/p&gt;

&lt;ul&gt;
&lt;li&gt;
&lt;strong&gt;Use direction prefixes on arguments only.&lt;/strong&gt; &lt;code&gt;in_&lt;/code&gt;, &lt;code&gt;out_&lt;/code&gt;, &lt;code&gt;io_&lt;/code&gt; for arguments, never for variables. A variable called &lt;code&gt;out_Something&lt;/code&gt; then looks wrong the moment you see it.&lt;/li&gt;
&lt;li&gt;
&lt;strong&gt;Assign the Out argument once.&lt;/strong&gt; If the value gets built up across several scopes, work on a local variable and assign it to the Out argument once, at the end of the workflow:
&lt;/li&gt;
&lt;/ul&gt;

&lt;div class="highlight js-code-highlight"&gt;
&lt;pre class="highlight vb"&gt;&lt;code&gt;&lt;span class="c1"&gt;' inside the scopes, work on a local variable&lt;/span&gt;
&lt;span class="n"&gt;resultBuilder&lt;/span&gt; &lt;span class="o"&gt;=&lt;/span&gt; &lt;span class="n"&gt;resultBuilder&lt;/span&gt; &lt;span class="o"&gt;&amp;amp;&lt;/span&gt; &lt;span class="n"&gt;currentRow&lt;/span&gt;&lt;span class="p"&gt;(&lt;/span&gt;&lt;span class="s"&gt;"Name"&lt;/span&gt;&lt;span class="p"&gt;).&lt;/span&gt;&lt;span class="n"&gt;ToString&lt;/span&gt;

&lt;span class="c1"&gt;' last activity in the workflow&lt;/span&gt;
&lt;span class="n"&gt;out_Result&lt;/span&gt; &lt;span class="o"&gt;=&lt;/span&gt; &lt;span class="n"&gt;resultBuilder&lt;/span&gt;
&lt;/code&gt;&lt;/pre&gt;

&lt;/div&gt;



&lt;h2&gt;
  
  
  Docs
&lt;/h2&gt;

&lt;ul&gt;
&lt;li&gt;Managing arguments: &lt;a href="https://docs.uipath.com/studio/standalone/latest/user-guide/managing-arguments" rel="noopener noreferrer"&gt;https://docs.uipath.com/studio/standalone/latest/user-guide/managing-arguments&lt;/a&gt;
&lt;/li&gt;
&lt;li&gt;Managing variables: &lt;a href="https://docs.uipath.com/studio/standalone/latest/user-guide/managing-variables" rel="noopener noreferrer"&gt;https://docs.uipath.com/studio/standalone/latest/user-guide/managing-variables&lt;/a&gt;
&lt;/li&gt;
&lt;li&gt;ST-NMG-006 Variable Overrides Argument: &lt;a href="https://docs.uipath.com/studio/standalone/latest/user-guide/st-nmg-006" rel="noopener noreferrer"&gt;https://docs.uipath.com/studio/standalone/latest/user-guide/st-nmg-006&lt;/a&gt;
&lt;/li&gt;
&lt;li&gt;ST-NMG-005 Variable Overrides Variable: &lt;a href="https://docs.uipath.com/studio/standalone/latest/user-guide/st-nmg-005" rel="noopener noreferrer"&gt;https://docs.uipath.com/studio/standalone/latest/user-guide/st-nmg-005&lt;/a&gt;
&lt;/li&gt;
&lt;li&gt;About Workflow Analyzer: &lt;a href="https://docs.uipath.com/studio/standalone/latest/user-guide/about-workflow-analyzer" rel="noopener noreferrer"&gt;https://docs.uipath.com/studio/standalone/latest/user-guide/about-workflow-analyzer&lt;/a&gt;
&lt;/li&gt;
&lt;li&gt;Keyboard shortcuts: &lt;a href="https://docs.uipath.com/studio/standalone/latest/user-guide/keyboard-shortcuts" rel="noopener noreferrer"&gt;https://docs.uipath.com/studio/standalone/latest/user-guide/keyboard-shortcuts&lt;/a&gt;
&lt;/li&gt;
&lt;/ul&gt;

&lt;p&gt;Do you run Workflow Analyzer before every publish, or only when something breaks? Tell me in the original forum thread, that's where the discussion is: &lt;a href="https://forum.uipath.com/t/uipath-pitfall-of-the-week-4-a-variable-with-the-same-name-as-your-argument/5786751" rel="noopener noreferrer"&gt;https://forum.uipath.com/t/uipath-pitfall-of-the-week-4-a-variable-with-the-same-name-as-your-argument/5786751&lt;/a&gt;&lt;/p&gt;

&lt;p&gt;&lt;em&gt;Originally posted on the UiPath Forum: &lt;a href="https://forum.uipath.com/t/uipath-pitfall-of-the-week-4-a-variable-with-the-same-name-as-your-argument/5786751" rel="noopener noreferrer"&gt;https://forum.uipath.com/t/uipath-pitfall-of-the-week-4-a-variable-with-the-same-name-as-your-argument/5786751&lt;/a&gt;&lt;/em&gt;&lt;/p&gt;




&lt;p&gt;Prefer to watch it? Here is the video version: &lt;a href="https://youtu.be/iDeKOYjRn-s" rel="noopener noreferrer"&gt;UiPath Studio: Out Argument Is Nothing After Do + Agent LLM HTTP 403&lt;/a&gt;.&lt;/p&gt;

&lt;p&gt;I post short UiPath fix videos from real forum questions on my YouTube channel, &lt;a href="https://www.youtube.com/@rpasimplified" rel="noopener noreferrer"&gt;RPA Simplified&lt;/a&gt;.&lt;/p&gt;

</description>
      <category>uipath</category>
      <category>rpa</category>
      <category>studio</category>
      <category>automation</category>
    </item>
    <item>
      <title>UiPath Pitfall #3: Expecting Orchestrator to Retry a Business Exception</title>
      <dc:creator>Pravinraj Vincent</dc:creator>
      <pubDate>Fri, 30 Oct 2026 00:55:15 +0000</pubDate>
      <link>https://dev.to/pravinraj/uipath-pitfall-3-expecting-orchestrator-to-retry-a-business-exception-4f52</link>
      <guid>https://dev.to/pravinraj/uipath-pitfall-3-expecting-orchestrator-to-retry-a-business-exception-4f52</guid>
      <description>&lt;p&gt;Your failed queue item never got retried. Or the opposite: the robot hit the same broken item three times in a row. Almost every time, it comes down to which exception type the item failed with.&lt;/p&gt;

&lt;p&gt;This question shows up on the UiPath Forum every few weeks in one form or another, so it got its own Pitfall of the Week.&lt;/p&gt;

&lt;h2&gt;
  
  
  The mistake
&lt;/h2&gt;

&lt;p&gt;Treating Business and Application exceptions as two labels for the same thing, "failed". Orchestrator treats them very differently.&lt;/p&gt;

&lt;ul&gt;
&lt;li&gt;
&lt;strong&gt;Application exception&lt;/strong&gt; + Auto Retry turned on for the queue: Orchestrator retries the item, up to the queue's max number of retries.&lt;/li&gt;
&lt;li&gt;
&lt;strong&gt;Business exception&lt;/strong&gt;: Orchestrator doesn't retry it by default.&lt;/li&gt;
&lt;/ul&gt;

&lt;p&gt;That's straight from the Set Transaction Status docs. With &lt;code&gt;ErrorType&lt;/code&gt; set to Application, the transaction is retried according to the settings in Orchestrator. With Business, it isn't.&lt;/p&gt;

&lt;h2&gt;
  
  
  It bites in two directions
&lt;/h2&gt;

&lt;ol&gt;
&lt;li&gt;
&lt;strong&gt;A technical problem thrown as a &lt;code&gt;BusinessRuleException&lt;/code&gt;.&lt;/strong&gt; Say a page timeout or a frozen app ends up in the wrong catch. The item goes straight to Failed and never gets a second chance, even though a retry would probably have worked.&lt;/li&gt;
&lt;li&gt;
&lt;strong&gt;A data problem thrown as a plain &lt;code&gt;System.Exception&lt;/code&gt;.&lt;/strong&gt; Missing invoice number, invalid customer ID. The item gets retried, fails the exact same way, and burns runtime. If the failing step is a login with bad credentials, repeated retries can also get the account locked.&lt;/li&gt;
&lt;/ol&gt;

&lt;h2&gt;
  
  
  In REFramework, what you throw is the decision
&lt;/h2&gt;

&lt;p&gt;REFramework makes the split for you. In &lt;code&gt;SetTransactionStatus.xaml&lt;/code&gt;, anything thrown as &lt;code&gt;BusinessRuleException&lt;/code&gt; is marked Business, and everything else is marked Application. So the exception type you throw in &lt;code&gt;Process.xaml&lt;/code&gt; decides whether the item gets another shot.&lt;/p&gt;

&lt;p&gt;A data check in &lt;code&gt;Process.xaml&lt;/code&gt; can be as simple as this (VB):&lt;br&gt;
&lt;/p&gt;

&lt;div class="highlight js-code-highlight"&gt;
&lt;pre class="highlight vb"&gt;&lt;code&gt;&lt;span class="k"&gt;If&lt;/span&gt; &lt;span class="kt"&gt;String&lt;/span&gt;&lt;span class="p"&gt;.&lt;/span&gt;&lt;span class="n"&gt;IsNullOrWhiteSpace&lt;/span&gt;&lt;span class="p"&gt;(&lt;/span&gt;&lt;span class="n"&gt;in_TransactionItem&lt;/span&gt;&lt;span class="p"&gt;.&lt;/span&gt;&lt;span class="n"&gt;SpecificContent&lt;/span&gt;&lt;span class="p"&gt;(&lt;/span&gt;&lt;span class="s"&gt;"InvoiceNumber"&lt;/span&gt;&lt;span class="p"&gt;).&lt;/span&gt;&lt;span class="n"&gt;ToString&lt;/span&gt;&lt;span class="p"&gt;)&lt;/span&gt; &lt;span class="k"&gt;Then&lt;/span&gt;
    &lt;span class="k"&gt;Throw&lt;/span&gt; &lt;span class="k"&gt;New&lt;/span&gt; &lt;span class="n"&gt;BusinessRuleException&lt;/span&gt;&lt;span class="p"&gt;(&lt;/span&gt;&lt;span class="s"&gt;"Invoice number is missing"&lt;/span&gt;&lt;span class="p"&gt;)&lt;/span&gt;
&lt;span class="k"&gt;End&lt;/span&gt; &lt;span class="k"&gt;If&lt;/span&gt;
&lt;/code&gt;&lt;/pre&gt;

&lt;/div&gt;



&lt;p&gt;In Studio you'd normally do this with an If activity and a Throw activity, but the idea is the same.&lt;/p&gt;

&lt;h2&gt;
  
  
  What I do now
&lt;/h2&gt;

&lt;ul&gt;
&lt;li&gt;
&lt;strong&gt;Ask one question before throwing:&lt;/strong&gt; would running this exact item again, unchanged, have a real chance of working? Yes means Application. No (bad or missing data, a business rule the item fails) means &lt;code&gt;BusinessRuleException&lt;/code&gt;.&lt;/li&gt;
&lt;li&gt;
&lt;strong&gt;Validate input data early&lt;/strong&gt; in &lt;code&gt;Process.xaml&lt;/code&gt; and throw &lt;code&gt;BusinessRuleException&lt;/code&gt; with a clear message. That way the reason shows up in the transaction details instead of a stack trace.&lt;/li&gt;
&lt;li&gt;
&lt;strong&gt;Handle a shared login failure in the Init state&lt;/strong&gt; and stop the job, instead of failing every queue item one by one.&lt;/li&gt;
&lt;li&gt;
&lt;strong&gt;Let one layer own retries.&lt;/strong&gt; The note on &lt;code&gt;MaxRetryNumber&lt;/code&gt; in the REFramework &lt;code&gt;Config.xlsx&lt;/code&gt; says it must be 0 when you're working with Orchestrator queues. So leave it at 0 and set retries on the queue itself (Edit queue, Auto Retry section).&lt;/li&gt;
&lt;li&gt;
&lt;strong&gt;Retry a single flaky step in place.&lt;/strong&gt; If one step just needs another try right away, wrap only that step in a Retry Scope (&lt;code&gt;Number Of Retries&lt;/code&gt;, &lt;code&gt;Retry Interval&lt;/code&gt;) instead of failing the whole item.&lt;/li&gt;
&lt;/ul&gt;

&lt;h2&gt;
  
  
  Docs
&lt;/h2&gt;

&lt;ul&gt;
&lt;li&gt;Business exception vs Application exception: &lt;a href="https://docs.uipath.com/orchestrator/automation-cloud/latest/user-guide/business-exception-vs-application-exception" rel="noopener noreferrer"&gt;https://docs.uipath.com/orchestrator/automation-cloud/latest/user-guide/business-exception-vs-application-exception&lt;/a&gt;
&lt;/li&gt;
&lt;li&gt;Set Transaction Status: &lt;a href="https://docs.uipath.com/activities/other/latest/workflow/set-transaction-status" rel="noopener noreferrer"&gt;https://docs.uipath.com/activities/other/latest/workflow/set-transaction-status&lt;/a&gt;
&lt;/li&gt;
&lt;li&gt;Managing queues in Orchestrator (Auto Retry): &lt;a href="https://docs.uipath.com/orchestrator/automation-cloud/latest/user-guide/managing-queues-in-orchestrator" rel="noopener noreferrer"&gt;https://docs.uipath.com/orchestrator/automation-cloud/latest/user-guide/managing-queues-in-orchestrator&lt;/a&gt;
&lt;/li&gt;
&lt;li&gt;Retry Scope: &lt;a href="https://docs.uipath.com/activities/other/latest/workflow/retry-scope" rel="noopener noreferrer"&gt;https://docs.uipath.com/activities/other/latest/workflow/retry-scope&lt;/a&gt;
&lt;/li&gt;
&lt;/ul&gt;

&lt;p&gt;Where do you draw the line on the gray areas, like a CAPTCHA, an expired password, or a target system down for maintenance? Join the discussion in the original forum thread: &lt;a href="https://forum.uipath.com/t/uipath-pitfall-of-the-week-3-expecting-orchestrator-to-retry-a-business-exception/5785256" rel="noopener noreferrer"&gt;https://forum.uipath.com/t/uipath-pitfall-of-the-week-3-expecting-orchestrator-to-retry-a-business-exception/5785256&lt;/a&gt;&lt;/p&gt;

&lt;p&gt;&lt;em&gt;Originally posted on the UiPath Forum: &lt;a href="https://forum.uipath.com/t/uipath-pitfall-of-the-week-3-expecting-orchestrator-to-retry-a-business-exception/5785256" rel="noopener noreferrer"&gt;https://forum.uipath.com/t/uipath-pitfall-of-the-week-3-expecting-orchestrator-to-retry-a-business-exception/5785256&lt;/a&gt;&lt;/em&gt;&lt;/p&gt;




&lt;p&gt;I post short UiPath fix videos from real forum questions on my YouTube channel, &lt;a href="https://www.youtube.com/@rpasimplified" rel="noopener noreferrer"&gt;RPA Simplified&lt;/a&gt;.&lt;/p&gt;

</description>
      <category>uipath</category>
      <category>rpa</category>
      <category>orchestrator</category>
      <category>automation</category>
    </item>
    <item>
      <title>UiPath Pitfall #2: Keeping Your Studio Project in a OneDrive Sync Folder</title>
      <dc:creator>Pravinraj Vincent</dc:creator>
      <pubDate>Fri, 30 Oct 2026 00:53:46 +0000</pubDate>
      <link>https://dev.to/pravinraj/uipath-pitfall-2-keeping-your-studio-project-in-a-onedrive-sync-folder-16fe</link>
      <guid>https://dev.to/pravinraj/uipath-pitfall-2-keeping-your-studio-project-in-a-onedrive-sync-folder-16fe</guid>
      <description>&lt;p&gt;If you're getting vague compile errors like a bare &lt;strong&gt;Error code 1&lt;/strong&gt;, or a project runs on one PC and won't open cleanly on another, check where the project folder lives. If it's inside OneDrive or any always-sync cloud folder, that's very likely your problem.&lt;/p&gt;

&lt;h2&gt;
  
  
  The mistake
&lt;/h2&gt;

&lt;p&gt;Putting the live Studio project folder inside OneDrive (or Dropbox, or any folder that syncs all the time) so you can "share" it between machines. It feels convenient. It isn't.&lt;/p&gt;

&lt;h2&gt;
  
  
  Why it bites
&lt;/h2&gt;

&lt;p&gt;Studio writes a lot of files while you work: &lt;code&gt;project.json&lt;/code&gt;, generated files, and a local cache in the hidden &lt;code&gt;.local&lt;/code&gt; folder. A sync client can lock those files, leave some of them online-only, or still be downloading subfolders when Studio tries to compile.&lt;/p&gt;

&lt;p&gt;On the forum that usually shows up as a bare Error code 1, or a project that works on PC1 and won't open on PC2 until someone copies it out of OneDrive. Network and shared drives cause the same kind of pain. Copy the project to a real local disk and it suddenly works.&lt;/p&gt;

&lt;h2&gt;
  
  
  The fix
&lt;/h2&gt;

&lt;p&gt;&lt;strong&gt;1. Keep the working project on a local path.&lt;/strong&gt;&lt;/p&gt;

&lt;p&gt;Studio's default location is:&lt;br&gt;
&lt;/p&gt;

&lt;div class="highlight js-code-highlight"&gt;
&lt;pre class="highlight plaintext"&gt;&lt;code&gt;%USERPROFILE%\Documents\UiPath
&lt;/code&gt;&lt;/pre&gt;

&lt;/div&gt;



&lt;p&gt;But watch out: if Windows has redirected Documents into OneDrive, that default is a synced folder too. In that case put your projects somewhere that isn't synced, for example:&lt;br&gt;
&lt;/p&gt;

&lt;div class="highlight js-code-highlight"&gt;
&lt;pre class="highlight plaintext"&gt;&lt;code&gt;C:\UiPathProjects\
&lt;/code&gt;&lt;/pre&gt;

&lt;/div&gt;



&lt;p&gt;&lt;strong&gt;2. Share and version with Git, not by syncing the live folder.&lt;/strong&gt;&lt;/p&gt;

&lt;p&gt;In Studio, go to the &lt;strong&gt;Team&lt;/strong&gt; tab and use &lt;strong&gt;GIT Init&lt;/strong&gt; to add the project to a local Git repository, commit, then push to a remote. TFS and SVN are also supported if that's what your team uses. Studio leaves the &lt;code&gt;.local&lt;/code&gt; folder out of source control by default, so your machine cache stays off the repo.&lt;/p&gt;

&lt;p&gt;&lt;strong&gt;3. Need a cloud copy just for backup?&lt;/strong&gt;&lt;/p&gt;

&lt;p&gt;Zip or copy the project after you close Studio. Just don't develop inside the synced folder.&lt;/p&gt;

&lt;p&gt;&lt;strong&gt;4. Don't confuse this with Sync to Cloud.&lt;/strong&gt;&lt;/p&gt;

&lt;p&gt;Studio's &lt;strong&gt;Sync to Cloud&lt;/strong&gt; option syncs your project with Studio Web in Automation Cloud. It has nothing to do with OneDrive.&lt;/p&gt;

&lt;h2&gt;
  
  
  Quick check
&lt;/h2&gt;

&lt;p&gt;Open File Explorer on your project path. If you see OneDrive sync icons on &lt;code&gt;project.json&lt;/code&gt; or your workflow files, move the project to a local folder before your next publish.&lt;/p&gt;

&lt;h2&gt;
  
  
  Docs
&lt;/h2&gt;

&lt;ul&gt;
&lt;li&gt;About automation projects (default location, &lt;code&gt;.local&lt;/code&gt;, Sync to Cloud): &lt;a href="https://docs.uipath.com/studio/standalone/latest/user-guide/about-automation-projects" rel="noopener noreferrer"&gt;https://docs.uipath.com/studio/standalone/latest/user-guide/about-automation-projects&lt;/a&gt;
&lt;/li&gt;
&lt;li&gt;Adding a project to Git: &lt;a href="https://docs.uipath.com/studio/standalone/latest/user-guide/adding-a-project-to-git" rel="noopener noreferrer"&gt;https://docs.uipath.com/studio/standalone/latest/user-guide/adding-a-project-to-git&lt;/a&gt;
&lt;/li&gt;
&lt;li&gt;About version control: &lt;a href="https://docs.uipath.com/studio/standalone/latest/user-guide/about-version-control" rel="noopener noreferrer"&gt;https://docs.uipath.com/studio/standalone/latest/user-guide/about-version-control&lt;/a&gt;
&lt;/li&gt;
&lt;/ul&gt;

&lt;p&gt;Had a sync folder bite you? Tell me about it in the original forum thread, that's where the discussion is: &lt;a href="https://forum.uipath.com/t/uipath-pitfall-of-the-week-2-keeping-your-studio-project-in-a-onedrive-sync-folder/5783803" rel="noopener noreferrer"&gt;https://forum.uipath.com/t/uipath-pitfall-of-the-week-2-keeping-your-studio-project-in-a-onedrive-sync-folder/5783803&lt;/a&gt;&lt;/p&gt;

&lt;p&gt;&lt;em&gt;Originally posted on the UiPath Forum: &lt;a href="https://forum.uipath.com/t/uipath-pitfall-of-the-week-2-keeping-your-studio-project-in-a-onedrive-sync-folder/5783803" rel="noopener noreferrer"&gt;https://forum.uipath.com/t/uipath-pitfall-of-the-week-2-keeping-your-studio-project-in-a-onedrive-sync-folder/5783803&lt;/a&gt;&lt;/em&gt;&lt;/p&gt;




&lt;p&gt;I post short UiPath fix videos from real forum questions on my YouTube channel, &lt;a href="https://www.youtube.com/@rpasimplified" rel="noopener noreferrer"&gt;RPA Simplified&lt;/a&gt;.&lt;/p&gt;

</description>
      <category>uipath</category>
      <category>rpa</category>
      <category>automation</category>
      <category>studio</category>
    </item>
    <item>
      <title>UiPath Pitfall #1: Relying on idx in Your Selectors</title>
      <dc:creator>Pravinraj Vincent</dc:creator>
      <pubDate>Fri, 30 Oct 2026 00:53:37 +0000</pubDate>
      <link>https://dev.to/pravinraj/uipath-pitfall-1-relying-on-idx-in-your-selectors-3pih</link>
      <guid>https://dev.to/pravinraj/uipath-pitfall-1-relying-on-idx-in-your-selectors-3pih</guid>
      <description>&lt;p&gt;If your selector works on some runs and fails on others, open it and look for an &lt;code&gt;idx&lt;/code&gt; attribute. Nine times out of ten, that's the problem.&lt;/p&gt;

&lt;p&gt;I see this one on the UiPath Forum constantly, so it was the first thing I wrote up in my Pitfall of the Week series.&lt;/p&gt;

&lt;h2&gt;
  
  
  Why idx breaks
&lt;/h2&gt;

&lt;p&gt;When you indicate an element and Studio can't tell it apart from similar elements, it adds something like &lt;code&gt;idx='3'&lt;/code&gt; to the selector. That works on your screen today. But &lt;code&gt;idx&lt;/code&gt; is just the element's position in the UI hierarchy, nothing more.&lt;/p&gt;

&lt;p&gt;Add a column to a table, get a popup, or have the page load a little differently, and position 3 is now a different element. Or nothing at all. The robot either clicks the wrong thing or times out, and it looks random because it depends on what the screen looked like on that run.&lt;/p&gt;

&lt;p&gt;UiPath's own Workflow Analyzer rule for this, UI-REL-001 (Large Idx in Selectors), says the same thing: the order &lt;code&gt;idx&lt;/code&gt; relies on can change when the elements you're automating change, so it should be avoided.&lt;/p&gt;

&lt;h2&gt;
  
  
  The fix
&lt;/h2&gt;

&lt;p&gt;Open the selector in UI Explorer and replace &lt;code&gt;idx&lt;/code&gt; with something that actually describes the element. Good candidates:&lt;/p&gt;

&lt;ul&gt;
&lt;li&gt;&lt;code&gt;aaname&lt;/code&gt;&lt;/li&gt;
&lt;li&gt;&lt;code&gt;id&lt;/code&gt;&lt;/li&gt;
&lt;li&gt;&lt;code&gt;automationid&lt;/code&gt;&lt;/li&gt;
&lt;li&gt;&lt;code&gt;innertext&lt;/code&gt;&lt;/li&gt;
&lt;/ul&gt;

&lt;p&gt;So instead of this:&lt;br&gt;
&lt;/p&gt;

&lt;div class="highlight js-code-highlight"&gt;
&lt;pre class="highlight xml"&gt;&lt;code&gt;&lt;span class="nt"&gt;&amp;lt;webctrl&lt;/span&gt; &lt;span class="na"&gt;tag=&lt;/span&gt;&lt;span class="s"&gt;'BUTTON'&lt;/span&gt; &lt;span class="na"&gt;idx=&lt;/span&gt;&lt;span class="s"&gt;'3'&lt;/span&gt; &lt;span class="nt"&gt;/&amp;gt;&lt;/span&gt;
&lt;/code&gt;&lt;/pre&gt;

&lt;/div&gt;



&lt;p&gt;you'd aim for something like this:&lt;br&gt;
&lt;/p&gt;

&lt;div class="highlight js-code-highlight"&gt;
&lt;pre class="highlight xml"&gt;&lt;code&gt;&lt;span class="nt"&gt;&amp;lt;webctrl&lt;/span&gt; &lt;span class="na"&gt;tag=&lt;/span&gt;&lt;span class="s"&gt;'BUTTON'&lt;/span&gt; &lt;span class="na"&gt;aaname=&lt;/span&gt;&lt;span class="s"&gt;'Submit invoice'&lt;/span&gt; &lt;span class="nt"&gt;/&amp;gt;&lt;/span&gt;
&lt;/code&gt;&lt;/pre&gt;

&lt;/div&gt;



&lt;p&gt;If the stable value changes a bit between runs, use a wildcard. &lt;code&gt;*&lt;/code&gt; replaces zero or more characters:&lt;br&gt;
&lt;/p&gt;

&lt;div class="highlight js-code-highlight"&gt;
&lt;pre class="highlight xml"&gt;&lt;code&gt;&lt;span class="nt"&gt;&amp;lt;webctrl&lt;/span&gt; &lt;span class="na"&gt;tag=&lt;/span&gt;&lt;span class="s"&gt;'SPAN'&lt;/span&gt; &lt;span class="na"&gt;aaname=&lt;/span&gt;&lt;span class="s"&gt;'Invoice *'&lt;/span&gt; &lt;span class="nt"&gt;/&amp;gt;&lt;/span&gt;
&lt;/code&gt;&lt;/pre&gt;

&lt;/div&gt;



&lt;p&gt;In Modern activities, add an anchor, meaning a nearby label that never moves, so the target is found relative to something stable. Then keep Fuzzy selector or Image as a fallback targeting method in case the strict selector misses.&lt;/p&gt;

&lt;h2&gt;
  
  
  A quick check before you publish
&lt;/h2&gt;

&lt;p&gt;Search your project for &lt;code&gt;idx=&lt;/code&gt;. Every hit is a possible failure in production.&lt;/p&gt;

&lt;p&gt;You can also let Workflow Analyzer do it for you. UI-REL-001 lists every selector with an &lt;code&gt;idx&lt;/code&gt; value above the threshold you set, one by one, so you can work through them before they bite.&lt;/p&gt;

&lt;h2&gt;
  
  
  Docs
&lt;/h2&gt;

&lt;ul&gt;
&lt;li&gt;UI-REL-001 Large Idx in Selectors: &lt;a href="https://docs.uipath.com/activities/other/latest/ui-automation/ui-rel-001" rel="noopener noreferrer"&gt;https://docs.uipath.com/activities/other/latest/ui-automation/ui-rel-001&lt;/a&gt;
&lt;/li&gt;
&lt;li&gt;Selectors with wildcards: &lt;a href="https://docs.uipath.com/activities/other/latest/ui-automation/selectors-with-wildcards" rel="noopener noreferrer"&gt;https://docs.uipath.com/activities/other/latest/ui-automation/selectors-with-wildcards&lt;/a&gt;
&lt;/li&gt;
&lt;li&gt;UI Explorer: &lt;a href="https://docs.uipath.com/activities/other/latest/ui-automation/ui-explorer" rel="noopener noreferrer"&gt;https://docs.uipath.com/activities/other/latest/ui-automation/ui-explorer&lt;/a&gt;
&lt;/li&gt;
&lt;li&gt;Advanced descriptor configuration (strict, fuzzy and image targeting): &lt;a href="https://docs.uipath.com/activities/other/latest/ui-automation/advanced-descriptor-configuration" rel="noopener noreferrer"&gt;https://docs.uipath.com/activities/other/latest/ui-automation/advanced-descriptor-configuration&lt;/a&gt;
&lt;/li&gt;
&lt;/ul&gt;

&lt;p&gt;Got a selector horror story? Share it in the original forum thread, that's where the discussion lives: &lt;a href="https://forum.uipath.com/t/uipath-pitfall-of-the-week-1-relying-on-idx-in-your-selectors/5783645" rel="noopener noreferrer"&gt;https://forum.uipath.com/t/uipath-pitfall-of-the-week-1-relying-on-idx-in-your-selectors/5783645&lt;/a&gt;&lt;/p&gt;

&lt;p&gt;&lt;em&gt;Originally posted on the UiPath Forum: &lt;a href="https://forum.uipath.com/t/uipath-pitfall-of-the-week-1-relying-on-idx-in-your-selectors/5783645" rel="noopener noreferrer"&gt;https://forum.uipath.com/t/uipath-pitfall-of-the-week-1-relying-on-idx-in-your-selectors/5783645&lt;/a&gt;&lt;/em&gt;&lt;/p&gt;




&lt;p&gt;I post short UiPath fix videos from real forum questions on my YouTube channel, &lt;a href="https://www.youtube.com/@rpasimplified" rel="noopener noreferrer"&gt;RPA Simplified&lt;/a&gt;.&lt;/p&gt;

</description>
      <category>uipath</category>
      <category>rpa</category>
      <category>automation</category>
      <category>selectors</category>
    </item>
  </channel>
</rss>
