<?xml version="1.0" encoding="utf-8" standalone="yes"?><rss version="2.0" xmlns:atom="http://www.w3.org/2005/Atom"><channel><title>Shell on Oracle Scripts</title><link>https://www.oraclescripts.com/tags/shell/</link><description>Recent content in Shell on Oracle Scripts</description><generator>Hugo -- gohugo.io</generator><language>en</language><copyright>OracleScripts.com</copyright><lastBuildDate>Mon, 25 May 2026 00:00:00 +0000</lastBuildDate><atom:link href="https://www.oraclescripts.com/tags/shell/index.xml" rel="self" type="application/rss+xml"/><item><title>Search UNIX Man Pages for Oracle DBA Tools with man -k</title><link>https://www.oraclescripts.com/post/oracle-dba-man-page-search-keyword/</link><pubDate>Mon, 25 May 2026 00:00:00 +0000</pubDate><guid>https://www.oraclescripts.com/post/oracle-dba-man-page-search-keyword/</guid><description>
&lt;h2 id="search-unix-man-pages-for-oracle-dba-tools-with-man--k"&gt;Search UNIX Man Pages for Oracle DBA Tools with man -k&lt;/h2&gt;
&lt;h2 id="purpose"&gt;Purpose&lt;/h2&gt;
&lt;p&gt;Where Google answers most &amp;quot;how do I do X in UNIX&amp;quot; questions for a DBA today, &lt;code&gt;man -k&lt;/code&gt; answers the narrower one — what tools the host actually has installed, and what each one does, without leaving the SSH session. Production Oracle hosts often run in change-controlled networks with no outbound HTTP, no browser, and a tight maintenance window; &lt;code&gt;man -k&lt;/code&gt; and its synonym &lt;code&gt;apropos&lt;/code&gt; are the offline discovery commands that close the gap between knowing a vague capability exists and finding the binary that delivers it.&lt;/p&gt;</description></item><item><title>Enable vi Command-Line Editing in ksh for Oracle DBA Shells</title><link>https://www.oraclescripts.com/post/ksh-vi-mode-shell-history-dba/</link><pubDate>Sun, 24 May 2026 00:00:00 +0000</pubDate><guid>https://www.oraclescripts.com/post/ksh-vi-mode-shell-history-dba/</guid><description>
&lt;h2 id="enable-vi-command-line-editing-in-ksh-for-oracle-dba-shells"&gt;Enable vi Command-Line Editing in ksh for Oracle DBA Shells&lt;/h2&gt;
&lt;h2 id="purpose"&gt;Purpose&lt;/h2&gt;
&lt;p&gt;Recalling and editing the prior command without retyping it is the single most-used operation in a long DBA shift on an Oracle host. On the Korn shell (&lt;code&gt;ksh&lt;/code&gt;) — the default login shell on AIX 7.x, Solaris 11, HP-UX 11i, and many older Oracle 11g/12c reference builds — that operation is gated behind one line in &lt;code&gt;.profile&lt;/code&gt;: &lt;code&gt;set -o vi&lt;/code&gt;. Without it, the up-arrow key prints a literal escape sequence and the only way to repeat a long &lt;code&gt;RMAN BACKUP DATABASE PLUS ARCHIVELOG ...&lt;/code&gt; invocation is to type it again.&lt;/p&gt;</description></item><item><title>Disable vi Autoindent Before Pasting SQL Scripts on Oracle Hosts</title><link>https://www.oraclescripts.com/post/vi-disable-autoindent-pasting-sql/</link><pubDate>Sat, 23 May 2026 00:00:00 +0000</pubDate><guid>https://www.oraclescripts.com/post/vi-disable-autoindent-pasting-sql/</guid><description>
&lt;h2 id="disable-vi-autoindent-before-pasting-sql-scripts-on-oracle-hosts"&gt;Disable vi Autoindent Before Pasting SQL Scripts on Oracle Hosts&lt;/h2&gt;
&lt;h2 id="purpose"&gt;Purpose&lt;/h2&gt;
&lt;p&gt;Pasting a SQL script copied from email, a Confluence page, or a ticket comment into a vi buffer over SSH — and watching every line shift one indent right of the previous one — is the most common and most frustrating vi failure mode an Oracle DBA encounters. The script that ran fine when executed directly from the wiki snippet now fails on the host with &lt;code&gt;SP2-0042: unknown command &amp;quot; INSERT&amp;quot;&lt;/code&gt; on every line. The cause is &lt;code&gt;autoindent&lt;/code&gt;: vi assumed each new line should match the indentation of the previous one, and the terminal cannot distinguish a real keypress from a pasted character, so the indent compounds.&lt;/p&gt;</description></item></channel></rss>