DBMS_SCHEDULER – Fiddling with File Watchers

I recently updated my blog profile picture. Let’s face it, it was about time. The old one was taken 14 years ago.
This time, I thought I’d go for a contemporary portrait, which has the added benefit of being less likely to frighten animals and small children.
Unfortunately, it is also crying out for a caption :

Just popping into the bank. Keep the engine running

One or two alternatives captions may well crop up in the following paragraphs.
Primarily though, I’ll be talking about the file watcher capabilities of DBMS_SCHEDULER.
Introduced in 11g, it enables developers to trigger actions when a file arrives without having to leave the comfort of their database.
There are a few moving parts and getting them to work together can take a bit of, well, fiddling.
Specifically, we’ll be looking at :

• configuring the frequency at which filewatchers are run
• creating a credential to use for file watching
• creating a file watcher to determine when a file has arrived
• creating a job for the file watcher to execute
• adding a program so that we can get some information about the file
• getting the filewatcher to wait for a large file transfer to complete

Before we get into all of that, we need to have a quick chat about environments…

Time flies when you’re having fun – The truth about Oracle Exception handlers (this week)

Oracle Twitter is a bit like Poker – you tend to take notice when Aces turn up.
I mention this because I’m indebted to Jacek Gebal, Sabine Heismath and Andy Sayer, whose discussion on this topic alerted me to the fact that PL/SQL exception handling had changed in a small but important way whilst I wasn’t looking.

Just in case you’re wondering whether that opening paragraph is a prelude to a post containing gratuitous references to a certain Motorhead song, all I can say is…if you like to gamble…

Making the most of Oracle 18c XE – Pluggable Databases and the Oracle Eco-System

This was going to be a simple post about creating multiple Pluggable Databases (PDBs) in Oracle 18cXE.
But you know what it’s like, you get your lego database set out and then you spot those unusual shaped bricks… and the transparent ones… oh, that one has a wheel on…and get a bit carried away.
What follows is a guide on how to create three concurrent PDBs on an Oracle 18cXE database. However, I have taken the opportunity to investigate other areas of the Express Edition platform and so will be making use of Advanced Compression as well as Oracle Enterprise Manager Express.
In fact, I wouldn’t be surprised if SQLDeveloper put in an appearance as well.
I’ve also included a couple of the more notable members of Oracle’s small-but-perfectly-formed eco-system into the final design. On top of all that, I’ll be performing a magic trick to ensure that I only have to install each piece of software once, even though it may end up in multiple PDBs…

Running a shell script from inside an Oracle database using DBMS_SCHEDULER

As I write, various parts of the United Kingdom are still under variations of the theme of “lockdown”. Not that I’m stir crazy or anything but I’ve now decided it’s time to explore one way of escaping, from the confinements of my database, at least.

Specifically I’m going to :

• create an OS user on the database server
• create an OS group to allow both the new user and the oracle user to access to a common location
• create a shell script owned by the OS user which writes to that common location
• create a credential to access that user from inside the database itself
• setup and run a scheduler job to execute a shell script as the new OS user
• read the output file generated by the shell script from inside the database

For this exercise, I’m using my trusty 18cXE database running on CentOS…

No Data ? No Problem ! Fun and Games with the NO_DATA_FOUND exception

I can’t remember how long we’ve been in lockdown now but between you and me, I think my Oracle Database may be cracking up.
As Exhibit A, I present the fact that RDBMS seems to raise the NO_DATA_FOUND exception only when it feels like it…

To Boldly gzip from PL/SQL

Lempel-Ziv. Sounds like something you might find on a starship. I mean, you can just imagine Scotty telling Kirk that they can’t go to warp because the Lempel-Ziv compressors are offline.
In fact, the Lempel-Ziv algorithm is what GNU’s Gzip program uses to compress files.
Fortunately, the UTL_COMPRESS package employs the same algorithm, which means that you can create archives using one tool that can be manipulated by the other…

Reading an OS file from inside an Oracle Database using an External Table

Two weeks into lockdown and some are looking to break the monotony by varying their diet…

For my part, I’ve decided to amuse myself by making a couple of “Top five” lists – the five best things about the lockdown and five things I’ve now finally got time to watch.

These lists are in text files are somewhere on my Database server and therefore provide the basis for the examples that follow.

When using Oracle, if you find yourself in a situation where you don’t have access to the Database Server file system, or would just rather use SQL to find what you’re looking for in a data set, you’re in luck.
The technique that used to be a neat way of viewing the database alert log from inside the database can now be re-purposed to look at any text file that’s located in a directory referenced by a Directory Object in the database.

What we’re going to look at here is :

• Setting up an external table for reading from files
• How to point the external table specific files and directories using EXTERNAL MODIFY
• How achieve the same thing in 11gR2 databases with the help of DBMS_LOCK

First of all though, just in case you’re wondering….

Generating CSV files from PL/SQL the Very Easy Way.

This post is dedicated to Morten Braten and William Robertson.
Thanks to both of you for saving me a lot of time (and typing)…

XML, YAML, JSON. When it comes to transferring data between systems, they’ve all been flavour of the month at one time or another. However, good old CSV is still the format of choice when the batch window opens.
Not only is it simple, it adds comparatively little overhead to the size of the data file being generated.

OK – CSV used to mean “Comma-delimited” but these days it’s become synonymous with delimited data.
Whatever separator character you use, generating CSV is considerably easier these days…unless you’re using PL/SQL in a batch (as opposed to interactive) process.
That said, Connor MacDonald does have a clever way of utilising the functionality available in a command line tool such as SQL*Plus by means of a DBMS_SCHEDULER job.

If for some reason that won’t work for you, it looks like you’re going to have to hack out some custom PL/SQL to get the job done…or you could have a look at a couple of the options available in what other technologies would refer to as “the ecosystem”.
What I’m going to cover is :

The pain of hand-cranking delimited SQL

A solution offered in Morten Braten’s Alexandria PL/SQL Utility Library

An alternative solution made available by William Robertson

The final PL/SQL procedure

Running a “background” job in PL/SQL

Teddy has observed the recent General Election campaign with some interest and has concluded that he has what it takes to be the next Prime Minister.

It’s not just the haircut, which does now look extremely Prime Ministerial…

Politics is a robust business but Teddy’s more than capable of playing “ruff” :

He firmly believes in the need to streamline Government at Cabinet level, which has the incumbent Chief Mouser to the Cabinet Office a little nervous.
He’s also well used to being followed around by a “pooper scooper”. And not to put too fine a point on it, there’s more than one reason that he’s known as a “shaggy” dog.

If he’s going to make it in politics, Teddy knows that he doesn’t have time to waste waiting for that pesky long-running job he’s just started. Oh no, he needs to use his evenings to get on with building his power base.
Fortunately, Oracle facilitates detached execution of PL/SQL blocks by means of the DBMS_SCHEDULER package. Now, I know what you’re thinking, that’s going to involve a lot of fiddly setup with schedules and windows and chains and stuff. Well, you may be pleasantly surprised…

Easy Listening with Datapump in the SQLDeveloper DBA Module

There are a number of ways to transfer data between Oracle Databases, one of which is to use the PL/SQL Datapump API – DBMS_DATAPUMP.
If you wish to avail yourself of this utility but find the syntax a bit fiddly, you always have the option of getting SQLDeveloper to do (most of) it for you.
What we’re talking about here is how to persuade the SQLDeveloper DB module to :

• Create and execute a custom Datapump export job
• do most of the work creating an import of a subset of the exported data

…