<?xml version="1.0"?>
<feed xmlns="http://www.w3.org/2005/Atom">
<title>Comments for Known Causes of Corruption in InterBase Databases</title>
<link rel="alternate" type="text/plain" href="http://edn.embarcadero.com/article/29515" title="Known Causes of Corruption in InterBase Databases" />
<link rel="self" type="application/atom+xml" href="http://edn.embarcadero.com/article/29515/feed" title="Comments for Known Causes of Corruption in InterBase Databases" />
<id>http://edn.embarcadero.com/article/29515</id>
<updated>2009-07-09T13:42:49-07:00</updated>
<entry>
<title>Known Causes of Corruption in InterBase Databases</title>
<author>
<name>Dmitri Kouzmenko</name>
<uri>http://threads.embarcadero.com/threads/threads.exe/userall?commentid=36336</uri>
</author>
<id>http://threads.embarcadero.com/threads/threads.exe/view?commentid=36336</id>
<updated>2004-03-14T03:47:48-07:00</updated>
<published>2004-03-14T03:47:48-07:00</published>
<summary>Known Causes of Corruption in InterBase Databases</summary>
<content>Server crush during &quot;Changing metadata while database in use&quot; is well known since SuperServer architecture was introdiced (IB 4.2 for Windows).Yes, itself it can't corrupt database, but if it crushes server, it can cause database corruption, since server is being terminated by itself during work.If you know, InterBase 5.5 introduced mechanism that described in Data Definition's Guide &quot;altering and dropping procedures in use&quot; section. This was exactly made to prevent server crush during metadata change.Of course, I do not recommend to change metadata at production server while it serves user requests, too.</content>
</entry>
</feed>
