Bug 225496

Summary: CVE-2007-0555 PostgreSQL arbitrary memory read flaws (CVE-2007-0556)
Product: Red Hat Enterprise Linux 5 Reporter: Josh Bressers <bressers>
Component: postgresqlAssignee: Tom Lane <tgl>
Status: CLOSED ERRATA QA Contact: David Lawrence <dkl>
Severity: medium Docs Contact:
Priority: medium    
Version: 5.0CC: hhorak, security-response-team
Target Milestone: ---Keywords: Security
Target Release: ---   
Hardware: All   
OS: Linux   
Whiteboard: reported=20070129,source=redhat,impact=moderate,embargo=20070205
Fixed In Version: RHSA-2007-0068 Doc Type: Bug Fix
Doc Text:
Story Points: ---
Clone Of: Environment:
Last Closed: 2007-03-14 15:02:19 UTC Type: ---
Regression: --- Mount Type: ---
Documentation: --- CRM:
Verified Versions: Category: ---
oVirt Team: --- RHEL 7.3 requirements from Atomic Host:
Cloudforms Team: --- Target Upstream Version:
Embargoed:
Bug Depends On: 227688    
Bug Blocks:    

Description Josh Bressers 2007-01-30 22:09:33 UTC 
+++ This bug was initially created as a clone of Bug #225493 +++

An authenticated PostgreSQL user has the ability to crash the database server or
possibly read arbitrary memory for the server process.  This is caused by
insufficient type checking for SQL-language functions.

CVE-2006-5556 also describes a similar flaw.  The description form the
PostgreSQL advisory describes it as such:

The risk scenarios are exactly the same as above, but the method to
exploit the hole is a bit different. The attacker must cause a query plan
to be prepared and saved (via PREPARE, or implicitly in a plpgsql
function) and then execute an ALTER COLUMN TYPE command to change the type
of one of the columns used in the query, and then execute the now-stale
query plan. Since ALTER COLUMN TYPE was introduced in PostgreSQL 8.0,
older versions are not vulnerable.
Comment 1 Mark J. Cox 2007-02-05 11:31:15 UTC 
removing embargo; public at http://www.postgresql.org/support/security.html
Comment 2 Tom Lane 2007-02-05 14:54:12 UTC 
Fix is built in postgresql-8.1.7-1.el5, awaiting RHEL5 0-day updates.
Comment 3 Tom Lane 2007-02-06 23:33:20 UTC 
It turns out that the security fix in 8.1.7 broke enough things that it's going
to have to be modified.  I'm withdrawing postgresql-8.1.7-1.el5 as a candidate
for RHEL5 0-day.  Upstream is going to spin an 8.1.8 with the fix shortly
(probably tonight) and I recommend we go with that instead.
Comment 9 Red Hat Bugzilla 2007-03-14 15:02:19 UTC 
An advisory has been issued which should help the problem
described in this bug report. This report is therefore being
closed with a resolution of ERRATA. For more information
on the solution and/or where to find the updated files,
please follow the link below. You may reopen this bug report
if the solution does not work for you.

http://rhn.redhat.com/errata/RHSA-2007-0068.html