Bug 136300

Summary: CAN-2004-0977 temporary file vulnerabilities in make_oidjoins_check script
Product: Red Hat Enterprise Linux 3 Reporter: Mark J. Cox <mjc>
Component: rh-postgresqlAssignee: Tom Lane <tgl>
Status: CLOSED ERRATA QA Contact:
Severity: low Docs Contact:
Priority: medium    
Version: 3.0CC: hhorak
Target Milestone: ---Keywords: Security
Target Release: ---   
Hardware: All   
OS: Linux   
Whiteboard: public=20040930,impact=low
Fixed In Version: Doc Type: Bug Fix
Doc Text:
Story Points: ---
Clone Of: Environment:
Last Closed: 2004-12-20 17:54:14 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:
Attachments:
Description Flags
tempfile patch from vendor-sec none

Description Mark J. Cox 2004-10-19 09:42:08 UTC 
On September 10th 2004, Trustix shared some temporary file
vulnerabilities with vendor-sec.  After some refinement these were
made public on Sep30.  These are minor issues (impact: LOW) and
therefore should be fixed in future updates, but don't deserve their
own security advisory.

Temporary file vulnerability in make_oidjoins_check script.  Patch
attached.  Part of rh-postgresql-contrib therefore affects RHEL3

This issue does not affect the RHEL2.1 postgresql packages which did
not contain this script.
Comment 1 Mark J. Cox 2004-10-19 09:43:25 UTC 
Created attachment 105429 [details]
tempfile patch from vendor-sec
Comment 2 Tom Lane 2004-10-19 14:16:57 UTC 
What we have here is an unportable solution to a non-problem. 
make_oidjoins_check is never run by users, and in any case the worst
someone could do is screw up the regression test script it creates,
which would certainly be noticed before the script got used in
production.

I might be more excited about applying this if the proposed mktemp
call hadn't failed outright on the system that I would actually be
likely to be running make_oidjoins_check on for future PG releases;
but as is, the patch is unacceptable upstream, and I don't think it's
worth the trouble to try to develop a portable invocation of mktemp
for this.

A more likely patch would be to remove the script from the RPM
entirely ... it seems like a waste of space to have it there.
Comment 4 Tom Lane 2004-10-19 15:37:00 UTC 
It certainly wouldn't ever be run as root --- it's basically a tool
that is used about once per release cycle to update the oidjoins
regression test script to match the system catalogs' foreign-key
structure.

Also the attacker couldn't control what was going to be written, so
the vulnerability seems essentially nil to me.  Taking the script out
of the distro is definitely the most reasonable response.
Comment 5 Tom Lane 2004-10-26 01:03:53 UTC 
This is repaired in rh-postgresql-7.3.8-1.
Comment 6 John Flanagan 2004-12-20 17:54:14 UTC 
An errata 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-2004-489.html