Issue 75734 - Get rid of nas completely (security issues fixed in nas 1.8b)
Summary: Get rid of nas completely (security issues fixed in nas 1.8b)
Status: CLOSED DUPLICATE of issue 81172
Alias: None
Product: utilities
Classification: Unclassified
Component: code (show other issues)
Version: OOo 2.2 RC4
Hardware: All All
: P3 Trivial (vote)
Target Milestone: ---
Assignee: caolanm
QA Contact: Unknown
URL:
Keywords: oooqa
Depends on:
Blocks:
 
Reported: 2007-03-26 09:40 UTC by rene
Modified: 2007-09-23 12:35 UTC (History)
4 users (show)

See Also:
Issue Type: DEFECT
Latest Confirmation in: ---
Developer Difficulty: ---


Attachments
patch (6.71 KB, patch)
2007-03-26 18:08 UTC, rene
no flags Details | Diff

Note You need to log in before you can comment on or make changes to this issue.
Description rene 2007-03-26 09:40:35 UTC
http://freshmeat.net/projects/nas/?branch_id=13568?release_id=250259

I don't know yet whether those fixes also apply to the internal nas 1.6 we ship
in the tree..

Only info AFAIS is from the homepage

--- snip ---
3/25/2007

NAS 1.8b (devel) is now available. See HISTORY for details. This version
includes some device open fixes, as well as fixes for serveral denial of service
vulnerabilites. 
--- snip ---
Comment 1 rene 2007-03-26 10:21:10 UTC
according to http://bugs.debian.org/cgi-bin/bugreport.cgi?bug=416038 our at
least 1.7 - shipped in Debian stable -- is also affected (maybe then the even
older 1.6, too?)

The changelog for the sid update says some details:

 nas (1.8-4) unstable; urgency=high
 .
    * High-urgency upload to fix multiple security holes (CVE-2007-1543,
      CVE-2007-1544, CVE-2007-1545, CVE-2007-1546 and CVE-2007-1547):
     + accept_att_local buffer overflow through USL connection
     + server termination through unexistent ID in AddResource
     + bcopy crash caused by integer overflow in ProcAuWriteElement
     + invalid memory pointer caused by big num_actions in
       ProcAuSetElements
     + another invalid memory pointer caused by big num_actions in
       ProcAuSetElements
     + invalid memory pointer in compileInputs
     + exploits bug 3 in read mode (requires something playing on
       the server)
     + NULL pointer caused by too much connections
     + Closes: #416038
Comment 2 pavel 2007-03-26 17:27:42 UTC
we should remove nas completely, I think none is using it.
Comment 3 rene 2007-03-26 18:07:39 UTC
probably. but in the meanwhile we should fix it, shouldn't we?
Or should we release 2.2.0 with this included? :)

Anyway, the interdiff between Debians 1.8-3 and 1.8-4 applies to our 1.6.
Attached is the patch.
Comment 4 rene 2007-03-26 18:07:47 UTC
probably. but in the meanwhile we should fix it, shouldn't we?
Or should we release 2.2.0 with this included? :) Or remove nas in a rc? :)

Anyway, the interdiff between Debians 1.8-3 and 1.8-4 applies to our 1.6.
Attached is the patch.
Comment 5 rene 2007-03-26 18:08:19 UTC
(soory for the double posting, hit submit too early)
Comment 6 rene 2007-03-26 18:08:55 UTC
Created attachment 43947 [details]
patch
Comment 7 rene 2007-03-26 18:21:08 UTC
as cloph rightly pointed oput (doh, why didn't I notice) the server-only is
affected...
Comment 8 lohmaier 2007-05-21 14:48:40 UTC
Yes, get rid of nas completely.

Nas works only where OOo uses its own sound-playing stuff. And the only place I
know that does play sound by itself it the preview button in the file-selection
dialog for sound-effects.

Yes, a "play" button in a file-dialog (and only in OOo's own dialog, the
gtk-dialog doesn't have that button). That's all. All other places I know ignore
nas completely and rely on the JMF.

See http://www.mail-archive.com/dev@gsl.openoffice.org/msg00421.html and
follow-ups for details.
Comment 9 Martin Hollmichel 2007-09-07 14:10:17 UTC
@pl, do you agree on removing nas ?
Comment 10 philipp.lohmann 2007-09-07 15:02:44 UTC
mh: yes I do

and I think cmc already has a CWS for that (or planned to), yes ?
Comment 11 caolanm 2007-09-23 12:30:51 UTC
yup, just waiting until I come back from holidays to push the workspace through

*** This issue has been marked as a duplicate of 81172 ***
Comment 12 caolanm 2007-09-23 12:35:15 UTC
close as dup