Bug 86447 - Crash importing malformed .ppt -- DoS
Summary: Crash importing malformed .ppt -- DoS
Status: RESOLVED WONTFIX
Alias: None
Product: LibreOffice
Classification: Unclassified
Component: filters and storage (show other bugs)
Version:
(earliest affected)
3.5.4 release
Hardware: x86-64 (AMD64) Linux (All)
: medium normal
Assignee: Not Assigned
URL:
Whiteboard:
Keywords:
Depends on:
Blocks:
 
Reported: 2014-11-19 00:36 UTC by Alexander Cherepanov
Modified: 2015-01-09 12:42 UTC (History)
3 users (show)

See Also:
Crash report or crash signature:


Attachments
Crasher (33.00 KB, application/vnd.ms-powerpoint)
2014-11-19 00:36 UTC, Alexander Cherepanov
Details
Valgrind log (15.31 KB, text/plain)
2014-11-19 00:37 UTC, Alexander Cherepanov
Details

Note You need to log in before you can comment on or make changes to this bug.
Description Alexander Cherepanov 2014-11-19 00:36:04 UTC
A crash while importing malformed .ppt file. According to valgrind (log attached) it is due to null pointer deref. Seem to be DoS only.
Tested on Debian Stable.
Comment 1 Alexander Cherepanov 2014-11-19 00:36:50 UTC
Created attachment 109699 [details]
Crasher
Comment 2 Alexander Cherepanov 2014-11-19 00:37:04 UTC
Created attachment 109700 [details]
Valgrind log
Comment 3 Jean-Baptiste Faure 2014-11-19 05:44:15 UTC
Did you really report this bug against version 3.5.4? This version is very old and not maintained anymore. If it is the case, did you check if the problem is present in current stable versions and in master? If the problem is not present in the current versions, I fear we should close this bug report as WontFix.

Set status to NEEDINFO. Please set it back to UNCONFIRMED once you have provided requested informations. Thank you for your understanding.

Best regards. JBF
Comment 4 Julien Nabet 2014-11-19 21:52:31 UTC
Precisely, 3.5.4 has been released more than 2 years ago, see https://wiki.documentfoundation.org/ReleasePlan#3.5_release
Last stable LO version is now 4.3.4
Comment 5 Alexander Cherepanov 2014-11-21 02:01:20 UTC
Yes, I really mean version 3.5.4. It is shipped with Debian Stable and is used by a sizable number of people.

I understand that this version is not supported upstream, so feel free to close the bug with any resolution you like. If the crash is present in the master I think it will be tracked by crashtest automatically.
Comment 6 Alexander Cherepanov 2014-11-21 02:09:51 UTC
(In reply to Julien Nabet from comment #4)
> Precisely, 3.5.4 has been released more than 2 years ago, see

Yes, and it is supposed to be supported in Debian for a couple of years more.
If you take Ubuntu then Ubuntu 12.04 LTS includes 3.5.7 and is supposed to be supported until 2017-04.
Comment 7 Julien Nabet 2014-11-21 06:35:26 UTC
(In reply to Alexander Cherepanov from comment #6)
> (In reply to Julien Nabet from comment #4)
> > Precisely, 3.5.4 has been released more than 2 years ago, see
> 
> Yes, and it is supposed to be supported in Debian for a couple of years more.
> If you take Ubuntu then Ubuntu 12.04 LTS includes 3.5.7 and is supposed to
> be supported until 2017-04.

I don't understand what do you expect. Even if we would have reproduced the problem with last sources (3.5 has been EOL since a long time) and would have fixed it, the LO version in these version distrib will be same.
So it's Ubuntu/Debian issue if they don't backport recent version, not LO matter.
Comment 8 Alexander Cherepanov 2014-11-21 17:06:55 UTC
FTR: the crash is inside libwps. Hence it's probably not a bug in LO. Similar to bug 86446.
Comment 9 Alexander Cherepanov 2014-11-21 17:35:44 UTC
(In reply to Julien Nabet from comment #7)
> I don't understand what do you expect. Even if we would have reproduced the
> problem with last sources (3.5 has been EOL since a long time) and would
> have fixed it, the LO version in these version distrib will be same.
> So it's Ubuntu/Debian issue if they don't backport recent version, not LO
> matter.

I don't expect anything else from this PR. At least two goals are reached:
- the crasher is uploaded to some bugzilla so that LO crashtest script will find it (IIUC it cannot work with Debian bugtracker);
- there is a central place where downstreams could be pointed to regarding this issue.
Further work (if any) is for downstreams.