Project: chromium Issues People Development process History Sign in
New issue
Advanced search Search tips
Note: Color blocks (like or ) mean that a user may not be available. Tooltip shows the reason.
Starred by 1 user
Status: Verified
Owner:
User never visited
Closed: Aug 2010
Cc:
Components:
EstimatedDays: ----
NextAction: ----
OS: Linux, Mac
Pri: 2
Type: Bug-Security
M-6

Restricted
  • Only users with EditIssue permission may comment.



Sign in to add a comment
Null deref when socket stream is closed during hostname resolution
Reported by camp...@iit.edu, Aug 11 2010 Back to list
This is a followup to the browser crashing part of  issue 51630 

Reproduction case attached as crash.html, crash.py

 
crash.html
129 bytes View Download
crash.py
1.6 KB View Download
Comment 1 by camp...@iit.edu, Aug 11 2010
I'm running Chromium 6.0.490.0 on Mac OS 10.6.4
Labels: -Pri-0 -Area-Undefined Pri-2 Area-Internals OS-Linux OS-Mac Mstone-6 Internals-Network SecSeverity-Low
Status: Started
Summary: Null deref when socket stream is closed during hostname resolution (was: NULL)
When a socket stream closes the delegate_ is nulled out. When this happens in the middle of a host resolution SocketStream::DoResolveHostComplete() will dereference the null pointer and crash the browser. 


There is also another issue that became apparent in testing this which I will be opening a new bug for. You can essentially DoS the browser and prevent it from creating additional connections by opens lots socket streams (websocket etc). I will look at how we can rate limit this or define a cutoff for maximum per origin socket streams.
Comment 3 by bugdro...@gmail.com, Aug 13 2010
The following revision refers to this bug:
    http://src.chromium.org/viewvc/chrome?view=rev&revision=56076 

------------------------------------------------------------------------
r56076 | cdn@chromium.org | 2010-08-13 13:29:05 -0700 (Fri, 13 Aug 2010) | 5 lines
Changed paths:
   M http://src.chromium.org/viewvc/chrome/trunk/src/net/socket_stream/socket_stream.cc?r1=56076&r2=56075

Added check for null delegate when a socket stream closes between DoResolveHost and DoResolveHostDone.

BUG= 51846 
TEST=None
Review URL: http://codereview.chromium.org/3145010
------------------------------------------------------------------------

 Issue 52080  has been merged into this issue.
Apparently there is still another crash here that is unique to release builds. Leaving this open for the moment to try and track that down.
campkei@ can you sync with trunk now that the null deref fix is in and try to reproduce. There has been a lot of cleanup in websockets in the past few days as a result of your original bug (thanks btw :) and I am wondering if this is still hitting with the patches. If not we might have to play with revisions to get a build where it will repro.
Comment 8 by kerz@chromium.org, Aug 18 2010
Labels: -Mstone-6 Mstone-7
Move back to 6 if needed.
Comment 9 by kerz@chromium.org, Aug 18 2010
Move back to 6 if needed.
Labels: -Mstone-7 Mstone-6
This is M6. Needs a merge and some checks to make sure that memory corruption is not happening.
Merged browser crash part to 472 branch.
The following revision refers to this bug:
    http://src.chromium.org/viewvc/chrome?view=rev&revision=56618 

------------------------------------------------------------------------
r56618 | inferno@chromium.org | 2010-08-18 16:35:59 -0700 (Wed, 18 Aug 2010) | 8 lines
Changed paths:
   M http://src.chromium.org/viewvc/chrome/branches/472/src/net/socket_stream/socket_stream.cc?r1=56618&r2=56617

Merge 56076 - Added check for null delegate when a socket stream closes between DoResolveHost and DoResolveHostDone.

BUG= 51846 
TEST=None
Review URL: http://codereview.chromium.org/3145010


Review URL: http://codereview.chromium.org/3170030
------------------------------------------------------------------------

Labels: -Restrict-View-SecurityTeam Restrict-View-SecurityNotify
Status: FixUnreleased
Unable to reproduce the memory corruption of any platform including Mac OS snow leopard. If you can still trigger it after the fixes by @ukai, please feel free to file a new bug.
Comment 14 by camp...@iit.edu, Aug 20 2010
Cannot reproduce memory corruption or even tab crash with Release build Chromium 7.0.500.0 (56798) on Mac OS 10.6.4
I think the fix for  issue 51630  has covered up the memory corruption problem (need to corrupt memory in the tab process to corrupt memory in the browser process)
(Sorry it took me so long, I've been quite busy)

@inferno you want me to merge the r56617,r56618 patch to Chromium 6.0.490.0 and see if I can still reproduce?

@campkei, don't worry about it. we had pretty rock solid audit from @cdn and awesome patch from @ukai. and i did test your repros to make sure they didn't reproduce on mac. also, our qa will do full round of testing on all platforms before it gets shipped to v6 stable.

thanks a lot for all your help. you can still do more fuzzing to see if you find anything new and get awarded, but i think all existing testcases from you are fixed.
Labels: -Restrict-View-SecurityNotify
Status: Fixed
Labels: Verifier-Rohitbm
Verified label updated by AutoAllocator, contact AmolK or KrisR for details
Status: Verified
7.0.530.0 (Official Build 60028) dev
Labels: Type-Security
Labels: SecImpacts-Stable
Batch update.
Project Member Comment 21 by bugdroid1@chromium.org, Oct 13 2012
Labels: Restrict-AddIssueComment-Commit
This issue has been closed for some time. No one will pay attention to new comments.
If you are seeing this bug or have new data, please click New Issue to start a new bug.
Project Member Comment 22 by bugdroid1@chromium.org, Mar 10 2013
Labels: -Area-Internals -Mstone-6 -Internals-Network -SecSeverity-Low -Type-Security -SecImpacts-Stable Security-Severity-Low Security-Impact-Stable M-6 Cr-Internals Cr-Internals-Network Type-Bug-Security
Project Member Comment 23 by bugdroid1@chromium.org, Mar 13 2013
Labels: -Restrict-AddIssueComment-Commit Restrict-AddIssueComment-EditIssue
Project Member Comment 24 by bugdroid1@chromium.org, Mar 21 2013
Labels: -Security-Severity-Low Security_Severity-Low
Project Member Comment 25 by bugdroid1@chromium.org, Mar 21 2013
Labels: -Security-Impact-Stable Security_Impact-Stable
Project Member Comment 26 by sheriffbot@chromium.org, Oct 1 2016
This bug has been closed for more than 14 weeks. Removing security view restrictions.

For more details visit https://www.chromium.org/issue-tracking/autotriage - Your friendly Sheriffbot
Project Member Comment 27 by sheriffbot@chromium.org, Oct 2 2016
This bug has been closed for more than 14 weeks. Removing security view restrictions.

For more details visit https://www.chromium.org/issue-tracking/autotriage - Your friendly Sheriffbot
Labels: allpublic
Sign in to add a comment