Debugging a classic ReaderWriterLock deadlock with SOSex.dllhttp://blogs.msdn.com/b/tess/archive/2010/04/27/debugging-a-classic-readerwriterlock-deadlock-with-sosex-dll.aspxI was helping out on an issue the other day where the process would stall if they added enough users in their load tests.&#160; Btw, serious kudos to them for making load tests, so much nicer to work with a problem in test rather than when it is gettingsv-SETelligent Evolution Platform Developer Build (Build: 5.6.50428.7875)re: Debugging a classic ReaderWriterLock deadlock with SOSex.dllhttp://blogs.msdn.com/b/tess/archive/2010/04/27/debugging-a-classic-readerwriterlock-deadlock-with-sosex-dll.aspx#10332851Tue, 24 Jul 2012 06:42:36 GMT91d46819-8472-40ad-a661-2c78acb4018c:10332851Bernhard<p>Hi,</p> <p>i try to investigate a hung with windbg. If I call the command</p> <p>!mlocks i got the following</p> <p>:000&gt; !mlocks</p> <p>Examining SyncBlocks...</p> <p>Scanning for ReaderWriterLock instances...</p> <p>Scanning for holders of ReaderWriterLock locks...</p> <p>Scanning for ReaderWriterLockSlim instances...</p> <p>Scanning for holders of ReaderWriterLockSlim locks...</p> <p>Examining CriticalSections...</p> <p>ClrThread &nbsp;DbgThread &nbsp;OsThread &nbsp; &nbsp;LockType &nbsp; &nbsp;Lock &nbsp; &nbsp; &nbsp; &nbsp; &nbsp; &nbsp; &nbsp;LockLevel</p> <p>--------------------------------------------------------------------------</p> <p>0x640064 &nbsp; -1 &nbsp; &nbsp; &nbsp; &nbsp; 0xffffffff &nbsp;RWLock &nbsp; &nbsp; &nbsp;000000000339a338 &nbsp;Writer &nbsp; &nbsp; &nbsp; &nbsp;</p> <p>0x6 &nbsp; &nbsp; &nbsp; &nbsp;7 &nbsp; &nbsp; &nbsp; &nbsp; &nbsp;0x1ea8 &nbsp; &nbsp; &nbsp;thinlock &nbsp; &nbsp;000000000343ddd8 &nbsp;(recursion:0)</p> <p>When executeing rwlocks i got the following:</p> <p>0:000&gt; !rwlocks 000000000339a338 &nbsp; &nbsp; &nbsp; &nbsp; &nbsp; </p> <p>No export rwlocks found</p> <p>0:000&gt; !rwlock 000000000339a338 &nbsp; &nbsp; &nbsp; &nbsp; &nbsp; </p> <p>WriterThread: &nbsp; &nbsp; &nbsp; &nbsp; &nbsp; 0x640064 (DEAD)</p> <p>WriterLevel: &nbsp; &nbsp; &nbsp; &nbsp; &nbsp; &nbsp;115</p> <p>WaitingWriterCount: &nbsp; &nbsp; 0</p> <p>WriterEvent: &nbsp; &nbsp; &nbsp; &nbsp; &nbsp; &nbsp;0</p> <p>WaitingWriterThreadIds: None</p> <p>ReaderCount: &nbsp; &nbsp; &nbsp; &nbsp; &nbsp; &nbsp;116</p> <p>CurrentReaderThreadIds: </p> <p>WaitingReaderCount: &nbsp; &nbsp; 576</p> <p>ReaderEvent: &nbsp; &nbsp; &nbsp; &nbsp; &nbsp; &nbsp;80400002</p> <p>WaitingReaderThreadIds: </p> <p>*This lock has 116 orphaned reader locks.</p> <p>0:007&gt; !rwlock</p> <p>Address &nbsp; &nbsp; &nbsp; &nbsp; &nbsp; &nbsp;ReaderCount &nbsp; WaitingReaderCount &nbsp; &nbsp; WriterThread &nbsp; WaitingWriterCount</p> <p>...</p> <p>000000000339a338 &nbsp; &nbsp; &nbsp; &nbsp; &nbsp; 116 &nbsp; &nbsp; &nbsp; &nbsp; &nbsp; &nbsp; &nbsp; &nbsp; &nbsp;576 &nbsp; &nbsp; &nbsp; &nbsp; &nbsp; &nbsp;0x640064 &nbsp; &nbsp; &nbsp; &nbsp; &nbsp; &nbsp; &nbsp; &nbsp; &nbsp; &nbsp;0</p> <p>...</p> <p>00000000053f0688 &nbsp; &nbsp; &nbsp; &nbsp; &nbsp; 568 &nbsp; &nbsp; &nbsp; &nbsp; &nbsp; &nbsp; &nbsp; &nbsp; &nbsp;499 &nbsp; &nbsp; &nbsp; &nbsp; &nbsp; &nbsp;-- &nbsp; &nbsp; &nbsp; &nbsp; &nbsp; &nbsp; &nbsp; &nbsp; &nbsp; &nbsp; &nbsp; 6</p> <p>...</p> <p>i got this.</p> <p>When I call </p> <p>0:000&gt; !dlk</p> <p>Examining SyncBlocks...</p> <p>Scanning for ReaderWriterLock instances...</p> <p>Scanning for holders of ReaderWriterLock locks...</p> <p>Scanning for ReaderWriterLockSlim instances...</p> <p>Scanning for holders of ReaderWriterLockSlim locks...</p> <p>Examining CriticalSections...</p> <p>Scanning for threads waiting on SyncBlocks...</p> <p>Scanning for threads waiting on ReaderWriterLock locks...</p> <p>Scanning for threads waiting on ReaderWriterLocksSlim locks...</p> <p>Scanning for threads waiting on CriticalSections...</p> <p>No deadlocks detected.</p> <p>then no deadlock will be detected.</p> <p>My question is if this threads have to do with my hunging application,</p> <p>and what the scenario could be.</p> <p>What does it mean that the thread is DEAD. Or do I have to find somewhere else </p> <p>to find out the root cause of my hunging application?</p> <p>Please help me to interpret this output.</p> <p>Regards,</p> <p>Bernhard</p> <div style="clear:both;"></div><img src="http://blogs.msdn.com/aggbug.aspx?PostID=10332851" width="1" height="1">re: Debugging a classic ReaderWriterLock deadlock with SOSex.dllhttp://blogs.msdn.com/b/tess/archive/2010/04/27/debugging-a-classic-readerwriterlock-deadlock-with-sosex-dll.aspx#10158177Tue, 26 Apr 2011 15:33:56 GMT91d46819-8472-40ad-a661-2c78acb4018c:10158177Yi Shen<p>I&#39;m wondering why didn&#39;t the &#39;timeout&#39; work. </p> <div style="clear:both;"></div><img src="http://blogs.msdn.com/aggbug.aspx?PostID=10158177" width="1" height="1">re: Debugging a classic ReaderWriterLock deadlock with SOSex.dllhttp://blogs.msdn.com/b/tess/archive/2010/04/27/debugging-a-classic-readerwriterlock-deadlock-with-sosex-dll.aspx#10003384Tue, 27 Apr 2010 18:52:59 GMT91d46819-8472-40ad-a661-2c78acb4018c:10003384Raghuraman<p>Thank you for the write up. It was Neatly Explained, Simple and short to the point.</p> <p>Regards</p> <div style="clear:both;"></div><img src="http://blogs.msdn.com/aggbug.aspx?PostID=10003384" width="1" height="1">