Jared Cantwell commented on ZOOKEEPER-897:
we are using the 3.3.2 release. i don't think the patch leaks memory because
destroy() will eventually get called (by the reentrant call to
zookeeper_close()), which calls cleanup_bufs() and frees those buffers, right?
Also, i had a test that reproduced this error, but it was easiest to reproduce
if i injected artificial sleeps into the zookeeper.c file. If that's ok, then
I can submit that. Otherwise, i'll see if i can devise a test that can
reproduce it otherwise.
> C Client seg faults during close
> Key: ZOOKEEPER-897
> URL: https://issues.apache.org/jira/browse/ZOOKEEPER-897
> Project: Zookeeper
> Issue Type: Bug
> Components: c client
> Reporter: Jared Cantwell
> Assignee: Jared Cantwell
> Fix For: 3.3.2, 3.4.0
> Attachments: ZOOKEEEPER-897.diff, ZOOKEEPER-897.patch
> We observed a crash while closing our c client. It was in the do_io() thread
> that was processing as during the close() call.
> #0 queue_buffer (list=0x6bd4f8, b=0x0, add_to_front=0) at src/zookeeper.c:969
> #1 0x000000000046234e in check_events (zh=0x6bd480, events=<value optimized
> out>) at src/zookeeper.c:1687
> #2 0x0000000000462d74 in zookeeper_process (zh=0x6bd480, events=2) at
> #3 0x0000000000469c34 in do_io (v=0x6bd480) at src/mt_adaptor.c:311
> #4 0x00007ffff7bc59ca in start_thread () from /lib/libpthread.so.0
> #5 0x00007ffff6f706fd in clone () from /lib/libc.so.6
> #6 0x0000000000000000 in ?? ()
> We tracked down the sequence of events, and the cause is that input_buffer is
> being freed from a thread other than the do_io thread that relies on it:
> 1. do_io() call check_events()
> 2. if(events&ZOOKEEPER_READ) branch executes
> 3. if (rc > 0) branch executes
> 4. if (zh->input_buffer != &zh->primer_buffer) branch executes
> .....in the meantime......
> 5. zookeeper_close() called
> 6. if (inc_ref_counter(zh,0)!=0) branch executes
> 7. cleanup_bufs() is called
> 8. input_buffer is freed at the end
> ..... back to check_events().........
> 9. queue_events() is called on a NULL buffer.
> I believe the patch is to only call free_completions() in zookeeper_close()
> and not cleanup_bufs(). The original reason cleanup_bufs() was added was to
> call any outstanding synhcronous completions, so only free_completions (which
> is guarded) is needed. I will submit a patch for review with this change.
This message is automatically generated by JIRA.
You can reply to this email to add a comment to the issue online.