Hi, On Thu, 26 Sep 2013, Dong Yuan wrote: > Hello, Sage. > > I meet the same bug as #5883. It is happy to know that the bug is > fixed, but although I spent lots of time to try to get why, I still > couldn't understand the reason. > > Is there any race conditions? Or which threads? I guess the threads > about this bug are Pipe.reader and DispatchQueue.entry, But I am not > sure. First, we need to carry the ref through teh function in case another thread (reader *or* writer) gets a socket error and tears down the socket. Second, we need to hold mds_lock to avoid races with other mds code when looking at the Session*. Hope that helps! sage > > Maybe you can give me more details. > > Thank you very much. > > -- > Dong Yuan > Email:yuandong1222@xxxxxxxxx > > -- To unsubscribe from this list: send the line "unsubscribe ceph-devel" in the body of a message to majordomo@xxxxxxxxxxxxxxx More majordomo info at http://vger.kernel.org/majordomo-info.html
