Here is the output from kill -3. This is with TThreadPoolServer, 128 worker threads. I hit the server with 100 clients.
http://pastebin.com/raw.php?i=axjYbau2 I noticed threads from pool-1-thread-22 to pool-1-thread-121 didn't show up in the dump. Does that mean they died? Thanks! --Michi On Apr 28, 2011, at 6:56 AM, Bryan Duxbury wrote: > It would probably be more interesting to get the java stack trace, rather > than the OS one. > > On Wed, Apr 27, 2011 at 8:55 PM, Michi Mutsuzaki <[email protected]>wrote: > >> Hi Bryan, >> >>> TThreadPoolServer does its IO in many threads, so it can be faster in >> some >>> circumstances. However, it's not "fair" in terms of requests - when a >> client >>> gets a thread, it monopolizes it until it closes, even if it's not even >>> making a request. In practice, this means you have to allow basically >>> unlimited threads. >> >> I think this is ok in my case. I know how many clients will be connecting >> to >> the server. >> >>> I don't know why thread pool server would stop responding after 30 >> minutes. >>> We'd need to see more information to have any chance of figuring that >> out. >> >> Here are the outputs from pstack and strace. I captured them after >> TThreadPoolServer stopped responding: >> >> http://pastebin.com/raw.php?i=2QzWFmz2 >> >> Let me know if there are other commands that might be helpful in figuring >> out >> what's going on. >> >> Thanks! >> --Michi >>
