On Sat, May 9, 2015 at 4:47 PM, Eric Wong <e...@80x24.org> wrote: > "Lin Jen-Shin (godfat)" <god...@godfat.org> wrote: >> A backtrace for knowing what's happening I think is quite enough for me now. >> Still curious though, could this worker do anything else if this happened? >> I am guessing that if the application no longer does anything, then this >> worker >> would not do anything. Or the socket might timeout eventually? > > It depends on the application structure. > Often apps have very different code paths for different endpoints so > some endpoint being fatally broken may not affect others. A simple > endpoint (e.g. static files) could function at 100% and serve other > clients without any problems. > > Eventually the socket will timeout if the client_expire_threshold is > reached, otherwise it's fairly harmless to keep the socket around > (aside from memory overhead).
Great! I was just worried that idled workers would get piled up and eventually no other workers would be able to do any work. As long as there's a timeout for this and it could recover itself, I think this is could be the best solution given all the trade off. Thank you!
