https://bz.apache.org/bugzilla/show_bug.cgi?id=63191
--- Comment #4 from Mark Thomas <ma...@apache.org> --- So far, you haven't provided any evidence to back up a claim of a Tomcat bug. That the changes between CometD 3 and CometD 4 trigger the issue look to be valid based on the evidence provided. The claim that there is an I/O bug in Tomcat doesn't make any sense at this point as there is no evidence supporting that claim. The stack-traces are not consistent with your claims. The longer stack trace you quote is from Tomcat reading an incoming WebSocket message. It is not triggered by a message write. I'm not seeing any link between a call to wsSession.getAsyncRemote().sendText(...) and that longer stack trace. The longer stack trace is blocked in CometD code waiting for a CometD Promise to complete. I can't find any trace of code that might trigger that completion in the full stack trace (but I don't know CometD so I might have missed something). My suggestion is to work with the CometD folks to figure out why that Promise isn't completing. I'm not ruling out a Tomcat bug at this point. The explanation of why that Promise isn't completing may point to a Tomcat issue but, if it does, we need that explanation to know where to look at what to look for. -- You are receiving this mail because: You are the assignee for the bug. --------------------------------------------------------------------- To unsubscribe, e-mail: dev-unsubscr...@tomcat.apache.org For additional commands, e-mail: dev-h...@tomcat.apache.org
