Thanks. For me issue came in when one had a async_connect, e.g. inside the handle_resolve (called from io service thread as an async_resolve) was done. When a boost exception was thrown, it wasn't caught anywhere as the io_service::run was started directly in a thread, causing the program to terminate. I'm currently catching the errors in a wrapper thread to io_service's run and then propagating these exceptions back to the main thread and rethrow them there. On Fri, 2009-05-29 at 11:51 +0300, Roman Shmelev wrote:
m.. I haven't read everything, but it seems ok for me that exceptions are thrown and catched in the threads where io_service.run() was called. At least, so was the behavior in my program - I just wrapped io_service.run() with try-catch block. _______________________________________________ Boost-users mailing list Boost-users@lists.boost.org http://lists.boost.org/mailman/listinfo.cgi/boost-users