02.11.2016 12:39, Markus Armbruster wrote: > Michael Tokarev <m...@tls.msk.ru> writes: > >> With current code, pid file is open after various >> sockets, chardevs, fsdevs and the like. This causes >> interesting effects, for example when monitor is a >> unix-socket, and another qemu instance is already >> running, new qemu first "damages" the socket and >> next complain that it can't acquire the pid file and >> exits, making running qemu unreachable. .. > ... is this early enough? For instance, we clobber the log file another > 200 lines further up.
Yeah, you're right. There's also some tracing initing in there and other potentially interesting stuff. I moved the pid file creation right after the os_daemonize() call. Before this point we don't even know our pid. > A "don't clobber anything externally visible before this spot" comment > could be useful. Hopefully this is "early enough" to not need this comment :) Thanks, /mjt
