#5212: waitForProcess: does not exist (No child processes) ------------------------------------------------+-------------------------- Reporter: chrisdone | Owner: Type: bug | Status: closed Priority: normal | Milestone: Component: Runtime System | Version: 7.4.2 Resolution: invalid | Keywords: Operating System: Linux | Architecture: x86_64 Type of failure: Incorrect result at runtime | (amd64) Test Case: | Difficulty: Blocking: | Unknown | Blocked By: | Related Tickets: ------------------------------------------------+-------------------------- Changes (by chrisdone): * cc: simonmar (added) * status: new => closed * resolution: => invalid Comment: The reason is because XMonad (and presumably some other library I was using two years ago) installs signal handlers for SIGCHILD: {{{ import Data.Maybe import Control.Monad import System.Posix.Signals import Control.Monad.Fix import Control.Exception import Control.Monad.Extra import Control.Monad.Trans import Control.Concurrent import System.Process import System.Posix.Process.ByteString main = do installSignalHandlers readProcess "ghc" ["--version"] "" >>= putStrLn -- | Ignore SIGPIPE to avoid termination when a pipe is full, and SIGCHLD to -- avoid zombie processes, and clean up any extant zombie processes. installSignalHandlers :: MonadIO m => m () installSignalHandlers = io $ do installHandler openEndedPipe Ignore Nothing installHandler sigCHLD Ignore Nothing (try :: IO a -> IO (Either SomeException a)) $ fix $ \more -> do x <- getAnyProcessStatus False False when (isJust x) more return () }}} This is pretty annoying because it breaks other uses of readProcess. My solution was to use a patched version of readProcess which handles exceptions thrown on waitForProcess. I guess this ticket can be closed… -- Ticket URL: http://ghc.haskell.org/trac/ghc/ticket/5212#comment:6 GHC http://www.haskell.org/ghc/ The Glasgow Haskell Compiler