tests: avoid hung processes due to gdb SIGCONT handling
commitfe02153e6475ae52f90cfccae1d1ba24baf35f9c
authorPádraig Brady <P@draigBrady.com>
Fri, 1 May 2015 17:03:37 +0000 (1 18:03 +0100)
committerPádraig Brady <P@draigBrady.com>
Mon, 11 May 2015 13:41:17 +0000 (11 14:41 +0100)
tree25006e80ac8a20eeebe3f402f4ceec6de74dc50e
parent8ee3ca4bb96f61ed5fc12da1d751a29fec6f7979
tests: avoid hung processes due to gdb SIGCONT handling

* tests/tail-2/inotify-race.sh: Add a `wait` to ensure that
we reap all background gdb and tail processes.  That resulted
in the test hanging intermittently and upon investigation was
due to gdb intermittently failing to terminate the child process
due to receiving a SIGCONT signal.  Therefore we avoid using
timeout(1) which sends that signal, and instead rely on tail's
inbuilt --pid monitoring on a background sleep process.
Given this new implementation, the VERY_EXPENSIVE guard was removed.
Related issues with this test hanging were previously discussed at:
https://lists.gnu.org/archive/html/bug-coreutils/2009-12/msg00025.html
tests/tail-2/inotify-race.sh