[3320] | 1 | Wait + signals
|
---|
| 2 |
|
---|
| 3 | We had some bugs here which are hard to test in testsuite.
|
---|
| 4 |
|
---|
| 5 | Bug 1280 (http://busybox.net/bugs/view.php?id=1280):
|
---|
| 6 | was misbehaving in interactive ash. Correct behavior:
|
---|
| 7 |
|
---|
| 8 | $ sleep 20 &
|
---|
| 9 | $ wait
|
---|
| 10 | ^C
|
---|
| 11 | $ wait
|
---|
| 12 | ^C
|
---|
| 13 | $ wait
|
---|
| 14 | ^C
|
---|
| 15 | ...
|
---|
| 16 |
|
---|
| 17 |
|
---|
| 18 | Bug 1984 (http://busybox.net/bugs/view.php?id=1984):
|
---|
| 19 | traps were not triggering:
|
---|
| 20 |
|
---|
| 21 | trap_handler_usr () {
|
---|
| 22 | echo trap usr
|
---|
| 23 | }
|
---|
| 24 | trap_handler_int () {
|
---|
| 25 | echo trap int
|
---|
| 26 | }
|
---|
| 27 | trap trap_handler_usr USR1
|
---|
| 28 | trap trap_handler_int INT
|
---|
| 29 | sleep 3600 &
|
---|
| 30 | echo "Please do: kill -USR1 $$"
|
---|
| 31 | echo "or: kill -INT $$"
|
---|
| 32 | while true; do wait; echo wait interrupted; done
|
---|
| 33 |
|
---|
| 34 |
|
---|
| 35 | Bug 189 (https://bugs.busybox.net/show_bug.cgi?id=189)
|
---|
| 36 |
|
---|
| 37 | func() {
|
---|
| 38 | sleep 1
|
---|
| 39 | }
|
---|
| 40 | while (true); do
|
---|
| 41 | func
|
---|
| 42 | echo Looping
|
---|
| 43 | done
|
---|
| 44 |
|
---|
| 45 | ^C was observed to make ash processes geometrically multiply (!) instead
|
---|
| 46 | of exiting. (true) in subshell does not seem to matter, as another user
|
---|
| 47 | reports the same with:
|
---|
| 48 |
|
---|
| 49 | trap "echo USR1" USR1
|
---|
| 50 | while true; do
|
---|
| 51 | echo Sleeping
|
---|
| 52 | sleep 5
|
---|
| 53 | done
|
---|
| 54 |
|
---|
| 55 | Compat note.
|
---|
| 56 | Bash version 3.2.0(1) exits this script at the receipt of SIGINT
|
---|
| 57 | _only_ if it had two last children die from it.
|
---|
| 58 | The following trace was obtained while periodically running
|
---|
| 59 | "killall -SIGINT sleep; sleep 0.1; kill -SIGINT <bash_PID>":
|
---|
| 60 |
|
---|
| 61 | 23:48:32.376707 clone(...) = 13528
|
---|
| 62 | 23:48:32.388706 waitpid(-1, 0xffc832ec, 0) = ? ERESTARTSYS (To be restarted)
|
---|
| 63 | 23:48:32.459761 --- SIGINT (Interrupt) @ 0 (0) ---
|
---|
| 64 | kill -SIGINT <bash_PID> is ignored, back to waiting:
|
---|
| 65 | 23:48:32.463706 waitpid(-1, [{WIFEXITED(s) && WEXITSTATUS(s) == 0}], 0) = 13528
|
---|
| 66 | sleep exited with 0
|
---|
| 67 | 23:48:37.377557 --- SIGCHLD (Child exited) @ 0 (0) ---
|
---|
| 68 | 23:48:37.378451 clone(...) = 13538
|
---|
| 69 | 23:48:37.390708 waitpid(-1, [{WIFSIGNALED(s) && WTERMSIG(s) == SIGINT}], 0) = 13538
|
---|
| 70 | sleep was killed by "killall -SIGINT sleep"
|
---|
| 71 | 23:48:38.523944 --- SIGCHLD (Child exited) @ 0 (0) ---
|
---|
| 72 | 23:48:38.524861 clone(...) = 13542
|
---|
| 73 | 23:48:38.538706 waitpid(-1, 0xffc832ec, 0) = ? ERESTARTSYS (To be restarted)
|
---|
| 74 | 23:48:38.624761 --- SIGINT (Interrupt) @ 0 (0) ---
|
---|
| 75 | kill -SIGINT <bash_PID> is ignored, back to waiting:
|
---|
| 76 | 23:48:38.628706 waitpid(-1, [{WIFEXITED(s) && WEXITSTATUS(s) == 0}], 0) = 13542
|
---|
| 77 | sleep exited with 0
|
---|
| 78 | 23:48:43.525674 --- SIGCHLD (Child exited) @ 0 (0) ---
|
---|
| 79 | 23:48:43.526563 clone(...) = 13545
|
---|
| 80 | 23:48:43.538709 waitpid(-1, [{WIFSIGNALED(s) && WTERMSIG(s) == SIGINT}], 0) = 13545
|
---|
| 81 | sleep was killed by "killall -SIGINT sleep"
|
---|
| 82 | 23:48:44.466848 --- SIGCHLD (Child exited) @ 0 (0) ---
|
---|
| 83 | 23:48:44.467735 clone(...) = 13549
|
---|
| 84 | 23:48:44.481706 waitpid(-1, 0xffc832ec, 0) = ? ERESTARTSYS (To be restarted)
|
---|
| 85 | 23:48:44.567757 --- SIGINT (Interrupt) @ 0 (0) ---
|
---|
| 86 | kill -SIGINT <bash_PID> is ignored, back to waiting:
|
---|
| 87 | 23:48:44.571706 waitpid(-1, [{WIFEXITED(s) && WEXITSTATUS(s) == 0}], 0) = 13549
|
---|
| 88 | sleep exited with 0
|
---|
| 89 | 23:48:49.468553 --- SIGCHLD (Child exited) @ 0 (0) ---
|
---|
| 90 | 23:48:49.469445 clone(...) = 13551
|
---|
| 91 | 23:48:49.481708 waitpid(-1, [{WIFSIGNALED(s) && WTERMSIG(s) == SIGINT}], 0) = 13551
|
---|
| 92 | sleep was killed by "killall -SIGINT sleep"
|
---|
| 93 | 23:48:50.515837 --- SIGCHLD (Child exited) @ 0 (0) ---
|
---|
| 94 | 23:48:50.516718 clone(...) = 13555
|
---|
| 95 | 23:48:50.530706 waitpid(-1, 0xffc832ec, 0) = ? ERESTARTSYS (To be restarted)
|
---|
| 96 | 23:48:50.615761 --- SIGINT (Interrupt) @ 0 (0) ---
|
---|
| 97 | kill -SIGINT <bash_PID> is ignored, back to waiting:
|
---|
| 98 | 23:48:50.619705 waitpid(-1, [{WIFSIGNALED(s) && WTERMSIG(s) == SIGINT}], 0) = 13555
|
---|
| 99 | sleep was killed by "killall -SIGINT sleep".
|
---|
| 100 | This is the second one in a row. Kill ourself:
|
---|
| 101 | 23:48:51.504604 kill(13515, SIGINT) = 0
|
---|
| 102 | 23:48:51.504689 --- SIGINT (Interrupt) @ 0 (0) ---
|
---|
| 103 | 23:48:51.504915 +++ killed by SIGINT +++
|
---|
| 104 |
|
---|
| 105 | As long as there is at least one "sleep 5" which exited successfully
|
---|
| 106 | (not killed by SIGINT), bash continues. This is not documented anywhere
|
---|
| 107 | AFAIKS.
|
---|
| 108 |
|
---|
| 109 | Why keyboard ^C acts differently?
|
---|
| 110 |
|
---|
| 111 | 00:08:07.655985 clone(...) = 14270
|
---|
| 112 | 00:08:07.669707 waitpid(-1, [{WIFEXITED(s) && WEXITSTATUS(s) == 0}], 0) = 14270
|
---|
| 113 | 00:08:12.656872 --- SIGCHLD (Child exited) @ 0 (0) ---
|
---|
| 114 | 00:08:12.657743 clone(...) = 14273
|
---|
| 115 | 00:08:12.671708 waitpid(-1, [{WIFSIGNALED(s) && WTERMSIG(s) == SIGINT}], 0) = 14273
|
---|
| 116 | 00:08:13.810778 --- SIGINT (Interrupt) @ 0 (0) ---
|
---|
| 117 | 00:08:13.818705 kill(14269, SIGINT) = 0
|
---|
| 118 | 00:08:13.820103 --- SIGINT (Interrupt) @ 0 (0) ---
|
---|
| 119 | 00:08:13.820925 +++ killed by SIGINT +++
|
---|
| 120 |
|
---|
| 121 | Perhaps because at the moment bash got SIGINT it had no children?
|
---|
| 122 | (it did not manage to spawn new sleep yet, see the trace)
|
---|