Anders Qvist
2002-10-22 16:00:39 UTC
Next one is a tricky beast. test_signal.py on both our sun4u solaris 8
machines hangs if run as a part of the test suite, but not if run on
its own. The top part of the stack trace is as follows:
#0 0xdf5da12c in _pause () from /usr/lib/libc.so.1
#1 0xdf4fa8bc in pause () from /usr/lib/libthread.so.1
#2 0x000792b0 in call_function (pp_stack=0xdfffded8, oparg=0)
at ../python/dist/src/Python/ceval.c:3214
#3 0x00076fd0 in eval_frame (f=0x1a1110)
at ../python/dist/src/Python/ceval.c:1993
#4 0x000785f0 in PyEval_EvalCodeEx (co=0x7333e0, globals=0x1a1110,
locals=0x0, args=0x0, argcount=0, kws=0x0, kwcount=0, defs=0x0,
defcount=0, closure=0x0) at ../python/dist/src/Python/ceval.c:2538
#5 0x0007b03c in PyEval_EvalCode (co=0x7333e0, globals=0x7adb70,
locals=0x7adb70) at ../python/dist/src/Python/ceval.c:478
#6 0x00091280 in PyImport_ExecCodeModuleEx (
name=0xdfffeab0 "test.test_signal", co=0x7333e0,
pathname=0xdfffe5c0 "/mp/slaskdisk/tmp/quest/python/dist/src/Lib/test/test_signal.py") at ../python/dist/src/Python/import.c:530
#7 0x000917d4 in load_source_module (name=0xdfffeab0 "test.test_signal",
pathname=0xdfffe5c0 "/mp/slaskdisk/tmp/quest/python/dist/src/Lib/test/test_signal.py", fp=0x100c00) at ../python/dist/src/Python/import.c:800
#8 0x000920dc in load_module (name=0xdfffeab0 "test.test_signal",
fp=0x100b78,
buf=0xdfffe5c0 "/mp/slaskdisk/tmp/quest/python/dist/src/Lib/test/test_signal.py", type=1) at ../python/dist/src/Python/import.c:1465
#9 0x00093090 in import_submodule (mod=0x19f3b0,
---Type <return> to continue, or q <return> to quit---
subname=0xdfffeab5 "test_signal", fullname=0xdfffeab0 "test.test_signal")
at ../python/dist/src/Python/import.c:2004
#10 0x00092b90 in load_next (mod=0x19f3b0, altmod=0x19f3b0, p_name=0xdfffeab5,
buf=0xdfffeab0 "test.test_signal", p_buflen=0xdfffeaac)
at ../python/dist/src/Python/import.c:1860
#11 0x00092694 in import_module_ex (name=0x0, globals=0x19f3b0,
locals=0x544390, fromlist=0x66feb0)
at ../python/dist/src/Python/import.c:1718
#12 0x00092898 in PyImport_ImportModuleEx (name=0x7868a4 "test.test_signal",
globals=0x119780, locals=0x544390, fromlist=0x66feb0)
at ../python/dist/src/Python/import.c:1752
I gather this comes from line 54 in test_signal.py:
try:
54> signal.pause()
if verbose:
print "pause() returned"
except HandlerBCalled:
if verbose:
print "HandlerBCalled exception caught"
else:
pass
The execution (with -v) generates the following output:
test_signal
starting pause() loop...
call pause()...
+ sleep 2
+ kill -5 23928
+ sleep 2
+ kill -2 23928
+ sleep 2
+ kill -3 23928
Whereas executing the test alone:
test_signal
starting pause() loop...
call pause()...
+ sleep 2
+ kill -5 25192
+ sleep 2
handlerA (5, <frame object at 0x15b6c0>)
pause() returned
call pause()...
+ kill -2 25192
+ sleep 2
handlerB (2, <frame object at 0x15b6c0>)
HandlerBCalled exception caught
call pause()...
+ kill -3 25192
KeyboardInterrupt (assume the alarm() went off)
1 test OK.
machines hangs if run as a part of the test suite, but not if run on
its own. The top part of the stack trace is as follows:
#0 0xdf5da12c in _pause () from /usr/lib/libc.so.1
#1 0xdf4fa8bc in pause () from /usr/lib/libthread.so.1
#2 0x000792b0 in call_function (pp_stack=0xdfffded8, oparg=0)
at ../python/dist/src/Python/ceval.c:3214
#3 0x00076fd0 in eval_frame (f=0x1a1110)
at ../python/dist/src/Python/ceval.c:1993
#4 0x000785f0 in PyEval_EvalCodeEx (co=0x7333e0, globals=0x1a1110,
locals=0x0, args=0x0, argcount=0, kws=0x0, kwcount=0, defs=0x0,
defcount=0, closure=0x0) at ../python/dist/src/Python/ceval.c:2538
#5 0x0007b03c in PyEval_EvalCode (co=0x7333e0, globals=0x7adb70,
locals=0x7adb70) at ../python/dist/src/Python/ceval.c:478
#6 0x00091280 in PyImport_ExecCodeModuleEx (
name=0xdfffeab0 "test.test_signal", co=0x7333e0,
pathname=0xdfffe5c0 "/mp/slaskdisk/tmp/quest/python/dist/src/Lib/test/test_signal.py") at ../python/dist/src/Python/import.c:530
#7 0x000917d4 in load_source_module (name=0xdfffeab0 "test.test_signal",
pathname=0xdfffe5c0 "/mp/slaskdisk/tmp/quest/python/dist/src/Lib/test/test_signal.py", fp=0x100c00) at ../python/dist/src/Python/import.c:800
#8 0x000920dc in load_module (name=0xdfffeab0 "test.test_signal",
fp=0x100b78,
buf=0xdfffe5c0 "/mp/slaskdisk/tmp/quest/python/dist/src/Lib/test/test_signal.py", type=1) at ../python/dist/src/Python/import.c:1465
#9 0x00093090 in import_submodule (mod=0x19f3b0,
---Type <return> to continue, or q <return> to quit---
subname=0xdfffeab5 "test_signal", fullname=0xdfffeab0 "test.test_signal")
at ../python/dist/src/Python/import.c:2004
#10 0x00092b90 in load_next (mod=0x19f3b0, altmod=0x19f3b0, p_name=0xdfffeab5,
buf=0xdfffeab0 "test.test_signal", p_buflen=0xdfffeaac)
at ../python/dist/src/Python/import.c:1860
#11 0x00092694 in import_module_ex (name=0x0, globals=0x19f3b0,
locals=0x544390, fromlist=0x66feb0)
at ../python/dist/src/Python/import.c:1718
#12 0x00092898 in PyImport_ImportModuleEx (name=0x7868a4 "test.test_signal",
globals=0x119780, locals=0x544390, fromlist=0x66feb0)
at ../python/dist/src/Python/import.c:1752
I gather this comes from line 54 in test_signal.py:
try:
54> signal.pause()
if verbose:
print "pause() returned"
except HandlerBCalled:
if verbose:
print "HandlerBCalled exception caught"
else:
pass
The execution (with -v) generates the following output:
test_signal
starting pause() loop...
call pause()...
+ sleep 2
+ kill -5 23928
+ sleep 2
+ kill -2 23928
+ sleep 2
+ kill -3 23928
Whereas executing the test alone:
test_signal
starting pause() loop...
call pause()...
+ sleep 2
+ kill -5 25192
+ sleep 2
handlerA (5, <frame object at 0x15b6c0>)
pause() returned
call pause()...
+ kill -2 25192
+ sleep 2
handlerB (2, <frame object at 0x15b6c0>)
HandlerBCalled exception caught
call pause()...
+ kill -3 25192
KeyboardInterrupt (assume the alarm() went off)
1 test OK.
--
Anders "Quest" Qvist
"We've all heard that a million monkeys banging on a million typewriters
will eventually reproduce the entire works of Shakespeare. Now, thanks
to the Internet, we know this is not true." -- Robert Wilensky
Anders "Quest" Qvist
"We've all heard that a million monkeys banging on a million typewriters
will eventually reproduce the entire works of Shakespeare. Now, thanks
to the Internet, we know this is not true." -- Robert Wilensky