Issue 124 in python-graph: test_find_cycle_on_very_deep_graph SIGSEGV

[python...@googlecode.com]

Status: New
Owner: ----
Labels: Type-Defect Priority-Medium

New issue 124 by sandro.tosi: test_find_cycle_on_very_deep_graph SIGSEGV
https://code.google.com/p/python-graph/issues/detail?id=124

Hello,
when running the test suite on a Debian system, we got this failure:

test_find_cycle_on_very_deep_graph (unittests-cycles.test_find_cycle) ...
Segmentation fault



--
You received this message because this project is configured to send all
issue notifications to this address.
You may adjust your notification preferences at:
https://code.google.com/hosting/settings

[python...@googlecode.com]

Comment #1 on issue 124 by sandro.tosi: test_find_cycle_on_very_deep_graph
SIGSEGV
https://code.google.com/p/python-graph/issues/detail?id=124

Some additional info from Debian bug
https://bugs.debian.org/cgi-bin/bugreport.cgi?bug=750362


(The full build log is attached.) Upon inspection via gdb it becomes
apparent
that this is caused by excessive recursion, which eventually causes a stack
overflow. This is due to the following lines:

http://sources.debian.net/src/python-graph/1.8.2-4/tests/unittests-sorting.py?hl=82,83#L82

Changing the numbers to 16876 and 16875, respectively, was the maximum I
got to
work.

[python...@googlecode.com]

Comment #2 on issue 124 by hgkam...@gmail.com:
test_find_cycle_on_very_deep_graph SIGSEGV
https://code.google.com/p/python-graph/issues/detail?id=124

confirming same on attempting to build fedora-21

[python...@googlecode.com]

Comment #3 on issue 124 by hgkam...@gmail.com:
test_find_cycle_on_very_deep_graph SIGSEGV
https://code.google.com/p/python-graph/issues/detail?id=124

$ ulimit -s 65536
is another way around.