Skip to content
This repository was archived by the owner on May 13, 2025. It is now read-only.
This repository was archived by the owner on May 13, 2025. It is now read-only.

PropertyGraph test_property_graph_mg.py::test_add_edge_data_with_ids seg faults when restricted to a single-GPU #13

@rlratzel

Description

@rlratzel

This is still light on details while being investigated more, but for now when the test is run using:

$> CUDA_VISIBLE_DEVICES=0 pytest -svx ./python/cugraph/cugraph/tests/data_store/test_property_graph_mg.py

A seg fault happens which produces the following stack trace:

python/cugraph/cugraph/tests/data_store/test_property_graph_mg.py::test_add_edge_data_with_ids[True] 2023-05-31 03:02:54,239 - distributed.preloading - INFO - Creating preload: dask_cuda.
initialize
2023-05-31 03:02:54,239 - distributed.preloading - INFO - Import preload module: dask_cuda.initialize

Dask client/cluster created using LocalCUDACluster
[ce950e66fa0c:104  :0:136] Caught signal 11 (Segmentation fault: address not mapped to object at address 0x1bd8)
==== backtrace (tid:    136) ====
 0  /usr/local/lib/libucs.so.0(ucs_handle_error+0x2dc) [0x7f6a70899c1c]
 1  /usr/local/lib/libucs.so.0(+0x30dff) [0x7f6a70899dff]
 2  /usr/local/lib/libucs.so.0(+0x31134) [0x7f6a7089a134]
 3  /usr/lib/x86_64-linux-gnu/libpthread.so.0(+0x14420) [0x7f6aff641420]
 4  /opt/conda/lib/python3.9/site-packages/cubinlinker/_nvlinklib.cpython-39-x86_64-linux-gnu.so(+0x358d40) [0x7f6a2cc0bd40]
 5  /opt/conda/lib/python3.9/site-packages/cubinlinker/_nvlinklib.cpython-39-x86_64-linux-gnu.so(__ptx37960+0x29) [0x7f6a2cc0ca49]
 6  /opt/conda/lib/python3.9/site-packages/cubinlinker/_nvlinklib.cpython-39-x86_64-linux-gnu.so(+0x3639f7) [0x7f6a2cc169f7]
 7  /opt/conda/lib/python3.9/site-packages/cubinlinker/_nvlinklib.cpython-39-x86_64-linux-gnu.so(+0x3639f7) [0x7f6a2cc169f7]
 8  /opt/conda/lib/python3.9/site-packages/cubinlinker/_nvlinklib.cpython-39-x86_64-linux-gnu.so(+0x3639f7) [0x7f6a2cc169f7]
 9  /opt/conda/lib/python3.9/site-packages/cubinlinker/_nvlinklib.cpython-39-x86_64-linux-gnu.so(+0x3639f7) [0x7f6a2cc169f7]
10  /opt/conda/lib/python3.9/site-packages/cubinlinker/_nvlinklib.cpython-39-x86_64-linux-gnu.so(+0x359d14) [0x7f6a2cc0cd14]
11  /opt/conda/lib/python3.9/site-packages/cubinlinker/_nvlinklib.cpython-39-x86_64-linux-gnu.so(+0x359d50) [0x7f6a2cc0cd50]
12  /opt/conda/lib/python3.9/site-packages/cubinlinker/_nvlinklib.cpython-39-x86_64-linux-gnu.so(__ptx37934+0x8e) [0x7f6a2cc0a7be]
13  /opt/conda/lib/python3.9/site-packages/cubinlinker/_nvlinklib.cpython-39-x86_64-linux-gnu.so(__ptx37958+0xb6) [0x7f6a2cc0d456]
14  /opt/conda/lib/python3.9/site-packages/cubinlinker/_nvlinklib.cpython-39-x86_64-linux-gnu.so(__cuda_CallJitEntryPoint+0x10ad) [0x7f6a2cc0faed]
15  /opt/conda/lib/python3.9/site-packages/cubinlinker/_nvlinklib.cpython-39-x86_64-linux-gnu.so(nvLinkerDestroy+0x15d) [0x7f6a2cbf218d]
16  /opt/conda/lib/python3.9/site-packages/cubinlinker/_nvlinklib.cpython-39-x86_64-linux-gnu.so(+0x33e885) [0x7f6a2cbf1885]
17  /opt/conda/bin/python3.9(+0x149df0) [0x557cb21bddf0]
18  /opt/conda/bin/python3.9(PyObject_Call+0x158) [0x557cb21bc1a8]
19  /opt/conda/bin/python3.9(_PyEval_EvalFrameDefault+0x5ba6) [0x557cb21a2d36]
20  /opt/conda/bin/python3.9(+0x12827a) [0x557cb219c27a]
21  /opt/conda/bin/python3.9(_PyObject_FastCallDictTstate+0x13e) [0x557cb21a5b1e]
22  /opt/conda/bin/python3.9(_PyObject_Call_Prepend+0x69) [0x557cb21b9209]
23  /opt/conda/bin/python3.9(+0x2110d9) [0x557cb22850d9]
24  /opt/conda/bin/python3.9(_PyObject_MakeTpCall+0x2fc) [0x557cb21a62bc]
25  /opt/conda/bin/python3.9(+0x1da59f) [0x557cb224e59f]
26  /opt/conda/bin/python3.9(+0x1ddedf) [0x557cb2251edf]
27  /opt/conda/bin/python3.9(+0x14c05c) [0x557cb21c005c]
28  /opt/conda/bin/python3.9(+0x1259e6) [0x557cb21999e6]
29  /opt/conda/bin/python3.9(+0x14bc74) [0x557cb21bfc74]
30  /opt/conda/bin/python3.9(+0x131ec8) [0x557cb21a5ec8]
31  /opt/conda/bin/python3.9(+0x1282b0) [0x557cb219c2b0]
32  /opt/conda/bin/python3.9(+0x147935) [0x557cb21bb935]
33  /opt/conda/bin/python3.9(_PyEval_EvalFrameDefault+0x123b) [0x557cb219e3cb]
34  /opt/conda/bin/python3.9(+0x139ef3) [0x557cb21adef3]
35  /opt/conda/bin/python3.9(_PyEval_EvalFrameDefault+0x68b) [0x557cb219d81b]
36  /opt/conda/bin/python3.9(+0x139ef3) [0x557cb21adef3]
37  /opt/conda/bin/python3.9(_PyEval_EvalFrameDefault+0x68b) [0x557cb219d81b]
38  /opt/conda/bin/python3.9(+0x139ef3) [0x557cb21adef3]
39  /opt/conda/bin/python3.9(_PyEval_EvalFrameDefault+0x68b) [0x557cb219d81b]
40  /opt/conda/bin/python3.9(+0x12827a) [0x557cb219c27a]
41  /opt/conda/bin/python3.9(_PyFunction_Vectorcall+0xd5) [0x557cb21adc05]
42  /opt/conda/bin/python3.9(+0x147a16) [0x557cb21bba16]
43  /opt/conda/bin/python3.9(_PyEval_EvalFrameDefault+0x402b) [0x557cb21a11bb]
44  /opt/conda/bin/python3.9(+0x12827a) [0x557cb219c27a]
45  /opt/conda/bin/python3.9(_PyObject_FastCallDictTstate+0x13e) [0x557cb21a5b1e]
46  /opt/conda/bin/python3.9(_PyObject_Call_Prepend+0x69) [0x557cb21b9209]
47  /opt/conda/bin/python3.9(+0x2110d9) [0x557cb22850d9]
48  /opt/conda/bin/python3.9(_PyObject_MakeTpCall+0x2fc) [0x557cb21a62bc]
49  /opt/conda/bin/python3.9(_PyEval_EvalFrameDefault+0x4d62) [0x557cb21a1ef2]
50  /opt/conda/bin/python3.9(+0x12827a) [0x557cb219c27a]
51  /opt/conda/bin/python3.9(_PyFunction_Vectorcall+0xd5) [0x557cb21adc05]
52  /opt/conda/bin/python3.9(_PyEval_EvalFrameDefault+0x123b) [0x557cb219e3cb]
53  /opt/conda/bin/python3.9(+0x12827a) [0x557cb219c27a]
54  /opt/conda/bin/python3.9(_PyFunction_Vectorcall+0xd5) [0x557cb21adc05]
55  /opt/conda/bin/python3.9(_PyEval_EvalFrameDefault+0x123b) [0x557cb219e3cb]
56  /opt/conda/bin/python3.9(+0x139ef3) [0x557cb21adef3]
57  /opt/conda/bin/python3.9(_PyEval_EvalFrameDefault+0x402b) [0x557cb21a11bb]
58  /opt/conda/bin/python3.9(+0x12827a) [0x557cb219c27a]
59  /opt/conda/bin/python3.9(_PyFunction_Vectorcall+0xd5) [0x557cb21adc05]
60  /opt/conda/bin/python3.9(_PyEval_EvalFrameDefault+0x68b) [0x557cb219d81b]
61  /opt/conda/bin/python3.9(+0x12827a) [0x557cb219c27a]
=================================
2023-05-31 03:03:13,384 - distributed.nanny - WARNING - Restarting worker
2023-05-31 03:03:14,600 - distributed.preloading - INFO - Creating preload: dask_cuda.initialize
2023-05-31 03:03:14,600 - distributed.preloading - INFO - Import preload module: dask_cuda.initialize
2023-05-31 03:03:17,364 - distributed.worker - WARNING - Compute Failed
Key:       ('join-indexed-3730f9db80b2f48b498c8c2119a7c40d', 1)
Function:  merge_chunk
args:      (           volume          time  card_num card_type  _SRC_  _DST_        _TYPE_
_EDGE_ID_
12           72.0  1.639085e+09      4321     DEBIT  78634     16  transactions, Empty DataFrame
Columns: [relationship_type, _SRC_, _DST_, _TYPE_]
Index: [])
kwargs:    {'how': 'outer', 'suffixes': ('', '_NEW_'), 'indicator': False, 'left_on': None, 'right_on': None, 'left_index': True, 'right_index': True, 'result_meta':            volume  time  card_num card_type  _SRC_  _DST_        _TYPE_  relationship_type  _SRC__NEW_  _DST__NEW_   _TYPE__NEW_
_EDGE_ID_
0             0.0   0.0         0       cat      0      0  transactions                  0           0           0  transactions
1             1.0   1.0         1       dog      1      1  transactions                  1           1           1  transactions}
Exception: 'TypeError("No dispatch for <class \'cudf.core.series.Series\'>")'

2023-05-31 03:03:17,411 - distributed.worker - WARNING - Compute Failed
Key:       ('join-indexed-3730f9db80b2f48b498c8c2119a7c40d', 2)
Function:  merge_chunk
args:      (           volume          time  card_num card_type  _SRC_  _DST_        _TYPE_
_EDGE_ID_
13          103.2  1.639085e+09     98124         V  32431      4  transactions, Empty DataFrame
Columns: [relationship_type, _SRC_, _DST_, _TYPE_]
Index: [])
kwargs:    {'how': 'outer', 'suffixes': ('', '_NEW_'), 'indicator': False, 'left_on': None, 'right_on': None, 'left_index': True, 'right_index': True, 'result_meta':            volume  time  card_num card_type  _SRC_  _DST_        _TYPE_  relationship_type  _SRC__NEW_  _DST__NEW_   _TYPE__NEW_
_EDGE_ID_
0             0.0   0.0         0       cat      0      0  transactions                  0           0           0  transactions
1             1.0   1.0         1       dog      1      1  transactions                  1           1           1  transactions}
Exception: 'TypeError("No dispatch for <class \'cudf.core.series.Series\'>")'

FAILED2023-05-31 03:03:47,792 - distributed.scheduler - ERROR - broadcast to tcp://127.0.0.1:32877 failed: OSError: Timed out trying to connect to tcp://127.0.0.1:32877 after 30 s
distributed.comm.core.CommClosedError: in <distributed.comm.tcp.TCPConnector object at 0x7fafa8b9c790>: ConnectionRefusedError: [Errno 111] Connection refused

Metadata

Metadata

Assignees

No one assigned

    Labels

    bugSomething isn't working

    Type

    No type

    Projects

    No projects

    Milestone

    No milestone

    Relationships

    None yet

    Development

    No branches or pull requests

    Issue actions