thick mode : segmentation fault using connection pool and auto_login wallet · Issue #425 · oracle/python-oracledb (original) (raw)

1. What versions are you using?
   • server : Oracle Database 19c
   • client:
     * Oracle Instant Client Basic 21.16.0.0.0
     * python:

platform.platform: Linux-4.18.0-553.5.1.el8_10.x86_64-x86_64-with-glibc2.28
sys.maxsize > 2**32: True
platform.python_version: 3.13.0
oracledb.__version__: 2.5.0

1. Is it an error or a hang or a crash?
   crash
2. What error(s) or behavior you are seeing?
   Segmentation fault when creating the connection pool using oracledb==2.5.0. This error does not happen when using oracledb==2.4.1.
   • using oracledb==2.5.0

$ PYTHONFAULTHANDLER=1 python test_oracle.py
platform.platform: Linux-4.18.0-553.5.1.el8_10.x86_64-x86_64-with-glibc2.28
sys.maxsize > 2**32: True
platform.python_version: 3.13.0
oracledb.__version__: 2.5.0
Fatal Python error: Segmentation fault

Thread 0x00007fb8ecc36700 (most recent call first):
  File "/usr/local/lib/python3.13/threading.py", line 1092 in join
  File "/usr/local/lib/python3.13/threading.py", line 992 in run
  File "/usr/local/lib/python3.13/threading.py", line 1041 in _bootstrap_inner
  File "/usr/local/lib/python3.13/threading.py", line 1012 in _bootstrap

Current thread 0x00007fb8f0efa740 (most recent call first):
  File "/opt/app-root/lib/python3.13/site-packages/oracledb/connection.py", line 579 in __init__
  File "/opt/app-root/lib/python3.13/site-packages/oracledb/connection.py", line 1194 in connect
  File "/opt/app-root/lib/python3.13/site-packages/oracledb/pool.py", line 406 in acquire
  File "/home/gdraelan/test_oracle.py", line 24 in main
  File "/home/gdraelan/test_oracle.py", line 36 in <module>

Extension modules: oracledb.base_impl, oracledb.thick_impl, _cffi_backend, oracledb.thin_impl (total: 4)
Segmentation fault (core dumped)

• using oracledb==2.4.1

$ PYTHONFAULTHANDLER=1 python test_oracle.py
platform.platform: Linux-4.18.0-553.5.1.el8_10.x86_64-x86_64-with-glibc2.28
sys.maxsize > 2**32: True
platform.python_version: 3.13.0
oracledb.__version__: 2.4.1
Database version: Oracle Database 19c Enterprise Edition Release 19.0.0.0.0 - Production
Database time: 2024-12-02 12:23:47.237628

1. Does your application call init_oracle_client()?
   Yes
2. Include a runnable Python script that shows the problem.

import faulthandler; faulthandler.enable()
import oracledb
import platform
import sys

def main() -> None:
    print("platform.platform:", platform.platform())
    print("sys.maxsize > 2**32:", sys.maxsize > 2**32)
    print("platform.python_version:", platform.python_version())
    print("oracledb.__version__:", oracledb.__version__)

    oracledb.init_oracle_client()
    param = oracledb.PoolParams(
        timeout=120,
        getmode=oracledb.POOL_GETMODE_TIMEDWAIT,
        wait_timeout=5000,
        externalauth=True,
        homogeneous=False
    )

    pool = None
    try:
        pool = oracledb.create_pool(dsn='xxx', params=param)
        with pool.acquire() as conn:
            with conn.cursor() as cursor:
                row = cursor.execute("select * from v$version where banner like 'Oracle%'").fetchone()
                print(f'Database version: {str(row[0])}')
                row = cursor.execute("select systimestamp from dual").fetchone()
                print(f'Database time: {str(row[0])}')
    finally:
        if pool is not None:
            pool.close()


if __name__ == "__main__":
    main()