错误：(‘’s 005‘，"[IM005] [unixODBC][Driver Manager] SQL_HANDLE_DBC上的驱动程序的SQLAllocHandle失败(0) (SQLDriverConnect)")

Question

我正在我的pyodbc脚本中运行这段代码，在那里我试图进行并行处理。

templst = [lineitem, orders, partsupp, region, cur_cur, T1, T2]
connstr = [DRIVER={libdbodbc17.so};host=lint16muthab.phl.sap.corp:8766;UID=dba;PWD=sql;CharSet=utf8, DRIVER={libdbodbc17.so};host=localhost:8767;UID=dba;PWD=sql;CharSet=utf8,
           DRIVER={libdbodbc17.so};host=localhost:8768;UID=dba;PWD=sql;CharSet=utf8,
           DRIVER={libdbodbc17.so};host=localhost:8769;UID=dba;PWD=sql;CharSet=utf8, DRIVER={libdbodbc17.so};host=localhost:8770;UID=dba;PWD=sql;CharSet=utf8]

def extract_single(q, cursorconn):
    while True:
        try:
            tableName = q.get_nowait()
            time.sleep(3)
            qry2 = "Select * FROM %s"% (tableName)
            print " extraction done of table:%s done by cursor:%s"%(tableName,cursorconn)
        except Queue.Empty:
            return

def main():
    q = multiprocessing.Queue()
    for item in templst:
        q.put(item) # add items to queue
    process = []
    for i in xrange(5):
        p = multiprocessing.Process(target=extract_single, args=(q, connstr[i]))
        process.append(p)
        p.start()
    for p in process:
        p.join()

if __name__ == '__main__':
    main()

产出如下：

  extraction done of table:lineitem done by cursor:DRIVER={libdbodbc17.so};host=lint16muthab.phl.sap.corp:8766;UID=dba;PWD=sql;CharSet=utf8

表的提取:由cursor:DRIVER={libdbodbc17.so};host=localhost:8767;UID=dba;PWD=sql;CharSet=utf8完成的订单

      extraction done of table:partsupp done by cursor:DRIVER={libdbodbc17.so};host=localhost:8768;UID=dba;PWD=sql;CharSet=utf8

      extraction done of table:region done by cursor:DRIVER={libdbodbc17.so};host=localhost:8769;UID=dba;PWD=sql;CharSet=utf8

      extraction done of table:cur_cur done by cursor:DRIVER={libdbodbc17.so};host=localhost:8770;UID=dba;PWD=sql;CharSet=utf8

     extraction done of table:T2 done by cursor:DRIVER={libdbodbc17.so};host=localhost:8767;UID=dba;PWD=sql;CharSet=utf8

     extraction done of table:T1 done by cursor:DRIVER={libdbodbc17.so};host=lint16muthab.phl.sap.corp:8766;UID=dba;PWD=sql;CharSet=utf8

但是当我将提取函数修改为

def extract_single(q, cursorconn):
    while True:
        try:
            tableName = q.get_nowait()
            time.sleep(3)
            conn = pyodbc.connect(cursorconn, timeout=0)
            cursor = connvar.cursor()
            qry2 = "Select * FROM %s"% (tableName)
            cursor.execute(qry2).fetchall()
            print " extraction done of table:%s done by cursor:%s"%(tableName,cursorconn)
        except Queue.Empty:
            return

我收到错误，无法打开连接和游标，以便执行此查询并将数据转储到文件中。

2Process Process-1:
Traceback (most recent call last):
  File "/sybopt/software/python/python/lib/python2.7/multiprocessing/process.py", line 258, in _bootstrap
    self.run()
  File "/sybopt/software/python/python/lib/python2.7/multiprocessing/process.py", line 114, in run
    self._target(*self._args, **self._kwargs)
  File "conn3.py", line 86, in extract_single
Process Process-2:
Traceback (most recent call last):
    connvar = pyodbc.connect(cursorconn, timeout=0)
  File "/sybopt/software/python/python/lib/python2.7/multiprocessing/process.py", line 258, in _bootstrap
Error: ('IM005', "[IM005] [unixODBC][Driver Manager]Driver's SQLAllocHandle on SQL_HANDLE_DBC failed (0) (SQLDriverConnect)")
    self.run()
  File "/sybopt/software/python/python/lib/python2.7/multiprocessing/process.py", line 114, in run
    self._target(*self._args, **self._kwargs)
  File "conn3.py", line 86, in extract_single
    connvar = pyodbc.connect(cursorconn, timeout=0)
Error: ('IM005', "[IM005] [unixODBC][Driver Manager]Driver's SQLAllocHandle on SQL_HANDLE_DBC failed (0) (SQLDriverConnect)")
Process Process-3:
Traceback (most recent call last):
  File "/sybopt/software/python/python/lib/python2.7/multiprocessing/process.py", line 258, in _bootstrap
    self.run()
  File "/sybopt/software/python/python/lib/python2.7/multiprocessing/process.py", line 114, in run
    self._target(*self._args, **self._kwargs)
  File "conn3.py", line 86, in extract_single
    connvar = pyodbc.connect(cursorconn, timeout=0)

为了所有的人脉。请让我知道如何解决这个错误。使用游标对象时获取错误

我希望使用以列表形式存储的游标对象，而不是再次打开新连接，每次打开游标。策展人看起来就像

<pyodbc.Cursor object at 0x7fcd9b47f270> <pyodbc.Cursor object at 0x7fcd9b47f330> <pyodbc.Cursor object at 0x7fcd9b47f390> <pyodbc.Cursor object at 0x7fcd9b47f3f0> <pyodbc.Cursor object at 0x7fcd9b47f450>

因此，我有5个游标对象，表列表与上面相同。

def extract_single(q, cursorconn):
    while True:
        try:
            tableName = q.get_nowait()

            qry2 = "Select count(*) FROM %s"%(tableName)
            cursorconn.execute(qry2)
            rowcnt = cursorconn.fetchone()[0]
            print " rows in tempdsc=",rowcnt
            print " extraction done of table:%s done by cursor:%s"%(tableName,cursorconn)

        except Queue.Empty:
            return



def main():
    q = multiprocessing.Queue()
    for item in tempdsclst:
        q.put(item) # add items to queue
    process = []
    for i in xrange(5):
        p = multiprocessing.Process(target=extract_single, args=(q, curlst[i]))
        process.append(p)
        p.start()
    for p in process:
        p.join()

它给出的错误就像

Process Process-1:
Traceback (most recent call last):
  File "/sybopt/software/python/python/lib/python2.7/multiprocessing/process.py", line 258, in _bootstrap
Process Process-2:
Traceback (most recent call last):
    self.run()
  File "/sybopt/software/python/python/lib/python2.7/multiprocessing/process.py", line 258, in _bootstrap
  File "/sybopt/software/python/python/lib/python2.7/multiprocessing/process.py", line 114, in run
    self._target(*self._args, **self._kwargs)
  File "conn4.py", line 87, in extract_single
    cursorconn.execute(qry2)
Error: ('HY000', 'The driver did not supply an error!')
    self.run()
  File "/sybopt/software/python/python/lib/python2.7/multiprocessing/process.py", line 114, in run
    self._target(*self._args, **self._kwargs)
  File "conn4.py", line 87, in extract_single
    cursorconn.execute(qry2)
Error: ('HY000', 'The driver did not supply an error!')
Process Process-3:
Traceback (most recent call last):
  File "/sybopt/software/python/python/lib/python2.7/multiprocessing/process.py", line 258, in _bootstrap
    self.run()
  File "/sybopt/software/python/python/lib/python2.7/multiprocessing/process.py", line 114, in run
    self._target(*self._args, **self._kwargs)
  File "conn4.py", line 87, in extract_single
    cursorconn.execute(qry2)
Error: ('HY000', 'The driver did not supply an error!')
Process Process-4:
Traceback (most recent call last):
  File "/sybopt/software/python/python/lib/python2.7/multiprocessing/process.py", line 258, in _bootstrap
    self.run()
  File "/sybopt/software/python/python/lib/python2.7/multiprocessing/process.py", line 114, in run
    self._target(*self._args, **self._kwargs)
  File "conn4.py", line 87, in extract_single
    cursorconn.execute(qry2)
Error: ('HY000', 'The driver did not supply an error!')
Process Process-5:
Traceback (most recent call last):
  File "/sybopt/software/python/python/lib/python2.7/multiprocessing/process.py", line 258, in _bootstrap
    self.run()
  File "/sybopt/software/python/python/lib/python2.7/multiprocessing/process.py", line 114, in run
    self._target(*self._args, **self._kwargs)
  File "conn4.py", line 87, in extract_single
    cursorconn.execute(qry2)
Error: ('HY000', 'The driver did not supply an error!')

Answer 1

除了您正在遇到的问题之外，基于您先前的问题，我假设cursorconn是一个已经打开的连接，将对每个查询重用。由于它现在似乎只是一个连接字符串(尽管在您的代码中它似乎不是一个实际的字符串，对于列表templst？中的项来说是相同的情况)，那么您应该在while True:语句之前进行调用conn = pyodbc.connect(cursorconn, timeout=0)，以便可以将相同的连接重用到多个查询中，然后在从函数返回之前关闭连接。

我认为问题在于cursor = connvar.cursor()语句，它应该是：cursor = conn.cursor()。因此，请尝试以下几点：

conn = pyodbc.connect(cursorconn, timeout=0)
while True:
    try:
        tableName = q.get_nowait()
        time.sleep(3) # why is this here?
        cursor = conn.cursor()
        qry2 = "Select * FROM %s"% (tableName)
        cursor.execute(qry2).fetchall()
        print " extraction done of table:%s done by cursor:%s"%(tableName,cursorconn)
        cursor.close() # should probably add this
    except Queue.Empty:
        return
    finally:
        conn.close()