Python Page Rank Streaming Application using Hadoop, py4j.protocol.Py4JJavaError: An error occurred while calling o27.partitions

Question

我正在嘗試將頁面排名算法從簡單的 python 代碼傳遞到在 spark 中使用 python 的流式應用程序，在這種情況下，根據特定時間（10）秒從另一個 python 腳本中獲取輸入本來應該根據時間來拿它們並分析它們，當我運行以下代碼時出現錯誤，我不知道是什么原因導致錯誤，我試圖獲取.csv文件，

import sys
from pyspark import SparkContext
from pyspark.sql import SparkSession
from pyspark.streaming import StreamingContext

def main(input_folder_location):
sc = SparkContext.getOrCreate()
ssc = StreamingContext(sc, 10)  # Streaming will execute in each 3 seconds
ssc.checkpoint(input_folder_location)  # 'mean directory name, Directory to be checked

links = spark.sparkContext.textFile(input_folder_location). \
    map(lambda line: line.split(',')). \
    map(lambda pages: (pages[0], pages[1])). \
    distinct(). \
    groupByKey(). \
    map(lambda x: (x[0], list(x[1])))

ranks = links.map(lambda element: (element[0], 1.0))
# iterations = int(sys.argv[3])
iterations = 4

    for x in range(iterations + 1):
    contribs = links.join(ranks).flatMap(lambda row: computeContribs(row[1][0], row[1][1]))
    print("\n")
    print("------- Iter: " + str(x) + " --------")
    ranks = contribs.reduceByKey(lambda v1, v2: v1 + v2).map(lambda x: (x[0], x[1] * 0.85 + 0.15))
    for rank in ranks.collect():
        print(rank)
    print("\n")

print("------- Final Results --------")
for rank in ranks.collect():
    print(rank)

ssc.start()
ssc.awaitTermination()

def computeContribs(neighbors, rank):
for neighbor in neighbors:
    yield (neighbor, rank / len(neighbors))

if __name__ == "__main__":
if len(sys.argv) < 2:
    sys.stderr.write(
        "Error: Usage: StreamingApp.py <input-file-directory>")
    sys.exit()

spark = SparkSession.builder.getOrCreate()
spark.sparkContext.setLogLevel("WARN")

main(sys.argv[1])

Answer 1

import sys
from pyspark import SparkContext
from pyspark.sql import SparkSession
from pyspark.streaming import StreamingContext

def computeContribs(neighbors, rank):
    for neighbor in neighbors:
        yield (neighbor, rank / len(neighbors))
def main(input_folder_location):
sc = SparkContext.getOrCreate()
ssc = StreamingContext(sc, 3)   #Streaming will execute in each 3 seconds
lines = ssc.textFileStream(input_folder_location)  #'log/ mean directory name
print(" hello from app")
counts = lines.map(lambda line: line.split(",")) \
.map(lambda pages:(pages[0],pages[1])) \
.transform(lambda rdd: rdd.distinct()) \
.groupByKey() \
.map(lambda x: (x[0],list(x[1])))

  ranks = counts.map(lambda element:(element[0],1.0))
iterations = 5
for x in range(1):
    contribs = counts.join(ranks).flatMap(lambda row: computeContribs(row[1][0],row[1][1]))
    print("\n")
    print(" iter --------------",x)
    ranks = contribs.reduceByKey(lambda v1,v2:v1+v2)
    print("\n")
#counts.pprint()
ranks.pprint()

print(" finishing the task")
ssc.start()
ssc.awaitTermination()


if __name__ == "__main__":
     if len(sys.argv) < 2:
        sys.stderr.write(
        "Error: Usage: StreamingApp.py <input-file-directory>")
        sys.exit()

spark = SparkSession.builder.getOrCreate()
spark.sparkContext.setLogLevel("WARN")

main(sys.argv[1])