HBase-HappyBase : Socket Timeout Error For Larger Files - Works Good With Smaller one's

Question

I use following piece of python code using happybase module to update hbase. This works perfectly for a file less than 30k records. But throws timeout error when exceeds 30k-35k. I tried options informed in other stack questions by editing hbase_site.xml and few other stuffs. But no help. Did anyone come across same issue ?

import happybase as hb

def loadIdPHSegmentPyBase() :
s = socket.socket()
s.settimeout(300)
connection = hb.Connection('XXXXX',9090,timeout=None,compat='0.92',transport='buffered')
table = connection.table('HBASE_D_L')
ReqFileToLoad = ("%segment.txt" %(dirName))
b = table.batch()
with open('%s' %(ReqFileToLoad)) as ffile1 :
     for line in ffile1 :
         line = line.strip()
         line = line.split('|')
         #print line[7] ,
         if line[7] == 'PH' :
           b.put(line[0],{'ADDR_IDPH:PHMIDDLE_NAME':line[1],'ADDR_IDPH:PHSUR_NAME' :line[2],'ADDR_IDPH:PHFIRST_NAME' :line[3],'ADDR_IDPH:PHFILLER1' :line[4],'ADDR_IDPH:PHFILLER2' :line[5],'ADDR_IDPH:PHFILLER3' :line[6],'ADDR_IDPH:TELEPHONE_SUBSEGMENT_ID' :line[7],'ADDR_IDPH:TELEPHONE_TYPE_CODE' :line[8],'ADDR_IDPH:PUBLISHED_INDICATOR' :line[9],'ADDR_IDPH:TELEPHONE_NUMBER' :line[10]})
         else :
           b.put(line[0],{'ADDR_IDPH:IDMIDDLE_NAME':line[1],'ADDR_IDPH:IDSUR_NAME' :line[2],'ADDR_IDPH:IDFIRST_NAME' :line[3],'ADDR_IDPH:IDFILLER1' :line[4],'ADDR_IDPH:IDFILLER2' :line[5],'ADDR_IDPH:IDFILLER3' :line[6],'ADDR_IDPH:IDSUBSEGMENT_IDENTIFIER' :line[7],'ADDR_IDPH:ID_TYPE' :line[8],'ADDR_IDPH:ID_VALIDITY_INDICATOR' :line[9],'ADDR_IDPH:ID_VALUE' :line[11]})
b.send()
s.close()

My error with larger files :

 File "thriftpy/protocol/cybin/cybin.pyx", line 429, in     cybin.TCyBinaryProtocol.read_message_begin (thriftpy/protocol/cybin/cybin.c:6325)
  File "thriftpy/protocol/cybin/cybin.pyx", line 60, in cybin.read_i32 (thriftpy/protocol/cybin/cybin.c:1546)
  File "thriftpy/transport/buffered/cybuffered.pyx", line 65, in thriftpy.transport.buffered.cybuffered.TCyBufferedTransport.c_read (thriftpy/transport/buffered/cybuffered.c:1881)
  File "thriftpy/transport/buffered/cybuffered.pyx", line 69, in thriftpy.transport.buffered.cybuffered.TCyBufferedTransport.read_trans (thriftpy/transport/buffered/cybuffered.c:1948)
  File "thriftpy/transport/cybase.pyx", line 61, in thriftpy.transport.cybase.TCyBuffer.read_trans (thriftpy/transport/cybase.c:1472)
  File "/usr/local/python27/lib/python2.7/site-packages/thriftpy/transport/socket.py", line 108, in read
    buff = self.sock.recv(sz)
socket.timeout: timed out

---

This was how it got resolved :

with open('%s' %(ReqFileToLoad)) as ffile1 :
     for line in ffile1 :
         line = line.strip()
         line = line.split('|')
         #print line[7] ,
         if line[7] == 'PH' :
           b = table.batch()
           b.put(line[0],{'ADDR_IDPH:PHMIDDLE_NAME':line[1],'ADDR_IDPH:PHSUR_NAME' :line[2],'ADDR_IDPH:PHFIRST_NAME' :line[3],'ADDR_IDPH:PHFILLER1' :line[4],'ADDR_IDPH:PHFILLER2' :line[5],'ADDR_IDPH:PHFILLER3' :line[6],'ADDR_IDPH:TELEPHONE_SUBSEGMENT_ID' :line[7],'ADDR_IDPH:TELEPHONE_TYPE_CODE' :line[8],'ADDR_IDPH:PUBLISHED_INDICATOR' :line[9],'ADDR_IDPH:TELEPHONE_NUMBER' :line[10]})
         else :
           b = table.batch()
           b.put(line[0],{'ADDR_IDPH:IDMIDDLE_NAME':line[1],'ADDR_IDPH:IDSUR_NAME' :line[2],'ADDR_IDPH:IDFIRST_NAME' :line[3],'ADDR_IDPH:IDFILLER1' :line[4],'ADDR_IDPH:IDFILLER2' :line[5],'ADDR_IDPH:IDFILLER3' :line[6],'ADDR_IDPH:IDSUBSEGMENT_IDENTIFIER' :line[7],'ADDR_IDPH:ID_TYPE' :line[8],'ADDR_IDPH:ID_VALIDITY_INDICATOR' :line[9],'ADDR_IDPH:ID_VALUE' :line[11]})
b.send()

Answer 1

i suggest that you use smaller batch sizes, or that you do not use a batch at all. batching is a client-side buffer without any limits, so it can cause huge thrift requests when it is sent. happybase also provides a helper for this: you can specify batch_size and the batch will be periodically flushed.

https://happybase.readthedocs.io/en/latest/api.html#happybase.Table.batch