Gavin Zhang created SPARK-3346:
----------------------------------
Summary: Error happened in using memcached
Key: SPARK-3346
URL: https://issues.apache.org/jira/browse/SPARK-3346
Project: Spark
Issue Type: Bug
Environment: CDH5.1.0
Spark1.0.0
Reporter: Gavin Zhang
I finished a distributed project in hadoop streaming and it worked fine with
using memcached storage during mapping. Actually, it's a python project.
Now I want to move it to Spark. But when I called the memcached library, two
errors was found during computing. (Both)
1. File "memcache.py", line 414, in get
rkey, rlen = self._expectvalue(server)
ValueError: too many values to unpack
2. File "memcache.py", line 714, in check_key
return key.translate(ill_map)
TypeError: character mapping must return integer, None or unicode
After adding exception handing, there was no successful cache got at all.
However, it works in hadoop streaming without any error. Why?
--
This message was sent by Atlassian JIRA
(v6.3.4#6332)
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
