shibin created FLUME-3105:
-----------------------------
Summary: tmp file left on Amazon S3 while using hadoop sinker to
save log files to Amazon S3 directly
Key: FLUME-3105
URL: https://issues.apache.org/jira/browse/FLUME-3105
Project: Flume
Issue Type: Bug
Affects Versions: 1.7.0
Reporter: shibin
While using flume to sync log files to amazon s3, once in a while there is some
file left as .tmp. after dig into this problem on internet, it seems that there
are many same cases reported. And someone says that it's S3's fault, it has
some API missing. But it is really an annoying problem. Is there anything we
can do to fix it, also I think a lot of people are storing their log files to
amazon S3, so it may have already affected a lot of flume users. Thanks.
--
This message was sent by Atlassian JIRA
(v6.3.15#6346)
