[
https://issues.apache.org/jira/browse/HADOOP-10195?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=14288022#comment-14288022
]
Steve Loughran commented on HADOOP-10195:
-----------------------------------------
I had lots of problems with socket timeouts on long-haul work; after a test
failure of {{testLongDirectoryList()}} I was
invariably left with lots of files of the form {{testfile-00004}}, which then
led to follow-on failures
# move {{testLongDirectoryList()}} to its own class, something like
ScaleTestLs; to be run with any other scale tests {{mvn test
-Dtest=ScaleTest\*}}
# test files to be created in a subdir
# setup operation to delete this directory before recreating it
# finally operation to delete this directory afterwards
> swiftfs object list stops at 10000 objects
> ------------------------------------------
>
> Key: HADOOP-10195
> URL: https://issues.apache.org/jira/browse/HADOOP-10195
> Project: Hadoop Common
> Issue Type: Bug
> Components: fs
> Affects Versions: 2.3.0, 2.6.0
> Reporter: David Dobbins
> Assignee: David Dobbins
> Attachments: hadoop-10195.patch, hadoop-10195.patch
>
>
> listing objects in a container in swift is limited to 10000 objects per
> request. swiftfs only makes one request and is therefore limited to the first
> 10000 objects in the container, ignoring any remaining objects
--
This message was sent by Atlassian JIRA
(v6.3.4#6332)
