[
https://issues.apache.org/jira/browse/HDFS-3869?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
]
Todd Lipcon updated HDFS-3869:
--
Attachment: hdfs-3869.txt
Found one small bug in this patch -- accidentally had synchronized on the inne
[
https://issues.apache.org/jira/browse/HDFS-3869?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
]
Todd Lipcon updated HDFS-3869:
--
Attachment: hdfs-3869.txt
This actually depends on HDFS-3863, so not committing until that one is in.
[
https://issues.apache.org/jira/browse/HDFS-3869?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
]
Todd Lipcon updated HDFS-3869:
--
Attachment: hdfs-3869.txt
Attached patch has a little cleanup (formatting and javadoc) and also adds the
[
https://issues.apache.org/jira/browse/HDFS-3869?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
]
Todd Lipcon updated HDFS-3869:
--
Attachment: lagging-jn.png
dir-failed.png
open-for-write.png
[
https://issues.apache.org/jira/browse/HDFS-3869?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
]
Todd Lipcon updated HDFS-3869:
--
Attachment: hdfs-3869.txt
Attached preliminary patch. In addition to the unit test, I ran some cluster