[
https://issues.apache.org/jira/browse/HDFS-7929?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=14364411#comment-14364411
]
Colin Patrick McCabe commented on HDFS-7929:
--------------------------------------------
The approach looks reasonable here.
{code}
131 File[] files = tmpDir.listFiles();
132 if (files != null) {
{code}
We should use {{IOUtils#listDirectory}} here. That function will throw an
exception if there is an IO error; in contrast, {{File#listFiles}} will
silently hide the error and just return null.
I am +1 pending that (and getting a good jenkins run)
We might want to file a follow-up JIRA for looking into whether we need any
changes in {{BackupJournalManager}} or {{BookKeeperJournalManager}}. Of
course, I think in most cases people configure local file journal managers to
go along with those journal managers, so it probably won't be an issue for most
systems.
> inotify unable fetch pre-upgrade edit log segments once upgrade starts
> ----------------------------------------------------------------------
>
> Key: HDFS-7929
> URL: https://issues.apache.org/jira/browse/HDFS-7929
> Project: Hadoop HDFS
> Issue Type: Bug
> Reporter: Zhe Zhang
> Assignee: Zhe Zhang
> Attachments: HDFS-7929-000.patch, HDFS-7929-001.patch
>
>
> inotify is often used to periodically poll HDFS events. However, once an HDFS
> upgrade has started, edit logs are moved to /previous on the NN, which is not
> accessible. Moreover, once the upgrade is finalized /previous is currently
> lost forever.
--
This message was sent by Atlassian JIRA
(v6.3.4#6332)
