[
https://issues.apache.org/jira/browse/HDFS-14235?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=16771054#comment-16771054
]
Surendra Singh Lilhore commented on HDFS-14235:
-----------------------------------------------
Thanks [~RANith] for patch.
1. No need to create sleep method, keep logic inside the if condition which
will check null for fsdataset.
{code:java}
if (dn.getFSDataset() != null) {
............
............
............
}
try {
Thread.sleep(detectionInterval);
} catch (InterruptedException e) {
LOG.error("Disk Outlier Detection thread interrupted", e);
Thread.currentThread().interrupt();
}
{code}
2. Fix the failed test cases. When the {{addSlowDiskForTesting()}} is called
from UT, daemon thread should not override the status.
3. Fix check-style warnings
> Handle ArrayIndexOutOfBoundsException in
> DataNodeDiskMetrics#slowDiskDetectionDaemon
> -------------------------------------------------------------------------------------
>
> Key: HDFS-14235
> URL: https://issues.apache.org/jira/browse/HDFS-14235
> Project: Hadoop HDFS
> Issue Type: Improvement
> Reporter: Surendra Singh Lilhore
> Assignee: Ranith Sardar
> Priority: Major
> Attachments: HDFS-14235.000.patch, NPE.png, exception.png
>
>
> below code throwing exception because {{volumeIterator.next()}} called two
> time without checking hashNext().
> {code:java}
> while (volumeIterator.hasNext()) {
> FsVolumeSpi volume = volumeIterator.next();
> DataNodeVolumeMetrics metrics = volumeIterator.next().getMetrics();
> String volumeName = volume.getBaseURI().getPath();
> metadataOpStats.put(volumeName,
> metrics.getMetadataOperationMean());
> readIoStats.put(volumeName, metrics.getReadIoMean());
> writeIoStats.put(volumeName, metrics.getWriteIoMean());
> }{code}
--
This message was sent by Atlassian JIRA
(v7.6.3#76005)
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
