[HDFS-6110] adding more slow action log in critical write path

• Type: Improvement
• Status: Closed
• Priority: Major
• Resolution: Fixed
• Affects Version/s: 2.3.0, 3.0.0-alpha1
• Fix Version/s: 2.5.0
• Component/s: datanode
• Hadoop Flags: Reviewed

Release Note

Log slow i/o. Set log thresholds in dfsclient and datanode via the below new configs:

dfs.client.slow.io.warning.threshold.ms (Default 30 seconds)
dfs.datanode.slow.io.warning.threshold.ms (Default 300ms)

• Assignee: Liang Xie
• Reporter: Liang Xie
• Created: 17/Mar/14 10:25
• Updated: 25/Apr/18 04:18
• Resolved: 26/May/14 19:47

Description

After digging a HBase write spike issue caused by slow buffer io in our cluster, just realize we'd better to add more abnormal latency warning log in write flow, such that if other guys hit HLog sync spike, we could know more detail info from HDFS side at the same time.
Patch will be uploaded soon.

Issue Links

• is related to: HDFS-8439 Adding more slow action log in critical read path
• relates to: HDFS-3751 DN should log warnings for lengthy disk IOs
• relates to: HDFS-3590 Print a WARN if the edit log sync period takes more than X time units