[HDFS-788] Datanode behaves badly when one disk is very low on space

• Type: Bug
• Status: Resolved
• Priority: Major
• Resolution: Fixed
• Component/s: datanode
• Assignee: Unassigned
• Reporter: Todd Lipcon
• Created: 28/Nov/09 04:56
• Updated: 30/Jul/14 00:00
• Resolved: 29/Jul/14 20:53

Description

FSDataset.getNextVolume() uses FSVolume.getAvailable() to determine whether to allocate a block on a volume. This doesn't factor in other in-flight blocks that have been "promised" space on the volume. The resulting issue is that, if a volume is nearly full but not full, multiple blocks will be allocated on that volume, and then they will all hit "Out of space" errors during the write.