[HDFS-14201] Ability to disallow safemode NN to become active

• Type: Improvement
• Status: Resolved
• Priority: Major
• Resolution: Fixed
• Affects Version/s: 3.1.1, 2.9.2
• Fix Version/s: 3.3.0
• Component/s: auto-failover
• Labels: multi-sbnn
• Target Version/s: 3.1.1, 2.9.2
• Hadoop Flags: Reviewed
• Assignee: Xiao Liang
• Reporter: Xiao Liang
• Created: 11/Jan/19 22:39
• Updated: 02/May/22 20:02
• Resolved: 18/Jun/19 16:58

Description

Currently with HA, Namenode in safemode can be possibly selected as active, for availability of both read and write, Namenodes not in safemode are better choices to become active though.

It can take tens of minutes for a cold started Namenode to get out of safemode, especially when there are large number of files and blocks in HDFS, that means if a Namenode in safemode become active, the cluster will be not fully functioning for quite a while, even if it can while there is some Namenode not in safemode.

The proposal here is to add an option, to allow Namenode to report itself as UNHEALTHY to ZKFC, if it's in safemode, so as to only allow fully functioning Namenode to become active, improving the general availability of the cluster.

Issue Links

• is related to: HDFS-16547 [SBN read] Namenode in safe mode should not be transfered to observer state
• relates to: HDFS-6440 Support more than 2 NameNodes
• links to: GitHub Pull Request #466