[YARN-1996] Provide alternative policies for UNHEALTHY nodes.

• Type: New Feature
• Status: Open
• Priority: Major
• Resolution: Unresolved
• Affects Version/s: 2.4.0
• Component/s: nodemanager, scheduler
• Assignee: Gera Shegalov
• Reporter: Gera Shegalov
• Created: 29/Apr/14 09:44
• Updated: 04/Jun/18 13:45

Description

Currently, UNHEALTHY nodes can significantly prolong execution of large expensive jobs as demonstrated by MAPREDUCE-5817, and downgrade the cluster health even further due to positive feedback. A container set that might have deemed the node unhealthy in the first place starts spreading across the cluster because the current node is declared unusable and all its containers are killed and rescheduled on different nodes.

To mitigate this, we experiment with a patch that allows containers already running on a node turning UNHEALTHY to complete (drain) whereas no new container can be assigned to it until it turns healthy again.

This mechanism can also be used for graceful decommissioning of NM. To this end, we have to write a health script such that it can deterministically report UNHEALTHY. For example with

if [ -e $1 ] ; then                                                                
  echo ERROR Node decommmissioning via health script hack                          
fi 

In the current version patch, the behavior is controlled by a boolean property yarn.nodemanager.unhealthy.drain.containers. More versatile policies are possible in the future work. Currently, the health state of a node is binary determined based on the disk checker and the health script ERROR outputs. However, we can as well interpret health script output similar to java logging levels (one of which is ERROR) such as WARN, FATAL. Each level can then be treated differently. E.g.,

• FATAL: unusable like today
• ERROR: drain
• WARN: halve the node capacity.
  complimented with some equivalence rules such as 3 WARN messages == ERROR, 2*ERROR == FATAL, etc.

Issue Links

• is duplicated by: YARN-8345 NodeHealthCheckerService to differentiate between reason for UnusableNodes for client to act suitably on it
• relates to: YARN-671 Add an interface on the RM to move NMs into a maintenance state
• relates to: MAPREDUCE-5817 Mappers get rescheduled on node transition even after all reducers are completed
• relates to: YARN-914 (Umbrella) Support graceful decommission of nodemanager