[ 
https://issues.apache.org/jira/browse/HDDS-4342?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
 ]

Aravindan Vijayan resolved HDDS-4342.
-------------------------------------
    Resolution: Fixed

Merged the PR.

> Add DataNode state and transitions for a node going through upgrade
> -------------------------------------------------------------------
>
>                 Key: HDDS-4342
>                 URL: https://issues.apache.org/jira/browse/HDDS-4342
>             Project: Hadoop Distributed Data Store
>          Issue Type: Sub-task
>            Reporter: Prashant Pogde
>            Assignee: Prashant Pogde
>            Priority: Major
>              Labels: pull-request-available
>
> A typical SCM sequence for driving datanodes through upgrade would be
> Something like following :                                                    
>               
>  - client sends Finalize                                                      
>  
>  - SCM moves to Finalizing states. If SCM crashes, and it comes up            
>  
>    it will always restart from this state.                                    
>  
>  - SCM disallows new pipeline creation, SCM in safe mode                      
>  
>     (SCM Freeze for new pipeline)                                             
>  
>  - SCM closes existing pieplines                                              
>  
>  - SCM updates MLV = SLV if not already so. Update on-disk MLV state.         
>  
>  - SCM moves all data nodes to HEALTHY_READONLY state. Please note that       
>  
>    initial state for all data node is HEALTHY_READONLY. For data nodes        
>  
>    to move from HEALTHY_READONLY -> HEALTHY, they need to send atleast        
>  
>    one heartbeat where DN.MLV == SCM.MLV                                      
>  
>  - SCM waits for few heartbeats                                               
>  
>  - SCM allows new pipeline creation (SCM thaw for new pipeline creation).     
>  
>    New Pipelines can be created if enough HEALTHY data nodes are found.       
>  
>  - If SCM comes across any data node heart beat with DN.MLV < SCM.MLV => SCM 
> sends
>    that data node finalize command 
> ======================================================================
>  
> As part of this, we would be introducing a new state HEALTHY-READONLY in 
> DataNode state machine maintained in SCM .
> This Jira will be used to make changes in the datanode state machine.         
>                                            



--
This message was sent by Atlassian Jira
(v8.3.4#803005)

---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]

Reply via email to