[ 
https://issues.apache.org/jira/browse/RATIS-2166?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
 ]

yuuka updated RATIS-2166:
-------------------------
    Description: 
In [RATIS-2166] Regarding the follow-up patch issue of RATIS-2148 - ASF JIRA 
(apache.org) the issue of receiving the last snapshot file without considering 
the possibility of intermediate blocks failing was fixed, which resulted in the 
incorrect triggering of loadStateMachine.
 
Some issues were missed in #2148:
 
For example, Leader L will send the following data to Follower F in the form of 
a snapshot: data [a, b, c,..., z].
L will split the data into 0: [a, b, c, d], 1: [e, f, g, h],...
F already has data [a, b, c, d, e]
F will reply with ALREADY_INSTALL after receiving data-0 and will not update 
nextChunkIndex.
 
This will result in failure to receive subsequent data_chunk correctly.

  was:
In Create Issue - ASF JIRA (apache.org) the issue of receiving the last 
snapshot file without considering the possibility of intermediate blocks 
failing was fixed, which resulted in the incorrect triggering of 
loadStateMachine.
 
Some issues were missed in #2148:
 
For example, Leader L will send the following data to Follower F in the form of 
a snapshot: data [a, b, c,..., z].
L will split the data into 0: [a, b, c, d], 1: [e, f, g, h],...
F already has data [a, b, c, d, e]
F will reply with ALREADY_INSTALL after receiving data-0 and will not update 
nextChunkIndex.
 
This will result in failure to receive subsequent data_chunk correctly.


> Regarding the follow-up patch issue of RATIS-2148
> -------------------------------------------------
>
>                 Key: RATIS-2166
>                 URL: https://issues.apache.org/jira/browse/RATIS-2166
>             Project: Ratis
>          Issue Type: Bug
>            Reporter: yuuka
>            Priority: Major
>
> In [RATIS-2166] Regarding the follow-up patch issue of RATIS-2148 - ASF JIRA 
> (apache.org) the issue of receiving the last snapshot file without 
> considering the possibility of intermediate blocks failing was fixed, which 
> resulted in the incorrect triggering of loadStateMachine.
>  
> Some issues were missed in #2148:
>  
> For example, Leader L will send the following data to Follower F in the form 
> of a snapshot: data [a, b, c,..., z].
> L will split the data into 0: [a, b, c, d], 1: [e, f, g, h],...
> F already has data [a, b, c, d, e]
> F will reply with ALREADY_INSTALL after receiving data-0 and will not update 
> nextChunkIndex.
>  
> This will result in failure to receive subsequent data_chunk correctly.



--
This message was sent by Atlassian Jira
(v8.20.10#820010)

Reply via email to