[jira] [Resolved] (RATIS-2183) Detect staled snapshot request

["Tsz-wo Sze (Jira)" <ji...@apache.org>]

     [ https://issues.apache.org/jira/browse/RATIS-2183?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ]

Tsz-wo Sze resolved RATIS-2183.
-------------------------------
    Fix Version/s: 3.2.0
       Resolution: Fixed

The pull request is now merged.  Thanks, [~tohsakarin__]!

> Detect staled snapshot request
> ------------------------------
>
>                 Key: RATIS-2183
>                 URL: https://issues.apache.org/jira/browse/RATIS-2183
>             Project: Ratis
>          Issue Type: Improvement
>          Components: snapshot
>            Reporter: yuuka
>            Assignee: yuuka
>            Priority: Major
>             Fix For: 3.2.0
>
>         Attachments: image-2024-10-30-21-29-17-511.png, image-2024-10-30-21-29-45-678.png, image-2024-10-30-21-30-11-453.png, image-2024-10-30-21-30-44-164.png, image-2024-10-30-21-31-25-035.png
>
>          Time Spent: 1h 10m
>  Remaining Estimate: 0h
>
> If the recipient's network fluctuates and there is a delay in receiving snapshotRequest and cannot reply to the sender in time, the sender will retry. This may cause confusion in receiving new and old snapshotRequest at the recipient
>  
> sender: 
> time1
> !image-2024-10-30-21-29-17-511.png!
> time2
> !image-2024-10-30-21-29-45-678.png!
>  
> time1 < time2
>  
> receiver: 
> !image-2024-10-30-21-30-11-453.png!
> !image-2024-10-30-21-31-25-035.png!
> Concurrently process requests for time1 and time2
>  
>  
> -I suggest using the timestamp field to filter out old requests- (we may use callId to detect old requests.)



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