[Storagesync] 回复: Re: Storagesync Digest, Vol 5, Issue 1

["Fei Song" <[e-mail]>]

--------------
Fei Song
>Sunday, Dec 6, 2015 10:17 PM Fei Song wrote:
>> Can we just treat the requests sent by peer B and peer C based on their timestamps (arriving at the server).
>> If the file must be kept as the same version, different modification reports can be generated independently with FCFS mode.
>> If this can not be finished automatically, a human intervention will be needed.
>
>No, this doesn't work even if you don't care about wiping out changes.   Think about this sequence:
>
>0: File F is at version 0
>1: Client A syncs to server
>2: Client B syncs to server
>3: File f changed on Client A -> Version 0.A
>4: File f changed on Client B -> Version 0.B
>5: Client B syncs to server, updating file F to version 0.B, modified time = 4
>6: Client A syncs to server: what happens?
>   a: File updated to version 0.A, modified time=3
>   b: Server overwrites version 0.A with version 0.B, changes in 0.A lost
>   c: Conflict resolution combines versions 0.A and 0.B, producing version 1, mod time=6
>
>I think C is the only right answer.   We want conflict resolution to be automatic whenever possible, but we do want to do conflict resolution, and not just wipe out changes, because those changes may contain real work that the user on client A did.

These "middle" changes are important. Yeah. they should be stroed at the server side.
Wiping out changes is just a simple appoach.
I also think c is a good option. Then we need to consider a suitable time window.
e.g. how long should be consider as a "simultaneous" updating.
otherwise F might be updated as Version 1 directly.


>
>
>--
>Sent from Whiteout Mail - https://whiteout.io
>
>My PGP key: https://keys.whiteout.io/[e-mail]