[Csync2] design question

[Giampaolo Tomassoni]

From: csync2-bounces at lists.linbit.com
> 
> I have been trying to set up csync2 for the last two weeks and now a 
> question comes to my mind: are the comparation/sync of the files on 
> different hosts only name-based ?

I guess so.


> What would happen if i rename a directory on one host ? Would all its 
> contents be first deleted on the remote hosts then copied again under 
> the new directory name ?

That's what happens to me.


> I this is the case wouldn't it be more efficient to use inode number 
> instead of name to mark files as dirty ?

I'm not in charge of anything in the csync2 project, so I can't really say why the author(s) didn't choose the i-node way to implement it.

I can only guess that the reason is due to the "volatile" nature of i-node numbers. Csync2 is not meant to replicate a filesystem: it is meant to replicate a VFS space. This way, a user mounting, say, a NFS fs somewhere in its VFS space and willing to sinchronize (also) the content from NFS, would put csync2 in trouble since NFS i-node numbers are not meant to be consistent across mounts. Other (often network-based) FS flavors follow this pattern.

Csync2 is meant to keep in synch almost-immutable files and hierarchies (say, your /etc dir and maybe the content from some static web sites). If you need to synch rapidly-changing hirarchies, you'd better to move to some other mean (i.e.: the coda fs or some other network-base redundant fs).


> Greeting,

Regards,

Giampaolo

> 
> --
> Eric
> _______________________________________________
> Csync2 mailing list
> Csync2 at lists.linbit.com
> http://lists.linbit.com/mailman/listinfo/csync2