Re: [Gluster-devel] gluster questions

[Jonathan Fine]

Hi Matt,

Thanks for answering my questions.  Now if you'll indulge me, I have  
one more for you(or anyone else that may be able to answer this):   
Regarding the scenario of starting with a single glusterfs node and  
then adding a second node later on:  If I do something like AFR, how  
does the system know which way to sync?  How can I be sure I will end  
with my data mirrored and not deleted?

So far, glusterfs sounds like the way to go..


thanks,


Jon



On Feb 4, 2008, at 5:31 PM, Matt Paine wrote:

> Hi Jon
>
>
> > I'm thinking about implementing gluster as a possible way of home   
> > directory server redundancy(using the afr translator), but I have  
> > a  few questions that I can't seem to find the answer to:
> > -Does installing gluster require a total format of the "bricks" or   
> > data volumes involved?  Can I just install it over the current  
> > system  and keep data intact? From what I read it just works over  
> > the current  filesystem (ext3, xfs, etc)?
>
> No. A gluster "brick" is simply a directory that is exported as a  
> filesystem (very similar to NFS). The underlying filesystem can be  
> whatever you want (as long as its readable from the kernel).
>
>
> > -We currently have a single node w/fibre channel raid (partitioned   
> > into 7 luns for different user groups) for our home directories.   
> > All  exported using NFS.  We are getting in a new server and raid  
> > unit,  which will become the primary server once it is setup.  Can  
> > I setup  the new server and raid as a single gluster node, get the  
> > data copied  over (via rsync or something similar), move clients  
> > over to the new  machine, then resetup the old server/raid combo  
> > and add it into the  gluster system?  Or is there a better way to  
> > do this?
>
> Sounds like a perfectly fine way to do it.
>
>
> > -Concerning NFS and tied into the above question: we use it for  
> > it's  compatibility with Mac, Linux and Solaris workstations and  
> > has worked  fairly well.  We'd like to stick with it.  Would that  
> > require running  a fuse/gluster client on one of the above nodes  
> > and re-exporting it as  NFS?  That is the way I seem to understand  
> > from my searches on the  mailing lists.    Or can you directly  
> > export a gluster brick via nfs?   And if running the fuse/gluster  
> > client on one of the server nodes w/  exported NFS is necessary, is  
> > this a safe way to do things?
>
> There were problems a while ago about exporting glusterfs volumes as  
> NFS exports, but I believe these problems have now been sorted out  
> (if you use the gluster patched fuse client). I am unable to tell  
> you if Mac and Solaris works with glusterfs exports, but all I can  
> sugest there is to download the source and give it a crack :) Or  
> someone else might be able to jump in with a better answer.
>
>
> Good luck :)
>
>
>
> Matt.
>
>
>
>
> _______________________________________________
> Gluster-devel mailing list
> address@hidden
> http://lists.nongnu.org/mailman/listinfo/gluster-devel

************************************************
Jonathan Fine
address@hidden
814-863-4465

Sys Admin/IT Manager
Astronomy and Astrophysics
Eberly College of Science
Pennsylvania State University