Mounting gluster volume from fstab at boot - Specifying multiple server nodes

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

 



Why not use an active rr dns ontop of your host naming?

When all hosts are up then  dns resolution would distribute the client
load between 2 nodes - when one host goes down the host dns entry would
be either removed or it's priority would be reduced to 0.


On Mon, 2010-03-15 at 17:05 +0900, Jon Swanson wrote: 
> Hi Tejas,
> Thanks for the response.
> 
> We have a cluster consisting of 6 machines, configured as 2 distributes 
> over 3 nodes, which are then replicated.
> 
> A-|\              /|-D
> B-|-Dist      Dist-|-E
> C-|/ |         |  \|-F
>       |         |
>       |         |
>       .Replicate.
> 
> In theory, we can lose a few different nodes here without any 
> interruption of service. The client, let's call it 'Z', has an fstab 
> entry to mount the gluster volume depicted above from 'A'. As soon as 
> the client connects, it gets the information for the entire volume of 
> course, so if 'A' goes down at any point after that there are no 
> problems, everything is transparent to the client.
> 
> If 'A' happens to be down when 'Z' boots however, it will never make any 
> connection.
> 
> It feels like there should be a way to specify multiple hosts at the 
> initial mount time, as as parsed through fstab, without having to 
> maintain a volume definition on every single client.
> 
> There's a big chance i'm misunderstanding something here, and 
> wholeheartedly welcome any corrections.
> 
> Thanks,
>   jon
> 
> 
> 
> Note: one solution is to just have multiple lines in the fstab. This 
> works but is very hacking and generates notifications in all cases:
> NodeA         /mnt glusterfs 
> rw,allow_other,default_permissions,max_read=131072 0 0
> NodeD         /mnt glusterfs 
> rw,allow_other,default_permissions,max_read=131072 0 0
> 
> 
> Note2: This is a /very/ limited case, and quite easy to diagnose so is 
> in no way a show stopper.
> 
> 
> On 03/15/2010 04:36 PM, Tejas N. Bhise wrote:
> > Jon,
> >
> > Why not just use replication ? and distribute over the replicated setup ? you can stack it up like that.
> >
> > Regards,
> > Tejas.
> > ----- Original Message -----
> > From: "Jon Swanson"<jswanson at valuecommerce.co.jp>
> > To: gluster-users at gluster.org
> > Sent: Monday, March 15, 2010 12:54:55 PM GMT +05:30 Chennai, Kolkata, Mumbai, New Delhi
> > Subject: Mounting gluster volume from fstab at boot - Specifying multiple server nodes
> >
> >
> > Hello. glusterfs is flat out awesome.
> >
> > Just had a quick question about mounting a volume via fstab at boot time
> > that did not seem to be answered
> > here:http://www.gluster.com/community/documentation/index.php/Mounting_a_GlusterFS_Volume
> >
> > I'd like to be able to specify multiple nodes at boot time, so if the
> > volume is spanned across Nodes A,B,C,D, and A goes down, the system will
> > still automatically mount the volume from B. It seems like you could do
> > that relatively easily referencing a volume file on the client.
> >
> > I would like to avoid maintaining separate volume files on each client
> > if at all possible though.
> >
> >
> > Is there a syntax for providing an fstab line for a gluster mount that
> > will allow the gluster client to try multiple hosts in the event one is
> > down?
> >
> >
> > Thanks,
> >   jon
> >
> > _______________________________________________
> > Gluster-users mailing list
> > Gluster-users at gluster.org
> > http://gluster.org/cgi-bin/mailman/listinfo/gluster-users
> >    
> _______________________________________________
> Gluster-users mailing list
> Gluster-users at gluster.org
> http://gluster.org/cgi-bin/mailman/listinfo/gluster-users




[Index of Archives]     [Gluster Development]     [Linux Filesytems Development]     [Linux ARM Kernel]     [Linux ARM]     [Linux Omap]     [Fedora ARM]     [IETF Annouce]     [Bugtraq]     [Linux OMAP]     [Linux MIPS]     [eCos]     [Asterisk Internet PBX]     [Linux API]

  Powered by Linux