Re: failover

[Thing <thing.thing@xxxxxxxxx>]

Hi,

Ah yes, I was just watching youtube videos yesterday and that came up, if only I had known earlier.  :(  Unfortunately I only have 2 hardware nodes.

Does the arbiter node have to be high spec?  ie I have a raspberrypi as my bastion host / system controller, if its a simple "write to node 2 as 1 is off"  sort of thing the Pi might cope. If its like that at all of course, the Pi cant do any bandwidth and i/o to speak of even with an added sata port.

On 10 November 2015 at 12:32, Lindsay Mathieson <lindsay.mathieson@xxxxxxxxx> wrote:

You probably have quorum problems – with 2 nodes its not safe to write to a node when one is down due to the possibility of it being up but uncontactable, so potentially both nodes could be written too, leading to split brain issues.

 

Ideally you should setup three nodes, replica 3 with client quorum enabled. So long as any *two* nodes stay in contact, they can be written to.

 

Alternatively two nodes plus one arbiter node will work as well.

 

Sent from Mail for Windows 10

 

 

From: Thing
Sent: Tuesday, 10 November 2015 9:16 AM
To: Gluster Users
Subject: failover

 

 

Hi,

I am just testing a KVM frontend with VMs and using a backend 2 node gluster setup mount with glusterfs.  this runs fine, but my understanding was that if the gluster node the front end is attached to goes down the client swaps to the other node and carries on?

So I switched gluster node 1 off and the VM just froze, so I am missing something?  or am I wrong and it cant autofailover?

 

 

_______________________________________________
Gluster-users mailing list
Gluster-users@xxxxxxxxxxx
http://www.gluster.org/mailman/listinfo/gluster-users