On Fri, 2003-03-28 at 09:24, Greg Freemyer wrote: > > >> As a side note, there is OpenGFS ( i don't know where it is in > >> alpha/beta release, but it is open source). > > Ignoring SPOF issues, I would say it is beta quality at best. > > To handle the SPOFs: > You must use an external RAID controller. ($$) > You have to setup a HA cluster to run the lockserver. I don't know if transparent failover can be achieved. > Trasparent.. good point.. I have active-active samba, and what I've found is this... Due to state issues, failover WILL cause any 'during failover' file operations to fail. But on the other side of that, the way that samba deals with file moves/copies ensures that no data will be lost.. simply, restart the operation immediately after the error. In our system, the error occurs the moment that the failover machine takes over. other than having to redo any 'in failover' operations, the failover is completely transparent to clients.. if no operations are occuring, then there will be no sign to clients that a failover occured. Matt Schillinger > _______________________________________________ > linux-lvm mailing list > linux-lvm@sistina.com > http://lists.sistina.com/mailman/listinfo/linux-lvm > read the LVM HOW-TO at http://tldp.org/HOWTO/LVM-HOWTO/ _______________________________________________ linux-lvm mailing list linux-lvm@sistina.com http://lists.sistina.com/mailman/listinfo/linux-lvm read the LVM HOW-TO at http://tldp.org/HOWTO/LVM-HOWTO/
