Matt Lawrence wrote: > It is an intentionally vague description since I want to keep an open > mind.
I have just been through this process. What inherited from my predecessor cost about £10,000 per year in support costs, was slow, ugly and not fault tolerant at all. For £7,000 I am now providing 3TB of fault-tolerant storage (12TB raw disk capacity), accessible via NFS and Samba, with NFS providing clients with faster I/O than they get at their local disks. It is provided for the purpose of all central storage needs for a fabless semiconductor company including silicon simulations with EDA tools, GridEngine and also general administrative storage. All the details you could need are here: http://blog.tpa.me.uk/high-availability-storage-with-slackware-drbd-pacemaker/ Bear in mind that the language used in the article shows frustration with the Linux HA stack, but I was approaching it cold from an unsupported distribution; from RHEL I think you will find it much easier to handle. Also, I did not use a Primary/Primary disk setup, only one host is serving files at a time from an XFS filesystem, but there's no reason you couldn't modify your setup to include GFS/OCFS2 with a Primary/Primary configuration. -- Zordrak [email protected] _______________________________________________ Tech mailing list [email protected] http://lopsa.org/cgi-bin/mailman/listinfo/tech This list provided by the League of Professional System Administrators http://lopsa.org/
