2 Replies Latest reply: Apr 26, 2012 1:04 AM by user12215372 RSS

    Second NFS Server - Extended RAC

    user12215372
      Hi everyone, i hope somebody can help me with a question,

      I am implementing a extended rac (11.2..3 in AIX 6.1) in redundancy normal with ASM, but my diskgroup for my votings files and ocr has 2 raw devices + 1 voting file by NFS, so my question is if i can add other NFS server for put a 4 voting file, if my first NFS Server died so i would have my second NFS Server as backup. Also if i put a 4 voting file in the diskgroup in redundancy normal the 4 files will be the same or only 3 votings files will be equal and the other will be offline?

      Thank you if somebody can help me.

      Regards.
        • 1. Re: Second NFS Server - Extended RAC
          Levi Pereira
          user12215372 wrote:
          Hi everyone, i hope somebody can help me with a question,

          I am implementing a extended rac (11.2..3 in AIX 6.1) in redundancy normal with ASM, but my diskgroup for my votings files and ocr has 2 raw devices + 1 voting file by NFS, so my question is if i can add other NFS server for put a 4 voting file, if my first NFS Server died so i would have my second NFS Server as backup. Also if i put a 4 voting file in the diskgroup in redundancy normal the 4 files will be the same or only 3 votings files will be equal and the other will be offline?
          Hi,
          Far as voting files are concerned, a node must be able to access more than the half of the voting files at any time (simple majority). In order to be able to tolerate a failure of n voting files, one must have at least 2n+1 configured. (n= number of voting files) for the cluster.
          The problem is if you are using two storage systems, which means that the site that hosts the majority of the voting files is a potential single point of failure for the entire cluster. If the storage or the site where n+1 voting files are configured fails, the whole cluster will go down, because Oracle Clusterware will loose the majority of voting files
          To prevent a full cluster outage, Oracle will support a third voting file on an inexpensive, lowend standard NFS mounted device somewhere in the network.

          For this reason Oracle introduced in 11.2 the option QUORUM DISK. This option should be used only when we have even number of Storage HW.
          The QUORUM DISK is a DISK on NFS or Third Location wich will not store DATA (i.e only VOTEDISK), because we cannot provide STORAGE HW only for this. And using these alternative storage option can cause perfomance problem if stored data.
          I believe that only one QUORUM voting disk is enough, because if your NFS die your cluster will be online since you don't lose another votedisk.
          You can have a another host configured with NFS running in Standby Mode, if primary NFS die you can online remove (from clusterware) the votedisk from failed NFS and configure third votedisk on Standby NFS.


          About use 4th Voting Disk is wrong.
          You should configure odd number of voting disks. (To add 4th on NFS you should choose 5 storage location "i.e 5 voting file" to be able configure a odd number)

          If you lose 1/2 or more of all of your voting disks, then nodes get evicted from the cluster, or nodes kick themselves out of the cluster.
          For this reason when using Oracle for the redundancy of your voting disks, Oracle recommends that customers use 3 or more voting disks .
          Note: It is recommended to use an odd number as 4 disks will not be any more highly available than 3 disks, 1/2 of 3 is 1.5...rounded to 2, 1/2 of 4 is 2, once you lose 2 disks, your cluster will fail with 4 voting disks.

          I recommend you read it:
          http://levipereira.wordpress.com/2012/01/11/explaining-how-to-store-ocr-voting-disks-and-asm-spfile-on-asm-diskgroup-rac-or-rac-extended/

          Regards,
          Levi Pereira
          • 2. Re: Second NFS Server - Extended RAC
            user12215372
            Thank you very much Levy, very good the explication, thank you really.

            Regards.