[DRBD-user] Optimizing for file copys over NFS?

[Mike Sweetser - Adhost]

Hello,

We have a DRBD installation with the following configuration that we're
experiencing serious performance issues with on the server when copying
a large amount of files, or a large file:

resource r1 {
  protocol C;
  handlers {
    pri-on-incon-degr "echo 'DRBD: primary requested but inconsistent!'
| wall; /etc/init.d/heartbeat stop"; #"halt -f";
    pri-lost-after-sb "echo 'DRBD: primary requested but lost!' | wall;
/etc/init.d/heartbeat stop"; #"halt -f";
  }

  startup {
    degr-wfc-timeout 120;    # 2 minutes.
    wfc-timeout 120;    # 2 minutes.
  }

  disk {
    on-io-error   detach;
  }

  net {
    timeout 120;
    connect-int 20;
    ping-int 20;
    max-buffers     2048;
    max-epoch-size  2048;
    ko-count 30;
    cram-hmac-alg "sha1";
    shared-secret "adhostdrbd";
    after-sb-0pri discard-least-changes;
    after-sb-1pri call-pri-lost-after-sb;
    after-sb-2pri disconnect;
  } 
    
  syncer {
    rate 3M;
    al-extents 257;
  }

  on NODE1 {
    device    /dev/drbd1;
    disk      /dev/sda7;
    address   172.16.2.1:7789;
    meta-disk /dev/sda6[1];
  } 
 
  on NODE2 {
    device    /dev/drbd1;
    disk      /dev/sda7;
    address   172.16.2.2:7789; 
    meta-disk /dev/sda6[1]; 
  }
}

What can we do to optimize this?  It's used as an NFS drive for several
web servers, so the faster it can go for this, the better.  Any help on
optimizing DRBD would be appreciated.  Also, are there any helpful hints
for using DRBD with NFS, as far as what parameters to use for the NFS
mounts to speed things up?

Thank You,
Mike Sweetser