[DRBD-user] Performance Problems with DRBD

[Drognat]

Hi,

 

i'm currently testing DRBD and i'm have read / write problems.

 

On the Local Raid 5 array with BBU i can read / write with around 800 MB/s.

If i use /dev/drbd100 it drops to 200 MB/s

 

resource drbd.test {
  protocol C;

  volume 0 {
    device      /dev/drbd100;
    disk        /dev/vgKVM.OS/drbdtest;
    meta-disk   internal;
  }

  net {
    shared-secret "drbd.drbdtest";
  }

  on KVMe001 {
    address     172.16.0.1:7100;
  }

  on KVMe002 {
    address     172.16.0.2:7100;
  }
}
 

# DRBD is the result of over a decade of development by LINBIT.
# In case you need professional services for DRBD or have
# feature requests visit http://www.linbit.com

global {
        usage-count no;
        # Decide what kind of udev symlinks you want for "implicit" volumes
        # (those without explicit volume <vnr> {} block, implied vnr=0):
        # /dev/drbd/by-resource/<resource>/<vnr>   (explicit volumes)
        # /dev/drbd/by-resource/<resource>         (default for implict)
        udev-always-use-vnr; # treat implicit the same as explicit volumes
        # minor-count dialog-refresh disable-ip-verification
        # cmd-timeout-short 5; cmd-timeout-medium 121; cmd-timeout-long 600;
}

common {
        handlers {
                #fence-peer "/usr/lib/drbd/crm-fence-peer.sh";
                #after-resync-target "/usr/lib/drbd/crm-unfence-peer.sh";
        }

        startup {
                # wfc-timeout degr-wfc-timeout outdated-wfc-timeout wait-after-sb
                wfc-timeout 30;
                degr-wfc-timeout 120;
                outdated-wfc-timeout 30;
        }

        syncer {
                rate 300M;
                verify-alg md5;
        }

        options {
                cpu-mask 2;
        }

        disk {
                on-io-error detach;
                disk-flushes no;
                disk-barrier no;
                c-plan-ahead 0;
                c-fill-target 24M;
                c-min-rate 80M;
                c-max-rate 720M;
                al-extents 6007;
                fencing resource-only;
        }

        net {
                # protocol timeout max-epoch-size max-buffers
                # connect-int ping-int sndbuf-size rcvbuf-size ko-count
                # allow-two-primaries cram-hmac-alg shared-secret after-sb-0pri
                # after-sb-1pri after-sb-2pri always-asbp rr-conflict
                # ping-timeout data-integrity-alg tcp-cork on-congestion
                # congestion-fill congestion-extents csums-alg verify-alg
                # use-rle
                allow-two-primaries no;
                after-sb-0pri discard-zero-changes;
                after-sb-1pri discard-secondary;
                after-sb-2pri disconnect;
                rr-conflict disconnect;
                max-buffers 8000;
                max-epoch-size 8000;
                sndbuf-size 2M;
                rcvbuf-size 2M;
                unplug-watermark 16;
        }
}
 

Ressource isn't connectet because need the other Server for another testings.

But finally both would are connected via 10GBe

 

Here some outputs from dd:

 

Bare metal

# dd if=/dev/vgKVM.OS/test of=/dev/null bs=1M count=10240
10240+0 Datensätze ein
10240+0 Datensätze aus
10737418240 bytes (11 GB, 10 GiB) copied, 13,6046 s, 789 MB/s
 

and with DRBD Device:

# dd if=/dev/drbd100 of=/dev/null bs=1M count=10240
10240+0 Datensätze ein
10240+0 Datensätze aus
10737418240 bytes (11 GB, 10 GiB) copied, 47,678 s, 225 MB/s
 

Thanks Alex

_______________________________________________
Star us on GITHUB: https://github.com/LINBIT
drbd-user mailing list
drbd-user@lists.linbit.com
http://lists.linbit.com/mailman/listinfo/drbd-user