[DRBD-user] full freeze on sync

Flavio Caetano cae2-drbd at col.psi.br
Fri Aug 12 23:56:49 CEST 2005

Note: "permalinks" may not be as permanent as we would like,
direct links of old sources may well be a few messages off.


Hi

I have two machines running debian 3.1 + kernel 2.6.12.4

I have setup DRBD on three resources. franciele-radius, franciele-named and 
franciele-mysql. When I start DRBD on both machines and tell what is the 
primary the sync begins.

The two first resources sync'ed ok, but another resource freezes my SyncTarget 
machine when it reaches 47% of sync'ed it full freeze, no logs, no keyboard. 
Defining "DBG_ALL_SYMBOLS" in drbd_config.h doesn't help, enabling 
nmi_watchdog on my processor doesn't generate an oops, magic sysrq doesn't 
work, memtest shows that my memory is ok, the machine just DIED.

After a hard reboot if I clean my meta-disk disk the sync begin again and 
machine freezes on 47%, if I dont clean meta-disk the machine freezes at the 
same time that I type: /etc/init.d/drbd start.

SyncSource machine have a SMP processor and two SATA disks
SyncTarget machine have an UP processor and two IDE disks (with no badblocks 
of course)
Both machines have DRBD 0.7.11


I'll paste the drbd.conf below

====================
resource franciele-radius {
  protocol B;
  incon-degr-cmd "echo '!DRBD! pri on incon-degr' | wall ; sleep 60 ; halt 
-f";

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

  disk {
    on-io-error   panic;
  }

  net {
    timeout       60;    #  6 seconds  (unit = 0.1 seconds)
    connect-int   10;    # 10 seconds  (unit = 1 second)
    ping-int      10;    # 10 seconds  (unit = 1 second)
    on-disconnect reconnect;

  }

  syncer {
    rate 150M;
  }

  on silvana {
    device     /dev/drbd0;
    disk       /dev/sdb5;
    address    10.0.0.7:7788;
    meta-disk  /dev/sda2[0];
  }

  on franciele {
    device    /dev/drbd0;
    disk      /dev/hdc5;
    address   10.0.0.40:7788;
    meta-disk /dev/hda2[0];
  }
}


resource franciele-named {
  protocol B;
  incon-degr-cmd "echo '!DRBD! pri on incon-degr' | wall ; sleep 60 ; halt 
-f";

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

  disk {
    on-io-error   panic;
  }

  net {
    timeout       60;    #  6 seconds  (unit = 0.1 seconds)
    connect-int   10;    # 10 seconds  (unit = 1 second)
    ping-int      10;    # 10 seconds  (unit = 1 second)
    on-disconnect reconnect;

  }

  syncer {
    rate 150M;
  }

  on silvana {
    device     /dev/drbd1;
    disk       /dev/sdb6;
    address    10.0.0.7:7789;
    meta-disk  /dev/sda2[1];
  }

  on franciele {
    device    /dev/drbd1;
    disk      /dev/hdc6;
    address   10.0.0.40:7789;
    meta-disk /dev/hda2[1];
  }
}


resource franciele-mysql {
  protocol B;
  incon-degr-cmd "echo '!DRBD! pri on incon-degr' | wall ; sleep 60 ; halt 
-f";

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

  disk {
    on-io-error   panic;
  }

  net {
    timeout       60;    #  6 seconds  (unit = 0.1 seconds)
    connect-int   10;    # 10 seconds  (unit = 1 second)
    ping-int      10;    # 10 seconds  (unit = 1 second)
    on-disconnect reconnect;

  }

  syncer {
    rate 10M;
  }

  on silvana {
    device     /dev/drbd2;
    disk       /dev/sdb7;
    address    10.0.0.7:7790;
    meta-disk  /dev/sda2[2];
  }

  on franciele {
    device    /dev/drbd2;
    disk      /dev/hdc7;
    address   10.0.0.40:7790;
    meta-disk /dev/hda2[2];
  }
}
====================

Thanks

Flávio Caetano



More information about the drbd-user mailing list