[DRBD-user] drbd heartbeat help

Alex E alex at pc171.com
Fri Jan 23 17:39:29 CET 2004

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


Hello .I am trying to implement drbd - heartbeat failover solution for our
webhosting company.
It is a 2 node cluster  master and hot slave.I did use heartbeat prior to
using drbd with simple ip failover.
I configured drbd and heartbeat in a following way:
drbd.conf:
resource drbd0 {
  protocol=C
  fsckcmd=fsck.ext3 -p -y
  load-only

disk {
     disk-size  = 979933
     }



net {
    sync-min    = 500k
    sync-max    = 2M    # maximal average syncer bandwidth
    tl-size     = 5000  # transfer log size, ensures strict write ordering
    timeout     = 60    # 0.1 seconds
    connect-int = 10    # seconds
    ping-int    = 10    # seconds
  }

  on node0 {
    device=/dev/nb0
    disk=/dev/hdc7
    address=192.168.1.3
    port=7789
  }

  on node1 {
    device=/dev/nb0
    disk=/dev/hda7
    address=192.168.1.4
    port=7789
  }
}
resource drbd1 {
  protocol=C
  fsckcmd=fsck.ext3 -p -y
  load-only

disk {
     disk-size  = 85401540
     }



net {
    sync-min    = 500k
    sync-max    = 2M    # maximal average syncer bandwidth
    tl-size     = 5000  # transfer log size, ensures strict write ordering
    timeout     = 60    # 0.1 seconds
    connect-int = 10    # seconds
    ping-int    = 10    # seconds
  }

  on node0 {
    device=/dev/nb1
    disk=/dev/md4
    address=192.168.1.3
    port=7790
  }

  on node1 {
    device=/dev/nb1
    disk=/dev/hda2
    address=192.168.1.4
    port=7790
  }
}
and my haresources:
node0.mydomain.com 192.168.1.2 datadisk::drbd0 datadisk::drbd1 psa xinetd
mailman
and i did proper adjustments in fstab

problem is than when i shut down the the main server it didnt remount the
devices on the slave
instead it left them in wait for connection mode but as soon as i started
the primary back it started resync in wrong direction.
and done my master as secondary.but still,didnt remount them.. and i had to
manually command them to primary on the master.
after i rebooted the slave it went to its normal syncall and corrrect
primary/secondary (master is primary,slave secondary).
Can anybody tell me what i do wrong? I have read about init timeout but i
cant figure out how can it help me because the datadisk suppose to cammnd
who is master and who is slave.
Any suggestions please!!!!
Thank you
Alex Einbinder
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.linbit.com/pipermail/drbd-user/attachments/20040124/bcf09394/attachment.htm>


More information about the drbd-user mailing list