> On Sun, 2006-06-11 at 20:59 +0100, centos at bathnetworks.com wrote: >> Hiya, >> >> I'm new to Centos but learning rapidly but I have been using FreeBSD. >> I'm >> trying to setup a HA NFS server using 2 machines. Both machines are >> running 4.3 updated to the latest via Yum. >> >> I did yum groupinstall drbd-heartbeat and >> yum install kernel-module-drbd-kernel-module-drbd-2.6.9-34.EL to match >> my >> kernel. >> >> The problem I have is that on 1 machine drbd works fine, but when I >> start >> drbd on the other I get an error: >> >> Starting DRBD resources: [ d0 ioctl(,SET_DISK_CONFIG,) failed: >> Invalid >> argument >> Lower device is already mounted. >> >> cmd /sbin/drbdsetup /dev/drbd0 disk /dev/hde1 internal -1 failed! >> >> It is not mounted as far as I can tell - df -a only shows /dev/hda and >> fstab has no entry for /dev/hde1. >> >> I have also tried deleting the partion with parted and recreating it. I >> have also tried using just /dev/hde. >> >> The /etc/drbd.conf is: >> >> resource drbd0 { >> protocol C; >> >> on fs1.bathnetworks.local { >> device /dev/drbd0; >> disk /dev/hde1; >> address 10.0.0.1:7788; >> meta-disk internal; >> } >> >> on shorty.bathnetworks.local { >> device /dev/drbd0; >> disk /dev/hdc1; >> address 10.0.0.2:7788; >> meta-disk internal; >> } >> } >> >> fs1 is the machine I am having trouble with. Any help with fixing this >> would be appreciated. >> >> Rob >> >> > > I think you are not understanding what drbd does. > > It is network RAID1 ... you can't mount and use the DRBD device on both > machines at the same time. It is designed so that if the first machine > fails, the second machine / partition is used instead. > Jonny, Thanks for the info. I understand that point. However, with DRBD not running or mounted on shorty, Fs1 still gives the same error. Rob