0

I am getting the following error when trying to set the Primary node for DRBD.

'node1' not defined in your config (for this host).

I know this is related to DNS/Hostname/Hosts and the config clusterdb.res. I know this because I originally got an error when trying to start clusterdb.res if node1 didn't resolve correctly. So what confuses me is that I can start the clusterdb.res if either use:

I have used this command on the hosts

hostnamectl set-hostname $(uname -n | sed s/\\..*//)

To make the hostname resolve to node1 instead of node1.localdomain

Or add node1.localdomain to the config, either works. But I have tried all combinations and can't seem to get this command to take :

drbdadm primary --force node1 && cat /proc/drbd

My Configs

/etc/drbd.d/clusterdb.res

resource clusterdb{
    protocol C;
    meta-disk internal;
    device /dev/drbd0;

startup {
    wfc-timeout 30;
    outdated-wfc-timeout 20;
    degr-wfc-timeout 30;
}

net {
    cram-hmac-alg sha1;
    shared-secret sync_disk;
}


syncer {
    rate 10M;
    al-extents 257;
    on-no-data-accessible io-error;
    verify-alg sha1;
}
on node1 {
    disk /dev/sda3;
    address 192.168.1.216:7788;
}
on node2 {
    disk /dev/sda3;
    address 192.168.1.217:7788;
}
}

/etc/hosts :

127.0.0.1   localhost localhost.localdomain localhost4 localhost4.localdomain4
::1         localhost localhost.localdomain localhost6 localhost6.localdomain6
192.168.1.216 node1
192.168.1.217 node2

/etc/hostname

node#

My full write up ATM (wip)

Edits :

 [root@node1 ~]# hostname
 node1
 [root@node1 ~]# cat /etc/hosts
 127.0.0.1   localhost localhost.localdomain localhost4      localhost4.localdomain4
 ::1         localhost localhost.localdomain localhost6      localhost6.localdomain6
 127.0.1.1     node1
 192.168.1.216 node1
 192.168.1.217 node2
 [root@node1 ~]#

Update: I have gotten this to work with LVM following this guide exactly, so I think my issue actually lies with the following lines of code. But for now I think i will stick with LVM since it works, unless somebody else really wants to work on this. (My working LVM writeup)

device /dev/drbd0;

or

 device /dev/drbd0; 

The reason I say this, is I used the same hosts/hostname/shortname/ip_addr but LVM and it worked, but maybe I missed something the first time, I fixed in my new VM Template (I started from scratch to build LVM)

  • Just to be sure - can you please add the output of hostname? Also please add the following line to your /etc/hosts: 127.0.1.1 node1 – geruetzel May 1 '16 at 8:32
1

You're not using the drbdadm command correctly. It wants the resource name, where you're giving it a node name.

Try this instead (from node1):
# drbdadm up clusterdb
# drbdadm primary --force clusterdb

As a side note, DRBD expects the hostnames in its config to be the same as uname - n.

| improve this answer | |
  • As soon as I read your answer I understood. Durp thanks! – FreeSoftwareServers May 2 '16 at 3:15

Not the answer you're looking for? Browse other questions tagged or ask your own question.