[DRBD-user] Understanding "classic" 3 node set up.

Igor Cicimov icicimov at gmail.com
Sat Feb 10 01:01:41 CET 2018


On 9 Feb 2018 7:30 pm, "Paul O'Rorke" <paul at tracker-software.com> wrote:

In a functional classic 3 node set up what is the expected state of the
secondary node that is the lower resource with regards the upper resource?

trk-kvm-01 (primary for both resource levels)

root at trk-kvm-01:~/scripts# drbd-overview
[...]
 12:convirt/0                Connected Primary/Secondary
UpToDate/UpToDate _convirt           vda virtio
[...]
120:convirt-U/0         ^^12 Connected Primary/Secondary
UpToDate/UpToDate

trk-kvm-02 (secondary lower resource)

root at trk-kvm-02:~/scripts# drbd-overview
[...]
 12:convirt/0                Connected    Secondary/Primary
UpToDate/UpToDate _convirt           vda virtio
[...]
120:convirt-U/0         ^^12 Unconfigured .
.

trk-kvm-03 (secondary on upper resource)

root at trk-kvm-03:/media/scratch# drbd-overview
[...]
120:convirt-U/0  Connected Secondary/Primary UpToDate/UpToDate

I see nowhere to set this node's IP yet the docs say this resource file
"must be distributed across all nodes in the cluster — in this case, three
nodes."  What is the point of the convirt-U res file on trk-kvm-02?

Well I guess if you loose the first node someone needs to take over its
role, right? Never used stacked resources so it is just a guess.
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.linbit.com/pipermail/drbd-user/attachments/20180210/5be81ab5/attachment.htm>


More information about the drbd-user mailing list