Disk Replication with DRBD: Difference between revisions

From Alpine Linux
(Category:Storage)
Line 27: Line 27:
It is recommended to have a dedicated heartbeat interface for DRBD where replication occurs.
It is recommended to have a dedicated heartbeat interface for DRBD where replication occurs.


Create DRBD device and bring it up on '''both nodes''':
Create the DRBD device and bring it up on '''both nodes''':


   #drbdadm create-md r0
   #drbdadm create-md r0
Line 36: Line 36:
   #drbdadm -- --overwrite-data-of-peer primary r0
   #drbdadm -- --overwrite-data-of-peer primary r0


If you want to temporarily speed up replication run (e.g. 1G):
If you want to temporarily speed up replication, run (e.g. 1G):


   #drbdsetup /dev/drbd1 syncer -r 1G
   #drbdsetup /dev/drbd1 syncer -r 1G

Revision as of 04:44, 5 July 2021

This material is work-in-progress ...

Do not follow instructions here until this notice is removed.
(Last edited by Bt129 on 5 Jul 2021.)

Disk Replication with DRBD

DRBD support is present only in SCST version of Alpine. At the time of this writing DRBD tools are currently in edge (testing).

 #apk add drbd

Load DRBD kernel driver:

 #modprobe drbd

Sample DRBD configuration in /etc/drbd.d/r0.res (resource 0):

 resource r0 {
   device /dev/drbd1;
   disk /dev/md0;
   meta-disk internal;
   on host1 {
     address	192.168.0.1:7789;
   }
   on host2 {
     address	192.168.0.2:7789;
   }
 }


It is recommended to have a dedicated heartbeat interface for DRBD where replication occurs.

Create the DRBD device and bring it up on both nodes:

 #drbdadm create-md r0
 #drbdadm up r0

On primary node start replication:

 #drbdadm -- --overwrite-data-of-peer primary r0

If you want to temporarily speed up replication, run (e.g. 1G):

 #drbdsetup /dev/drbd1 syncer -r 1G

Control replication status:

 #cat /proc/drbd