DRBD Command

[root@server1 ~]# drbdadm -- disconnect all
[root@server1 ~]# drbdadm -- connect all
[root@server2 ~]# drbdadm -- disconnect all
[root@server2 ~]# drbdadm -- connect all

or

## node at old/bad drbd data
[root@server2 ~]# drbdadm -- disconnect all
[root@server2 ~]# drbdadm -- --discard-my-data connect all
## node at the latest/good drbd data
[root@server1 ~]# drbdadm connect all

If you found the problem between primary and secondary server:
cs:WFConnection st:Primary/Unknown ds:UpToDate/DUnknown C r---
cs:StandAlone st:Secondary/Unknown ds:UpToDate/DUnknown r---

It can help to fix the "Unknown problem". The nodes will start to resynchronize, transferring data from the good-data node (will appears as SyncSource in /proc/drbd) to the bad-data node(SyncTarget).

[root@primary ~]# drbdsetup primary all

or

[root@primary ~]# drbdsetup /dev/drbd0 primary -o

Make the primary drbd to current server

[root@secondary ~]# drbdsetup secondary all

or

[root@secondary ~]# drbdsetup /dev/drbd0 secondary

Make the secondary drbd to current server

[root@primary ~]# drbdadm connect all

Connect all drbd resource manually
You can use it if you have secondary server waiting to join.