Proxmox HA Cluster with DRBD - DRBD fails to start

[Fatal Error]

Hi there

After i solved my problem with connecting the Servers i tried to make a DRBD Cluster.
This is a preparation for my practical apprenticeship ending exam hold in may.

After i built the Cluster and added the iSCSI Target to the Storages, i tried to make a DRBD.
I followed the HowTo on the DRBD Page in the wiki and did each step exactly like you (leaving aside our network specific eth0 config and the interface on which DRBD should run) and get an error when i try to start the DRBD Daemon. (/etc/init.d/drbd start)

host:~# /etc/init.d/drbd start
Starting DRBD resources:[ ]0: State change failed: (-2) Refusing to be Primary without at least one UpToDate disk
Command '/sbin/drbdsetup 0 primary' terminated with exit code 17
0: State change failed: (-2) Refusing to be Primary without at least one UpToDate disk
Command '/sbin/drbdsetup 0 primary' terminated with exit code 17
0: State change failed: (-2) Refusing to be Primary without at least one UpToDate disk
Command '/sbin/drbdsetup 0 primary' terminated with exit code 17
0: State change failed: (-2) Refusing to be Primary without at least one UpToDate disk
Command '/sbin/drbdsetup 0 primary' terminated with exit code 17
0: State change failed: (-2) Refusing to be Primary without at least one UpToDate disk
Command '/sbin/drbdsetup 0 primary' terminated with exit code 17

thats the output i get on my Clustermaster...

host:~# /etc/init.d/drbd start
Starting DRBD resources:[ ]0: State change failed: (-2) Refusing to be Primary without at least one UpToDate disk
Command '/sbin/drbdsetup 0 primary' terminated with exit code 17
0: State change failed: (-2) Refusing to be Primary without at least one UpToDate disk
Command '/sbin/drbdsetup 0 primary' terminated with exit code 17
0: State change failed: (-2) Refusing to be Primary without at least one UpToDate disk
Command '/sbin/drbdsetup 0 primary' terminated with exit code 17
0: State change failed: (-2) Refusing to be Primary without at least one UpToDate disk
Command '/sbin/drbdsetup 0 primary' terminated with exit code 17
0: State change failed: (-2) Refusing to be Primary without at least one UpToDate disk
Command '/sbin/drbdsetup 0 primary' terminated with exit code 17

thats the output on the Clusternode.

i did each step of the installation according to the Wiki Page...

does anybody knows anything about that error?

Thanks in advance
Fatal Error

 

[Randy]

Try the command:

drbdadm -- --do-what-I-say primary all

It looks like the error I got and if I remember correctly this is how I fixed it.

 

[Fatal Error]

Try the command:

drbdadm -- --do-what-I-say primary all

It looks like the error I got and if I remember correctly this is how I fixed it.

Err i got this error on both nodes...

host:~# drbdadm -- --do-what-I-say primary all
drbdsetup primary: unrecognized option '--do-what-I-say'
Command 'drbdsetup 0 primary --do-what-I-say' terminated with exit code 20
drbdadm primary r0: exited with code 20

 

[Fatal Error]

Hey there

it seems like it is working now, don't know what the solution was, perhaps the command primary all was it, really don't know...

Thanks to Randy for this command...

So long
Fatal Error

 

[Randy]

No I don't think it was the command. That command is for the older version of Drdb (0.71). I think that error you got happens when the time on your servers is out of sync.

 

[Fatal Error]

No I don't think it was the command. That command is for the older version of Drdb (0.71). I think that error you got happens when the time on your servers is out of sync.

that was definitely not the problem. After my previous problem with the Cluster Sync where the node had a time one month ago i learned my stuff and checked this one first^^

 

[tex]

Hi,
I have the same problem, and the command " drbdadm -- --do-what-I-say primary all " have any effect.
How you have resolved?

thanks

 

[Fatal Error]

try
/etc/init.d/drbd status

perhaps drbd is already running and you try to start it again...
if so, proceed with the metadata building and so on.