Satellite does not support the following layers: [DRBD]

[tiptop-labs]

I get the following error message when following approx. the example in the User's Guide and auto-placing resources/volumes:

ERROR: Description: Satellite 'node-2' does not support the following layers: [DRBD]

The kernel module for DRBD 9 is loaded on node-2. What is the reason for this error message and how can I rectify the root cause?

[rp-]

you probably don't have drbd correctly setup on that node. check the satellite log and if drbd-utils and drbd9 kernel module is loaded

[anencore94]

This error seems to be occured when the user create node with wrong hostname on 'node create {hostname} {IP}' command. I had the same error, but resolved it by checking the hostname.

[IsaackRasmussen]

I've run into the same problem and I dont see any issues with my hostnames.

How do I verify kernel modules? I presume for DRBD 9, I can run this: lsmod | grep -i drbd Which gives me:

drbd_transport_tcp     24576  0
drbd                  528384  1 drbd_transport_tcp
libcrc32c              16384  6 nf_conntrack,nf_nat,dm_persistent_data,drbd,raid456,ip_vs

But what about drbd-utils ?

And my Controller + Satellite logs are empty. I only have the individual error reports which gives me the same error message.

[ghernadi]

drbdadm --version or cat /proc/drbd

[IsaackRasmussen]

Thanks - the output also looks good to me

**drbdadm --version**
DRBDADM_BUILDTAG=GIT-hash:\ 147860216bf307433e2d0d7cfd6fad3642027b8d\ build\ by\ buildd@lgw01-amd64-060\,\ 2019-06-13\ 14:21:26
DRBDADM_API_VERSION=2
DRBD_KERNEL_VERSION_CODE=0x090013
DRBD_KERNEL_VERSION=9.0.19
DRBDADM_VERSION_CODE=0x090a00
DRBDADM_VERSION=9.10.0

**cat /proc/drbd**
version: 9.0.19-1 (api:2/proto:86-115)
GIT-hash: 6f5fa5d348a99e5eeb09d83c49853d72e614fd07 build by root@server013, 2019-08-26 13:26:53
Transports (api:16): tcp (9.0.19-1)

[ghernadi]

And the uname -n run on the satellite matches exactly the node name of the satellite?

[IsaackRasmussen]

Thanks for double-checking even though I claimed it was correct. I had not understood that the computer hostname had to match the linstor configured node name. Because it appeared to be communicating just fine.

Its working now - thanks again-

[raltnoeder]

To be more precise, the local node name must match the name of the node as it is registered in LINSTOR, because that name is used by LINSTOR to write DRBD configuration files, and DRBD requires the names in those configuration files to match that node's local node name.