To finish up this monologue - "all_squash" needed to be set on the NFS export. That solved this issue. On Fri, Jan 2, 2015 at 9:25 PM, Johan Kooijman <mail@johankooijman.com> wrote:
And also, when creating the storage domain, some stuff is written:
root@head1-1:/cloud/ovirt-data# ls -la total 10 drwxrwxrwx 3 36 36 4 Jan 2 21:24 . drwxr-xr-x 5 root root 5 Jan 2 19:24 .. drwxr-xr-x 4 36 36 4 Jan 2 21:24 24b198c4-41c1-4311-8e11-c4cd7a70db5f -rwxr-xr-x 1 36 36 0 Jan 2 21:24 __DIRECT_IO_TEST__
It just fails with AcquireHostIdFailure.
On Fri, Jan 2, 2015 at 8:49 PM, Johan Kooijman <mail@johankooijman.com> wrote:
Forgot to mention: engine & nodes are all CentOS 6.6
On Fri, Jan 2, 2015 at 8:44 PM, Johan Kooijman <mail@johankooijman.com> wrote:
Hi,
I have a datacenter, 2 nodes and an NFS storage domain. The issue is when I try activate a storage domain, logs are here: https://plakbord.cloud.nl/p/P1itMAEoTVIUHQv1U0AAZ5Xy.
In the past I've seen issues with sanlock & selinux, in this case I have no selinux running. When look at the 2 nodes, they actually have the storage domains mounted:
[root@hv2 mnt]# mount | grep nfs sunrpc on /var/lib/nfs/rpc_pipefs type rpc_pipefs (rw) 10.0.24.30:/cloud/ovirt-data on /rhev/data-center/mnt/10.0.24.30:_cloud_ovirt-data type nfs (rw,soft,nosharecache,timeo=600,retrans=6,nfsvers=3,addr=10.0.24.30) 10.0.24.30:/cloud/ovirt-iso on /rhev/data-center/mnt/10.0.24.30:_cloud_ovirt-iso type nfs (rw,soft,nosharecache,timeo=600,retrans=6,nfsvers=3,addr=10.0.24.30) 10.0.24.30:/cloud/ovirt-export on /rhev/data-center/mnt/10.0.24.30:_cloud_ovirt-export type nfs (rw,soft,nosharecache,timeo=600,retrans=6,nfsvers=3,addr=10.0.24.30)
Am I missing something here?
-- Met vriendelijke groeten / With kind regards, Johan Kooijman
-- Met vriendelijke groeten / With kind regards, Johan Kooijman
-- Met vriendelijke groeten / With kind regards, Johan Kooijman
-- Met vriendelijke groeten / With kind regards, Johan Kooijman