Creating NFS storage domain fails with sanlock errors and NFS timeouts.
Issue
-
Can't attach NFS storage domain to datacenter.
-
The following error was reported in the Admin Portal;
Failed to attach Storage Domain domain-X to Data Center dc-X.
- VDSM reported various "sanlock" errors;
ReleaseHostIdFailure: Cannot release host id: ('2cc79db8-de44-45e8-85d8-f53a2ac9b288', SanlockException(115, 'Sanlock lockspace remove failure', 'Operation now in progress'))
..........
AcquireLockFailure: Cannot obtain lock: "id=2cc79db8-de44-45e8-85d8-f53a2ac9b288, rc=5, out=Cannot acquire cluster lock, err=(5, 'Sanlock resource not acquired', 'Input/output error')"
-
The NFS filesystem for the storage domain was mounted on the host successfully.
-
When it was mounted by VDSM, simple shell commands like
ls
would "hang" and either eventually complete after several minutes or timeout. However, these commands would also respond normally intermittently. -
When it was mounted manually on the host, all commands appeared to respond normally.
-
-
The
messages
file contained;
Dec 19 17:56:20 host-A kernel: nfs: server xxx.yyy.com not responding, timed out
Environment
- Red Hat Enterprise Virtualization (RHEV) 3.4
- Red Hat Enterprise Virtualization Hypervisor (RHEV-H) 6.5 version 20141017.0
- vdsm-4.14.17-1
Subscriber exclusive content
A Red Hat subscription provides unlimited access to our knowledgebase, tools, and much more.