On Mon, Jul 19, 2021 at 09:37:43AM -0400, Jeff Layton wrote:
What version of ganesha and ceph cluster are you using? What OS is all of this running on?
The NFS server is Fedora 34, so that's nfs-ganesha-3.5-7.fc34.x86_64. The Ceph "cluster" (aka single node test environment) was deployed using cephadm from F24 and appears to be 16.2.5.
This looks like a terrible idea...
And more than that, as I mentioned, it didn't have any impact on the failover time, so there are two reasons to remove it :).
With the rados_cluster backend, you can also configure a nodeid to use. If you set up a trivial rados_cluster with one (active/passive) host, I expect it'll work more like you want.
I'll give that a shot and see what happens. -- Lars Kellogg-Stedman <lars@redhat.com> | larsks @ {irc,twitter,github} http://blog.oddbit.com/ | N1LKS