Il 02/07/2015 02:36, Christopher Young ha scritto:
I'm sure I have worked through this before, but I've been banging my head
against this one for a while now, and I think I'm too close to the issue.
Basically, my hosted-engine won't start anymore. I do recall attempting to
migrate it to a
Il 02/07/2015 16:04, Christopher Young ha scritto:
[root@orldc-dev-vnode02 ~]# getenforce
It looks like it isn't SELinux.
Checked also on the storage server ovirt-gluster-nfs?
On Thu, 2015-07-02 at 09:53 +0200, Sandro Bonazzola wrote:
Il 02/07/2015 02:36, Christopher Young
That's actually a local NFS implementation (running on Gluster) so I
wouldn't expect that it would factor in, but I just verified that all
gluster hosts are in Permissive mode. Like I mentioned, I think I may
have some ids wrong or bad links or something since this had previously
If anyone has an experience of the various IDs in hosted-engine.conf
and vm.conf for the Hosted Engine, I believe I need to just verify
everything. I tried a couple of changes, but I feel like I'm just
making this worse, so I've reverted them.
One thing I do not understand well is how a
Ok. It turns out that I had a split-brain on that volume file and
didn't even realize it. Thank both of you guys for keeping me thinking
and putting me on track. I had so many thoughts running through my
head and had run 'gluster volume status engine' (etc.) without checking
for heal status.
Looks normal, hosted engine uses some extra files/leases to track some of it’s
Looks like you might have a gluster problem though, that IO error appears to be
on your hosted engines disk image. Check for split brains and try and initiate
a heal on the files, see what you get.
Mail list logo