Hello,
I'm trying to mount a Ceph filesystem from a client within my Kubernetes
cluster. The client is dual-stacked (IPv4/IPv6) and the whole Ceph
platform is IPv6-only. The client can reach the monitor, osd and mds
servers via it's own default gateway. Ping and manual mount [1] in
terminal is possible.
But if I'm trying to mount the Ceph filesystem via Kubernetes/Docker (in
our case Rancher is on top, but that shouldn't matter here), I got in
trouble with name resolution, but I don't know why.
dockerd-current[1179]: I0816 13:39:39.569482 1567 reconciler.go:252]
operationExecutor.MountVolume started for volume "cephfs-rancher-ctest"
(UniqueName: "kubernetes.>
kernel: libceph: parse_ips bad ip
'prod-mon-101.ceph.example,prod-mon-201.ceph.example.com,prod-mon-301.ceph.exemple.com'
kdockerd-current[1179]: E0816 13:39:39.576555 1567
mount_linux.go:151] Mount failed: exit status 32
The client is running Fedora Atomic Host 28 (kernel 4.17.11-200.fc28
with ceph version 12.2.7). As far as I can see key.dns_resolver isn't
used because it is built as a module and not loaded. I have experienced
similar trouble with CoreOS which has dns_resolver enabled in Kernel config.
Have I missed something somewhere? Dig/host resolves a quad-A record for
each of the used monitor servers.
Kind regards
Jan
[1] like: mount -t ceph -o name=user,secretfile=/etc/ceph/secret
prod-mon-101.ceph.example.com,prod-mon-201.ceph.example.com,prod-mon-301.ceph.example.com:/user/userdir
/var/mnt/
_______________________________________________
ceph-users mailing list
ceph-users@xxxxxxxxxxxxxx
http://lists.ceph.com/listinfo.cgi/ceph-users-ceph.com
