daemon-base: bump nfs-ganesha 4 > 5 (quincy/el9)

[guits]

Quincy el9 based image should ship nfs-ganesha-5.
(nfs-ganesha-4 doesn't exist for CentOS 8)

Fixes: https://tracker.ceph.com/issues/66253

[guits]

@ljflores FYI

[ljflores]

Actually @guits it looks like the CI failures below only occur on c9 jobs:

CONTAINER ID   IMAGE     COMMAND   CREATED   STATUS    PORTS     NAMES
CONTAINER ID   IMAGE     COMMAND   CREATED   STATUS    PORTS     NAMES
Error response from daemon: No such container: ceph-demo
Error: Process completed with exit code 1.

I would recheck the "ceph-demo" references.

[guits]

Actually @guits it looks like the CI failures below only occur on c9 jobs:

CONTAINER ID   IMAGE     COMMAND   CREATED   STATUS    PORTS     NAMES
CONTAINER ID   IMAGE     COMMAND   CREATED   STATUS    PORTS     NAMES
Error response from daemon: No such container: ceph-demo
Error: Process completed with exit code 1.

I would recheck the "ceph-demo" references.

No. You are seeing this because the ceph-release RPM isn't available at download.ceph.com for both squid and quincy.
Those github workflow tests run a set of command, including a podman ps -a when a failure is caught. Obviously, if an error happens before the container is started, you won't see anything.

By the way, reef el9 test which has well its ceph-release RPM at download.ceph.com works, see [1]

[1] https://github.com/ceph/ceph-container/actions/runs/9289060252/job/25586978591?pr=2213

[yuriw]

@guits pls merge!

[guits]

closing in favor of #2218