CI

Another timeout on braggi10: https://jenkins.ceph.com/job/ceph-windows-pull-requests/35396/

The original timeout also happened on braggi10. I'm seeing other checks on main PRs pass on other nodes, so it might be node-specific.

Looks like it's not really a Windows job or environment issue, but an actual regression introduced by that specific patch.

Messages from Windows hosts are getting can no longer be decoded on the OSD side:

2024-02-17T15:28:20.348+0000 7f3bead9b640 -1 failed to decode message of type 42 v9: void tracing::decode(jspan_context&, ceph::buffer::v15_2_0::list::const_iterator&) decode past end of struct encoding: Malformed input [buffer:3]

We need to ensure that builds without telemetry are compatible with these that have telemetry enabled. We actually had a similar issue before but I think that work got reverted.

https://github.com/ceph/ceph/pull/51043
https://github.com/ceph/ceph/pull/47457
https://github.com/ceph/ceph/pull/51331