Hi Tony,
Thank you for your response.
The reason I raised this here is because our OpenStack deployment uses Ceph as the backend storage, and we’re exploring end-to-end tracing and observability across both OpenStack and Ceph components.
Since tracing on the Ceph side affects visibility for OpenStack services that rely on it (like Cinder, Glance, and Nova), I wanted to check if anyone in the community has encountered or addressed this integration scenario.
If this still falls completely under Ceph’s scope, I’ll move the discussion to the Ceph mailing list accordingly.
Thanks again for your clarification and support.
On Tue, 21 Oct 2025 at 15:58, Thamanna Farhath <thamanna.f@zybisys.com> wrote:
>
> Hi Ceph Team,
>
> We’ve enabled Jaeger tracing in our Ceph cluster using:
>
>
> ceph config set global jaeger_tracing_enable true
>
> Currently, only OSD traces (e.g., enqueue_op, op-request-created) are visible, while MON, MGR, and RGW traces are missing.
>
> I noticed PR #59365 adds more tracers but hasn’t been merged yet. Could you please confirm:
>
> If this is an ongoing bug or current limitation?
>
> Whether there’s an alternative branch, backport, or workaround to achieve full tracing coverage?
Hi Thamanna,
I'm not certain you've sent this to the correct mailing list.
OpenStack services can be backed by ceph resources but doesn't have
any specific sway or insight into that project's development process
or governance.
If there's an aspect of your query that is OpenStack specific please
reframe and make that clearer.
Tony.