Description of problem:
The cephadm serve loop sequentially processes daemons managed by ceph. Our use-case provisions/deprovisions nfs ganesha clusters frequently and is consistently queuing work on the serve loop thread causing long provision/deprovision times. This also affects upgrade and failover.
Request to leverage asyncio tasks to concurrently process daemons within the serve loop wherever applicable to optimize the daemon orchestration process, especially around the nfs daemons.
Version-Release number of selected component (if applicable): 8.0z3
How reproducible: 100%
Steps to Reproduce:
1. Provision multiple nfs ganesha clusters (each with multiple daemons)
2. Notice the time complexity it takes to completely provision all daemons
3. Delete in mass many nfs clusters
4. Provision a new set of nfs clusters
5. Notice while the serve loop thread is still in the daemon deletion handling, all the new provisions are stuck provisioning until the deletion thread is finished.
Actual results:
Sequential processing of all daemon orchestration
Expected results:
Concurrent processing of nfs daemons (and other daemons if applicable)
Additional info:
Since the problem described in this bug report should be
resolved in a recent advisory, it has been closed with a
resolution of ERRATA.
For information on the advisory (Moderate: Red Hat Ceph Storage 9.0 Security and Enhancement update), and where to find the updated
files, follow the link below.
If the solution does not work for you, open a new bug report.
https://access.redhat.com/errata/RHSA-2026:1536