Description of problem:
NFS cluster qos enablement is not support via orch apply. When applying nfs clusters at scale, issuing explicit nfs cluster qos enable causes the nfs daemons to get restarted. Since the restart behavior is synchronous, the backlog of work causes the orch module to get hung up on restarts. Rather than sending all the daemons through an explicit restart, there should be support on the orch apply when applying a new nfs spec to specify qos config options at deploy time.
Version-Release number of selected component (if applicable): 8.0z4
How reproducible:
Always
Steps to Reproduce:
1. Create 100 nfs clusters from specs via orch apply
2. Apply QOS bandwidth to all 100 nfs clusters
3. Apply QOS ops to all 100 clusters
4. Notice that ceph-mgr starts issuing systemctl reset-failed and systemctl restarts to all the deployed daemons
Actual results:
All daemons are restarted when QOS is applied even if immediately after being provisioned.
Expected results:
All clusters should be able to provision with QOS config and no dameons should be restarted.
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 (Red Hat Ceph Storage 8.1 security and bug fix updates), 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/RHBA-2025:14015