What were you trying to do that didn't work?

We use NFS with krb5 authentication. When a users' ticket expires and still has processes running, rpc.gssd starts to consume a lot of cpu. rpc.gssd seems to continuously poll sssd-kcm which also starts to consume 100% at this point.

Please provide the package NVR for which bug is seen:nfs-utils-2.5.4-25.el9.x86_64

How reproducible: 100%

Steps to reproduce

1. Setup long running job that reads and writes a simple file on kerberized nfs
2. use a ticket with a short lifetime to access the nfs files filesystem
3. ticket expires
4. watch rpc.gssd begin to consume more and more resources as it queues the read/write ops

Expected results

rpc.gssd does not cripple the system

Actual results

NFS access to home directory is blocked and rpc.gssd (and as a result sssd-kcm) starts to consume large amounts of cpu.

Additional information:

https://bugzilla.redhat.com/show_bug.cgi?id=1906712