ceph-mgr crash when enabling perf stats

[ikogan]

I'm having some performance issues on CephFS I'm trying to track down. I tried enabling stats following the information on https://docs.ceph.com/en/quincy/cephfs/cephfs-top/. The second I do ceph mgr module enable stats, all of my MDSes start to puke the following:

2023-02-19T21:19:31.164-0500 7f8e215ef700  1 mgr handle_mgr_map respawning because set of enabled modules changed!
    metric_features = int(metadata[CLIENT_METADATA_KEY]["metric_spec"]["metric_flags"]["feature_bits"], 16)
  File "/usr/share/ceph/mgr/stats/fs/perf_stats.py", line 174, in notify_cmd
    self.fs_perf_stats.notify_cmd(notify_id)
  File "/usr/share/ceph/mgr/stats/module.py", line 32, in notify
2023-02-19T21:19:31.032-0500 7f8e11bbb700 -1 mgr notify Traceback (most recent call last):
2023-02-19T21:19:31.032-0500 7f8e11bbb700 -1 mgr notify stats.notify:
ValueError: invalid literal for int() with base 16: '0x'
    metric_features = int(metadata[CLIENT_METADATA_KEY]["metric_spec"]["metric_flags"]["feature_bits"], 16)
  File "/usr/share/ceph/mgr/stats/fs/perf_stats.py", line 174, in notify_cmd
    self.fs_perf_stats.notify_cmd(notify_id)
  File "/usr/share/ceph/mgr/stats/module.py", line 32, in notify

All of my nodes are on Ceph 17.2.5. It looks like the value of feature_bits is likely 0x, which I see some other projects having issues with. How are these feature bits controlled?