mgr/devicehealth: fix daemon filtering before scraping device

Scraping health metrics of mon devices was introduced in Nautilus, then
disabled (only in Nautilus) since the 'tell' mechanism of mons was not
reliable.
This commit fixes a bug when filtering the daemons on the device to be
scrapped (and allows scraping osd devices solely).

When:
$ ceph device scrape-health-metrics seagate_123
Error EAGAIN: device seagate_123 not claimed by any active OSD daemons

But:
$ ceph device ls
DEVICE        	HOST:DEV     	DAEMONS	  LIFE EXPECTANCY
seagate_123	hostname:sdc	osd.1

See:
Introducing the feature in Nautilus:
PR: https://github.com/ceph/ceph/pull/24151
commit: 8d786003a44313c85c038cbdc9bd9875da197ab1

Disabling the feature in Nautilus:
PR: https://github.com/ceph/ceph/pull/31446
commit: d592e56e74d94c6a05b9240fcb0031868acefbab

This issue exists only in Nautilus and does not require fixes in any other branch.

Fixes: https://tracker.ceph.com/issues/46871
Signed-off-by: Yaarit Hatuka <yaarit@redhat.com>

diff --git a/src/pybind/mgr/devicehealth/module.py b/src/pybind/mgr/devicehealth/module.py
index 53826d094de28ec92834c4420230d93c1deffd20..9bddc66f5f33d78570d7e64940caaedcf3a2d73f 100644 (file)
--- a/src/pybind/mgr/devicehealth/module.py
+++ b/src/pybind/mgr/devicehealth/module.py
@@ -351,7 +351,7 @@ class Module(MgrModule):
         r = self.get("device " + devid)
         if not r or 'device' not in r.keys():
             return -errno.ENOENT, '', 'device ' + devid + ' not found'
-        daemons = [d for d in r['device'].get('daemons', []) if not d.startswith('osd.')]
+        daemons = [d for d in r['device'].get('daemons', []) if d.startswith('osd.')]
         if not daemons:
             return (-errno.EAGAIN, '',
                     'device ' + devid + ' not claimed by any active OSD daemons')