mgr/telemetry: add stats about crush map

Signed-off-by: Sage Weil <sage@redhat.com>

diff --git a/PendingReleaseNotes b/PendingReleaseNotes
index de3dfa6e472de09a244ee18a5d84907ae951e6de..9edf474c8faf463b82f091140eb9df97e5525ac8 100644 (file)
--- a/PendingReleaseNotes
+++ b/PendingReleaseNotes
@@ -214,6 +214,7 @@
     - whether a separate OSD cluster network is being used
     - how many RBD pools and images are in the cluster, and how many pools have RBD mirroring enabled
     - how many RGW daemons, zones, and zonegroups are present; which RGW frontends are in use
+    - aggregate stats about the CRUSH map, like which algorithms are used, how big buckets are, how many rules are defined, and what tunables are in use
 
   If you had telemetry enabled, you will need to re-opt-in with::
 

diff --git a/src/pybind/mgr/telemetry/module.py b/src/pybind/mgr/telemetry/module.py
index c7cc5738d75bbfffe8963cda6ad275914ede8c9d..8eea3b21eda4738a2cdf114b2e6f456a66e017ff 100644 (file)
--- a/src/pybind/mgr/telemetry/module.py
+++ b/src/pybind/mgr/telemetry/module.py
@@ -57,6 +57,7 @@ REVISION = 3
 #   - whether an OSD cluster network is in use
 #   - rbd pool and image count, and rbd mirror mode (pool-level)
 #   - rgw daemons, zones, zonegroups; which rgw frontends
+#   - crush map stats
 
 class Module(MgrModule):
     config = dict()
@@ -245,6 +246,45 @@ class Module(MgrModule):
 
         return metadata
 
+    def gather_crush_info(self):
+        osdmap = self.get_osdmap()
+        crush_raw = osdmap.get_crush()
+        crush = crush_raw.dump()
+
+        def inc(d, k):
+            if k in d:
+                d[k] += 1
+            else:
+                d[k] = 1
+
+        device_classes = {}
+        for dev in crush['devices']:
+            inc(device_classes, dev.get('class', ''))
+
+        bucket_algs = {}
+        bucket_types = {}
+        bucket_sizes = {}
+        for bucket in crush['buckets']:
+            if '~' in bucket['name']:  # ignore shadow buckets
+                continue
+            inc(bucket_algs, bucket['alg'])
+            inc(bucket_types, bucket['type_id'])
+            inc(bucket_sizes, len(bucket['items']))
+
+        return {
+            'num_devices': len(crush['devices']),
+            'num_types': len(crush['types']),
+            'num_buckets': len(crush['buckets']),
+            'num_rules': len(crush['rules']),
+            'device_classes': list(device_classes.values()),
+            'tunables': crush['tunables'],
+            'compat_weight_set': '-1' in crush['choose_args'],
+            'num_weight_sets': len(crush['choose_args']),
+            'bucket_algs': bucket_algs,
+            'bucket_sizes': bucket_sizes,
+            'bucket_types': bucket_types,
+        }
+
     def gather_configs(self):
         # cluster config options
         cluster = set()
@@ -472,6 +512,9 @@ class Module(MgrModule):
                 'cluster_network': cluster_network,
             }
 
+            # crush
+            report['crush'] = self.gather_crush_info()
+
             # cephfs
             report['fs'] = {
                 'count': len(fs_map['filesystems']),