From eff32aab979f87c7ed59b8b3843c5c882bc491c8 Mon Sep 17 00:00:00 2001
From: Sage Weil <sage@redhat.com>
Date: Fri, 4 Oct 2019 15:03:53 -0500
Subject: [PATCH] mgr/telemetry: include device telemetry

- anonymize device id
- anonymize host names
- strip out the serial number

Signed-off-by: Sage Weil <sage@redhat.com>
---
 src/pybind/mgr/devicehealth/module.py |  8 +++--
 src/pybind/mgr/telemetry/module.py    | 47 +++++++++++++++++++++++++--
 2 files changed, 49 insertions(+), 6 deletions(-)

diff --git a/src/pybind/mgr/devicehealth/module.py b/src/pybind/mgr/devicehealth/module.py
index 1955bec661e71..78bbc3166755f 100644
--- a/src/pybind/mgr/devicehealth/module.py
+++ b/src/pybind/mgr/devicehealth/module.py
@@ -641,6 +641,8 @@ class Module(MgrModule):
         except:
             return -1, '', 'unable to invoke diskprediction local or remote plugin'
 
-    def gather_device_report(self):
-        # write me
-        return {}
+    def get_recent_device_metrics(self, devid, min_sample):
+        return self._get_device_metrics(devid, min_sample=min_sample)
+
+    def get_time_format(self):
+        return TIME_FORMAT
diff --git a/src/pybind/mgr/telemetry/module.py b/src/pybind/mgr/telemetry/module.py
index 769fc72bc62a5..8a260b27176cd 100644
--- a/src/pybind/mgr/telemetry/module.py
+++ b/src/pybind/mgr/telemetry/module.py
@@ -11,7 +11,7 @@ import re
 import requests
 import uuid
 import time
-from datetime import datetime
+from datetime import datetime, timedelta
 from threading import Event
 from collections import defaultdict
 
@@ -129,7 +129,7 @@ class Module(MgrModule):
             'name': 'channel_device',
             'type': 'bool',
             'default': True,
-            'description': 'Share device health metrics (e.g., SMART data)',
+            'description': 'Share device health metrics (e.g., SMART data, minus potentially identifying info like serial numbers)',
         },
     ]
 
@@ -293,9 +293,50 @@ class Module(MgrModule):
 
     def gather_device_report(self):
         try:
-            return self.remote('devicehealth', 'gather_device_report')
+            time_format = self.remote('devicehealth', 'get_time_format')
         except:
             return None
+        cutoff = datetime.utcnow() - timedelta(hours=self.interval * 2)
+        min_sample = cutoff.strftime(time_format)
+
+        devices = self.get('devices')['devices']
+
+        res = {}
+        for d in devices:
+            devid = d['devid']
+            try:
+                m = self.remote('devicehealth', 'get_recent_device_metrics',
+                                devid, min_sample)
+            except:
+                continue
+
+            # anonymize host id
+            try:
+                host = d['location'][0]['host']
+            except:
+                continue
+            anon_host = self.get_store('host-id/%s' % host)
+            if not anon_host:
+                anon_host = str(uuid.uuid1())
+                self.set_store('host-id/%s' % host, anon_host)
+            m['host_id'] = anon_host
+
+            # anonymize device id
+            (vendor, model, serial) = devid.split('_')
+            anon_devid = self.get_store('devid-id/%s' % devid)
+            if not anon_devid:
+                anon_devid = '%s_%s_%s' % (vendor, model, uuid.uuid1())
+                self.set_store('devid-id/%s' % devid, anon_devid)
+
+            self.log.info('devid %s / %s, host %s / %s' % (devid, anon_devid,
+                                                           host, anon_host))
+
+            # anonymize the smartctl report itself
+            for k in ['serial_number']:
+                m.pop(k)
+
+            res[anon_devid] = m
+        return res
 
     def compile_report(self, channels=[]):
         if not channels:
-- 
2.39.5