From 2b4d980685f84fea69ed8aeb84017d9d015b4e9b Mon Sep 17 00:00:00 2001
From: "Joshua M. Boniface" <joshua@boniface.me>
Date: Thu, 13 Aug 2020 15:06:19 -0400
Subject: [PATCH] Display Ceph health in PVC status as well

Makes this output a little more realistic and allows proper monitoring
of the Ceph cluster status (separate from the PVC status which is
tracking only OSD up/in state).
---
 api-daemon/pvcapid/flaskapi.py |  4 ++++
 client-cli/cli_lib/cluster.py  |  9 +++++++++
 daemon-common/cluster.py       | 10 ++++++++++
 3 files changed, 23 insertions(+)

diff --git a/api-daemon/pvcapid/flaskapi.py b/api-daemon/pvcapid/flaskapi.py
index 5f4a95f3..c9e083f1 100755
--- a/api-daemon/pvcapid/flaskapi.py
+++ b/api-daemon/pvcapid/flaskapi.py
@@ -359,6 +359,10 @@ class API_Status(Resource):
                   type: string
                   description: The overall cluster health
                   example: Optimal
+                storage_health:
+                  type: string
+                  description: The overall storage cluster health
+                  example: Optimal
                 primary_node:
                   type: string
                   description: The current primary coordinator node
diff --git a/client-cli/cli_lib/cluster.py b/client-cli/cli_lib/cluster.py
index 8cb08600..f7f58657 100644
--- a/client-cli/cli_lib/cluster.py
+++ b/client-cli/cli_lib/cluster.py
@@ -92,10 +92,19 @@ def format_info(cluster_information, oformat):
     else:
         health_colour = ansiprint.yellow()
 
+    if cluster_information['storage_health'] == 'Optimal':
+        storage_health_colour = ansiprint.green()
+    elif cluster_information['storage_health'] == 'Maintenance':
+        storage_health_colour = ansiprint.blue()
+    else:
+        storage_health_colour = ansiprint.yellow()
+
     ainformation = []
     ainformation.append('{}PVC cluster status:{}'.format(ansiprint.bold(), ansiprint.end()))
     ainformation.append('')
     ainformation.append('{}Cluster health:{}      {}{}{}'.format(ansiprint.purple(), ansiprint.end(), health_colour, cluster_information['health'], ansiprint.end()))
+    ainformation.append('{}Storage health:{}      {}{}{}'.format(ansiprint.purple(), ansiprint.end(), storage_health_colour, cluster_information['storage_health'], ansiprint.end()))
+    ainformation.append('')
     ainformation.append('{}Primary node:{}        {}'.format(ansiprint.purple(), ansiprint.end(), cluster_information['primary_node']))
     ainformation.append('{}Cluster upstream IP:{} {}'.format(ansiprint.purple(), ansiprint.end(), cluster_information['upstream_ip']))
     ainformation.append('')
diff --git a/daemon-common/cluster.py b/daemon-common/cluster.py
index 35e0edd8..df083880 100644
--- a/daemon-common/cluster.py
+++ b/daemon-common/cluster.py
@@ -127,6 +127,15 @@ def getClusterInformation(zk_conn):
     else:
         cluster_health = 'Optimal'
 
+    # Find out our storage health from Ceph
+    ceph_health = zkhandler.readdata(zk_conn, '/ceph').split('\n')[2].split()[-1]
+    if maint_state == 'true':
+        storage_health = 'Maintenance'
+    elif ceph_health != 'HEALTH_OK':
+        storage_health = 'Degraded'
+    else:
+        storage_health = 'Optimal'
+
     # State lists
     node_state_combinations = [
         'run,ready', 'run,flush', 'run,flushed', 'run,unflush',
@@ -174,6 +183,7 @@ def getClusterInformation(zk_conn):
     # Format the status data
     cluster_information = {
         'health': cluster_health,
+        'storage_health': storage_health,
         'primary_node': common.getPrimaryNode(zk_conn),
         'upstream_ip': zkhandler.readdata(zk_conn, '/upstream_ip'),
         'nodes': formatted_node_states,