You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
Copy file name to clipboardExpand all lines: articles/sap/monitor/provider-ha-pacemaker-cluster.md
+50-22Lines changed: 50 additions & 22 deletions
Display the source diff
Display the rich diff
Original file line number
Diff line number
Diff line change
@@ -21,53 +21,86 @@ In this how-to guide, you learn how to create a high-availability (HA) Pacemaker
21
21
22
22
## Install an HA agent
23
23
24
-
Before you add providers for HA (Pacemaker) clusters, install the appropriate agent for your environment in each cluster node.
24
+
Before you add providers for HA (Pacemaker) clusters, install the appropriate agent for RHEL/SUSE in your environment in each of the cluster node.
25
25
26
26
For SUSE-based clusters, install **ha_cluster_provider** in each node. For more information, see the [HA cluster exporter installation guide](https://github.com/ClusterLabs/ha_cluster_exporter#installation). Supported SUSE versions include SLES for SAP 12 SP3 and later versions.
27
27
28
+
For SUSE-based Pacemaker clusters, Please follow below steps to install in each of the cluster node
29
+
30
+
### Install an HA cluster exporter on SUSE
31
+
32
+
1. Install the required packages for Prometheus cluster exporter on the system.
1. Data is then collected in the system by ha_cluster_exporter. You can export the data via URL `http://<ip address of the server>:9644/metrics`.
49
+
To check if the metrics are fetched via URL on the server where the ha_cluster_exporter is installed, Run below command on the server.
50
+
51
+
```bash
52
+
curl http://localhost:9644/metrics
53
+
```
54
+
28
55
For RHEL-based clusters, install **performance co-pilot (PCP)** and the **pcp-pmda-hacluster** subpackage in each node. For more information, see the [PCP HACLUSTER agent installation guide](https://access.redhat.com/articles/6139852). Supported RHEL versions include 8.2, 8.4, and later versions.
29
56
30
-
For RHEL-based Pacemaker clusters, also install [PMProxy](https://access.redhat.com/articles/6139852)in each node.
57
+
For RHEL-based Pacemaker clusters, Please follow below steps to install in each of the cluster node
31
58
32
59
### Install an HA cluster exporter on RHEL
33
60
34
-
1. Install the required packages on the system.
61
+
1. Install the required packages for PCP on the system.
35
62
36
63
```bash
37
-
yum install pcp pcp-pmda-hacluster
64
+
sudo yum install pcp pcp-pmda-hacluster
38
65
```
39
66
40
67
1. Enable and start the required PCP Collector Services.
41
68
42
69
```bash
43
-
systemctl enable pmcd
70
+
sudo systemctl start pmcd
44
71
```
45
72
46
73
```bash
47
-
systemctl start pmcd
74
+
sudo systemctl enable pmcd
48
75
```
49
76
50
-
1. Install and enable the HA cluster PMDA. Replace `$PCP_PMDAS_DIR` with the path where `hacluster` is installed. Use the `find`commandin Linux to find the path.
77
+
1. Install and enable the HA cluster PMDA. Replace `$PCP_PMDAS_DIR` with the path where `hacluster` is installed. Use the `find`commandin Linux to find the path of "hacluster" bits. usually hacluster will be in path "/var/lib/pcp/pmdas".
78
+
Example :cd /var/lib/pcp/pmdas/hacluster
51
79
52
80
```bash
53
81
cd$PCP_PMDAS_DIR/hacluster
54
82
```
55
83
56
84
```bash
57
-
./Install
85
+
sudo ./Install
58
86
```
59
87
60
88
1. Enable and start the `pmproxy` service.
61
89
62
90
```bash
63
-
systemctl enable pmproxy
91
+
sudo systemctl start pmproxy
64
92
```
65
93
66
94
```bash
67
-
systemctl start pmproxy
95
+
sudo systemctl enable pmproxy
68
96
```
69
97
70
-
1. Data is then collected in the system by PCP. You can export the data by using `pmproxy` at `http://<SERVER-NAME-OR-IP-ADDRESS>:44322/metrics?names=ha_cluster`. Replace `<SERVER-NAME-OR-IP-ADDRESS>` with your server name or IP address.
98
+
1. Data is then collected in the system by PCP. You can export the data by using `pmproxy` via URL `http://<ipaddress of the serrver>:44322/metrics?names=ha_cluster`.
99
+
To check if the metrics are fetched via URL on the server where the hacluster is installed, Run below command on the server.
@@ -88,22 +121,17 @@ To [enable TLS 1.2 or higher](enable-tls-azure-monitor-sap-solutions.md), follow
88
121
1. Configure providers for each node of the cluster by entering the endpoint URL for **HA Cluster Exporter Endpoint**.
89
122
90
123
1. For SUSE-based clusters, enter `http://<IP-address>:9664/metrics`.
91
-
124
+
92
125

93
126
94
127
1. For RHEL-based clusters, enter `http://<'IP address'>:44322/metrics?names=ha_cluster`.
95
128
96
129

97
130
98
-
1. Enter the system identifiers, host names, and cluster names. For the system identifier, enter a unique SAP system identifier for each cluster. For the hostname, the value refers to an actual hostname in the VM. Use `hostname -s` for SUSE- and RHEL-based clusters.
99
-
100
-
1. Select **Add provider** to save.
101
-
102
-
1. Continue to add more providers as needed.
103
-
104
-
1. Select **Review + create** to review the settings.
131
+
1. Enter the SID - SAP system ID, Hostname - SAP hostname of the Virtual machine (Command `hostname -s` for SUSE and RHEL based servers will give hostname detail.), Cluster - Provide any custom name that is easy to identify the SAP system cluster - this Name will be visible in the workbook for metrics (need not have to be the cluster name configured on the server).
105
132
106
-
1. Select **Create** to finish creating the resource.
133
+
1. Click on "Start test" under "Pre-requisite check (Preview) - highly-recommended" - This test will help validate the connectivity from AMS subnet to the SAP source system and list out if any error's found - which need to be addressed before provider creation otherwise the provider creation will fail with error.
134
+
1. Select **Create** to finish creating the Provider.
107
135
108
136
## Troubleshooting
109
137
@@ -116,13 +144,13 @@ When the provider settings validation operation fails with the code `PrometheusU
116
144
1. Restart the HA cluster exporter agent.
117
145
118
146
```bash
119
-
systemctl start pmproxy
147
+
sudo systemctl start pmproxy
120
148
```
121
149
122
150
1. Reenable the HA cluster exporter agent.
123
151
124
152
```bash
125
-
systemctl enable pmproxy
153
+
sudo systemctl enable pmproxy
126
154
```
127
155
128
156
1. Verify that the Prometheus endpoint is reachable from the subnet that you provided when you created the Azure Monitor for SAP solutions resource.
0 commit comments