Monitoring
Monitoring endpoints are part of the Admin Console REST interface.
REST API
To enable monitoring "rest user" with "rest_getMetrics" permission must be added to the webswing.config file.
For example:
{
"/" : {
"path" : "/",
"restUsers": [
{
"permissions": [
"rest_getMetrics"
],
"password": "pwd",
"username": "rest_user"
}
]
}
}
Services
There are 3 rest services:
| Name | Description | Parameters to be monitored | |
|---|---|---|---|
| 1. | /rest/metrics/app | Info about connected, disconnected, idle and total applications and their stats | app_count app_total_count app_memoryUsed_percent app_cpuUtilization_percent app_latency_second app_latencyNetworkTransfer_second |
| 2. | /rest/metrics/sp | Info about connected, disconnected, idle and total applications | sp_apps_count sp_apps_total_count |
| 3. | /rest/metrics | Info about cluster server, session pools and their stats | overview_basic_stats_count overview_server_stats_count overview_server_flasg_info overview_sp_stats_count overview_server_flags_info overview_count |
These services can be found in the file:
\webswing-home\webswing-admin\webswing-admin-server\src\main\resources\openapi\webswing_api.yaml
Metrics list
| Name | Description | Labels | |
|---|---|---|---|
| 1. | app_count | Informations about applications | app: "Application name" state: "connected, disconnected, idle, total" |
| 2. | app_total_count | Sum informations about applications | state: "connected, disconnected, idle, total" |
| 3. | app_memoryUsed_percent | Memory used for application in percent | app: "Application name" aggr: "MIN, MAX, AVG" |
| 4. | app_cpuUtilization_percent | CPU utilization for application in percent | app: "Application name" aggr: "MIN, MAX, AVG" |
| 5. | app_latency_second | Latency in seconds | app: "Application name" aggr: "MIN, MAX, AVG" |
| 6. | app_latencyNetworkTransfer_second | Latency network transfer in seconds | app: "Application name" aggr: "MIN, MAX, AVG" |
| 7. | sp_apps_count | Informations about sessionpools | sessionpool: "Sessionpool ID" app: "Application name" state: "connected, disconnected, idle, total" |
| 8. | sp_apps_total_count | Sum informations about sessionpools | sessionpool: "Sessionpool ID" state: "connected, disconnected, idle, total" |
| 9. | overview_basic_stats_count | Informations about basic stats | flag: |
| 10. | overview_server_stats_count | Informations about servers | server: "Server name" flag: "session_pool, available_apps, running_apps, active_users, sessions, sessions_connected, connections" |
| 11. | overview_server_flags_info | Informations about servers | server: "Server name" flag: "verbose_logging" |
| 12. | overview_sp_stats_count | Informations about sessionpools | sessionpool: "Sessionpool ID" flag: "available_apps, running_apps, connected_servers, idle_apps, max_instances, max_idle_instances, priority" |
| 13. | overview_sp_flags_info | Informations about sessionpools | sessionpool: "Sessionpool ID" flag: "drain_mode, drain_auto_mode, static_sp, verbose_logging" |
| 14. | overview_count | Informations about servers and sessionpools | flag: "active_servers", "active_sp" |
Prometheus
To use this with Prometheus, you must add jobs to your prometheus.yml file.
Here is a docker example:
docker-compose.yml
prometheus:
image: prom/prometheus:v3.1.0
container_name: prometheus
depends_on:
- adminconsole
volumes:
- ./prometheus.yml:/etc/prometheus/prometheus.yml
ports:
- "9090:9090"
prometheus.yml
global:
scrape_interval: 30s
scrape_configs:
- job_name: 'ws-metrics-app'
metrics_path: '/rest/metrics/app'
basic_auth:
username: 'rest_user'
password: 'pwd'
static_configs:
- targets:
- 'host.docker.internal:8083'
- job_name: 'ws-metrics-sp'
metrics_path: '/rest/metrics/sp'
basic_auth:
username: 'rest_user'
password: 'pwd'
static_configs:
- targets:
- 'host.docker.internal:8083'
- job_name: 'ws-metrics'
metrics_path: '/rest/metrics'
basic_auth:
username: 'rest_user'
password: 'pwd'
static_configs:
- targets:
- 'host.docker.internal:8083'