feat(client): configurable metrics/process intervals with profile persistence; docs updated
This commit is contained in:
@@ -190,6 +190,12 @@ First time you specify a new `--profile/-P` name together with a URL (and option
|
||||
socktop --profile prod ws://prod-host:3000/ws
|
||||
# With TLS pinning:
|
||||
socktop --profile prod-tls --tls-ca /path/to/cert.pem wss://prod-host:8443/ws
|
||||
|
||||
You can also set custom intervals (milliseconds):
|
||||
|
||||
```bash
|
||||
socktop --profile prod --metrics-interval-ms 750 --processes-interval-ms 3000 ws://prod-host:3000/ws
|
||||
```
|
||||
```
|
||||
|
||||
If a profile already exists you will be prompted before overwriting:
|
||||
@@ -214,7 +220,7 @@ socktop --profile prod
|
||||
socktop -P prod-tls # short flag
|
||||
```
|
||||
|
||||
The stored URL (and TLS CA path, if any) will be used. TLS auto-upgrade still applies if a CA path is stored alongside a ws:// URL.
|
||||
The stored URL (and TLS CA path, if any) plus any saved intervals will be used. TLS auto-upgrade still applies if a CA path is stored alongside a ws:// URL.
|
||||
|
||||
### Interactive selection (no args)
|
||||
|
||||
@@ -238,7 +244,12 @@ An example `profiles.json` (pretty‑printed):
|
||||
{
|
||||
"profiles": {
|
||||
"prod": { "url": "ws://prod-host:3000/ws" },
|
||||
"prod-tls": { "url": "wss://prod-host:8443/ws", "tls_ca": "/home/user/certs/prod-cert.pem" }
|
||||
"prod-tls": {
|
||||
"url": "wss://prod-host:8443/ws",
|
||||
"tls_ca": "/home/user/certs/prod-cert.pem",
|
||||
"metrics_interval_ms": 500,
|
||||
"processes_interval_ms": 2000
|
||||
}
|
||||
},
|
||||
"version": 0
|
||||
}
|
||||
@@ -248,6 +259,7 @@ Notes:
|
||||
- The `tls_ca` path is stored as given; if you move or rotate the certificate update the profile by re-running with `--profile NAME --save`.
|
||||
- Deleting a profile: edit the JSON file and remove the entry (TUI does not yet have an in-app delete command).
|
||||
- Profiles are client-side convenience only; they do not affect the agent.
|
||||
- Intervals: `metrics_interval_ms` controls the fast metrics poll (default 500 ms). `processes_interval_ms` controls process list polling (default 2000 ms). Values below 100 ms (metrics) or 200 ms (processes) are clamped.
|
||||
|
||||
---
|
||||
|
||||
|
||||
Reference in New Issue
Block a user