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
- Clean separation between app runtime and managed database
64
113
65
114
### Cons
66
-
- Multiple connection patterns to support temporarily
115
+
- Requires Kubernetes + Helm + ingress setup
116
+
- Requires platform/DNS coordination
67
117
68
118
---
69
119
70
-
## Client Routing Rules
120
+
## Pointing Clients at Your Deployment
71
121
72
122
Compilation/test clients route as follows:
73
123
74
124
1. If `DEVFEEDBACK_URL` is set, use it.
75
125
2. Otherwise, use default `http://compilation-metrics`.
76
126
77
-
Recommended enterprise pattern:
78
-
- Keep client defaults untouched.
79
-
- Control destination centrally with DNS.
127
+
### Preferred enterprise pattern: DNS default host
80
128
81
-
---
129
+
Most corporate networks already use internal DNS for service discovery (for internal APIs, proxies, package mirrors, etc.).
82
130
83
-
## Example environment variables
131
+
Because clients default to `http://compilation-metrics`, your network/workstation team can create an internal DNS record for `compilation-metrics` pointing to your telemetry API endpoint (ingress/internal LB/service host). Once this is in place:
84
132
85
-
### API host
133
+
- no per-developer setup is required,
134
+
- telemetry works out-of-the-box,
135
+
- rollout and changes stay centralized with infra/network teams.
0 commit comments