[GH-ISSUE #418] Management Service Read Timeout #621

Closed
opened 2026-08-05 00:39:20 -04:00 by saavagebueno · 2 comments
Owner

Originally created by @diseq on GitHub (Aug 8, 2022).
Original GitHub issue: https://github.com/netbirdio/netbird/issues/418

i'm testing netbird behind nginx. Is there a recommended read timeout value for the management service?

per default, gprc proxy terminates after one minute leading to this:

time="2022-08-08T19:21:14+02:00" level=warning msg="disconnected from the Management service but will retry silently. Reason: rpc error: code = Internal desc = stream terminated by RST_STREAM with error code: INTERNAL_ERROR"

i configured nginx now with this, though not sure if 1h is too much. are there regular management service checkins?

location /management.ManagementService {
  grpc_pass grpc://127.0.0.1:8081;
  grpc_read_timeout 60m;              
}
Originally created by @diseq on GitHub (Aug 8, 2022). Original GitHub issue: https://github.com/netbirdio/netbird/issues/418 i'm testing netbird behind nginx. Is there a recommended read timeout value for the management service? per default, gprc proxy terminates after one minute leading to this: ``` time="2022-08-08T19:21:14+02:00" level=warning msg="disconnected from the Management service but will retry silently. Reason: rpc error: code = Internal desc = stream terminated by RST_STREAM with error code: INTERNAL_ERROR" ``` i configured nginx now with this, though not sure if 1h is too much. are there regular management service checkins? ``` location /management.ManagementService { grpc_pass grpc://127.0.0.1:8081; grpc_read_timeout 60m; } ```
Author
Owner

@braginini commented on GitHub (Aug 8, 2022):

Hey @diseq
There is no polling mechanism (check-ins). Instead, every NetBird agent keeps an open gRPC stream with the Management service. Once there is something to send (e.g., network map update) the Management service uses this stream to deliver updates to the agents. So there should be no timeout at all leaving the stream open indefinitely. If the connection breaks, agents will reconnect automatically and continue to operate normally.

<!-- gh-comment-id:1208437744 --> @braginini commented on GitHub (Aug 8, 2022): Hey @diseq There is no polling mechanism (check-ins). Instead, every NetBird agent keeps an open gRPC stream with the Management service. Once there is something to send (e.g., network map update) the Management service uses this stream to deliver updates to the agents. So there should be no timeout at all leaving the stream open indefinitely. If the connection breaks, agents will reconnect automatically and continue to operate normally.
Author
Owner

@diseq commented on GitHub (Aug 9, 2022):

thanks. well understood!

<!-- gh-comment-id:1208970888 --> @diseq commented on GitHub (Aug 9, 2022): thanks. well understood!
Sign in to join this conversation.
No Label
1 Participants
Notifications
Due Date
No due date set.
Dependencies

No dependencies set.

Reference: DYNR/netbird#621