Gitlab agent install from Helm - Args timeout?

Hi all,
I have Omnibus Gitlab CE 14.10.5 on premise.
I have a k8s cluster on a domestic cloud provider.
VPN site to site is setup and working.
When I install gitlab agents on k8s cluster to connect to my Gitlab projects , it takes longtime until connection status is connected.
I found logs from agent pod
{"level":"error","time":"2023-10-24T08:20:57.353Z","msg":"Error handling a connection","mod_name":"reverse_tunnel","error":"Connect(): rpc error: code = Unavailable desc = connection error: desc = \"transport: Error while dialing failed to WebSocket dial: failed to send handshake request: Get \\\"\\\": net/http: TLS handshake timeout\""}
Is there anyway to set this timeout for agent pods ?
I read values.yaml · main · / charts / gitlab-agent · GitLab and found there is

extraArgs: []
# Add additional args settings to the pod.

But I’m not sure how to do it , please give me some advice, thank you very much.