Only one of the three proxy pods can provide services, and the other two pods that cannot provide services have been continuously reporting

[jackin853]

Only one of the three proxy pods can provide services, and the other two pods that cannot provide services have been reported continuously：
not proxying to master address since we aren't in the enabled proxies list {"address": "177.177.208.189:5432"}"
And I directly connected to this good proxy pod through the psql command, and I was able to successfully access the keeper instance, while the other two failed
The entire environment uses etcd as storage. It has been normal during long-term operation. Suddenly, our business service reported that postgresql could not be connected. After positioning, it was found that the proxy could not be accessed through the k8s service. Two of the proxy could not provide services, only One is good, both sentinel and keeper pods are ok: