From 08285f5de7a0690f981a6ea28d1db750a0d05376 Mon Sep 17 00:00:00 2001 From: Lai Jiang Date: Wed, 20 Dec 2023 15:37:29 -0500 Subject: [PATCH] Greatly increase the upper limit of proxy instances in production (#2259) From our investigation, the Monday night WHOIS storm does not cause any strain to the backend system. The backend latency metrics are all well within the limits. The latency measured from the proxy matches observed latency by the prober, and we see that the "used" CPU is 1.5x of "requested" CPU during the time when the latency is above the threshold. Making this change hopefully removes the proxy as the bottleneck and ameliorate the pages. --- proxy/kubernetes/proxy-service.yaml | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/proxy/kubernetes/proxy-service.yaml b/proxy/kubernetes/proxy-service.yaml index c75e709ed..da162c83c 100644 --- a/proxy/kubernetes/proxy-service.yaml +++ b/proxy/kubernetes/proxy-service.yaml @@ -46,5 +46,5 @@ spec: apiVersion: apps/v1 kind: Deployment name: proxy-deployment - maxReplicas: 10 - minReplicas: 1 + maxReplicas: 50 + minReplicas: 10