From a365b82d42a577774e327ad3bf8a2ef3dbf6a00f Mon Sep 17 00:00:00 2001
From: larryruili <larryruili@google.com>
Date: Fri, 16 Feb 2018 13:49:38 -0800
Subject: [PATCH] Update publish queue with practical retry params

The unlimited exponential backoff makes cascading failure a serious problem,
when encountering burst DNS load. Originally, it was exponential backoff, with min 1 sec max 1 hour.

This changes it to be linearly scaling from
30 seconds to 10 minutes. Min 30 seconds is used to avoid over-retrying due to lock contention. Max 10 minutes allows for more retries within our 1 hour SLA. Finally, we're
switching to linear scaling to increase the number of 'quick' retries for low
backoff time, before ultimately settling on the upper bound of 10 minutes (if a
task ever gets to that point, it's probably misconfigured.)

-------------
Created by MOE: https://github.com/google/moe
MOE_MIGRATED_REVID=186041553
---
 java/google/registry/env/common/default/WEB-INF/queue.xml | 6 ++++++
 1 file changed, 6 insertions(+)

diff --git a/java/google/registry/env/common/default/WEB-INF/queue.xml b/java/google/registry/env/common/default/WEB-INF/queue.xml
index 61e29cfee..77b050968 100644
--- a/java/google/registry/env/common/default/WEB-INF/queue.xml
+++ b/java/google/registry/env/common/default/WEB-INF/queue.xml
@@ -10,6 +10,12 @@
     <name>dns-publish</name>
     <rate>100/s</rate>
     <bucket-size>100</bucket-size>
+    <!-- 30 sec backoff increasing linearly up to 10 minutes. -->
+    <retry-parameters>
+      <min-backoff-seconds>30</min-backoff-seconds>
+      <max-backoff-seconds>600</max-backoff-seconds>
+      <max-doublings>0</max-doublings>
+    </retry-parameters>
   </queue>
 
   <queue>