Azure / Azure/Azure-Spring-Apps

partitionOwnershipExpirationInterval setting not applied

オープン
#60 コメント 1 件 リアクション 0 件 担当者 0 名 GitHub で見る
主要言語
言語のデータがありません
スター
9
フォーク
11
PR マージ指標
30日以内にマージされた PR はありません

説明

**Describe the bug**
We are using **spring-cloud-azure-starter-eventhubs** version 7.4.0. We have configured rebalancing and partition ownership expiration internal as per documentation:

```
spring:
cloud:
azure:
eventhubs:
processor:
load-balancing:
partition-ownership-expiration-interval: 30s
update-interval: 10s
```

We can clearly see in logs that load balancing is performed every 10s, as configured, but partition ownership is only re-assigned after the default 2 minutes. The configuration property is ignored.

**To Reproduce**
Create an eventhubs consumer with above configuration. It is sufficient to make it listen to a topic with only one partition configured. Then stop the pod/instance, at which point a new instance will automatically be started up. In the logs, it can be seen that the new instance performs load balancing every 10s, but only after 2 minutes it obtains ownership of the partition that was released by the earlier terminated listener.

**Expected behavior**
The new consumer should be granted ownership of the partition after 30s, as per configuration.

**Screenshots**
n/a

**Additional context**
n/a

**Can we contact you for additional details?**
Y

If yes, please send us your contact information to AzureSpringCloud-Talk@service.microsoft.com and include the issue number in the email title.

コントリビューションガイド

このリポジトリのコントリビューションガイドは索引されていません

調査の方向性

Start by tracing the Spring Cloud Azure Event Hubs processor configuration for partition-ownership-expiration-interval and its load-balancing setup. Reproduce with a single-partition topic, a 30-second expiration interval, and a terminated instance; done means the replacement consumer obtains ownership after about 30 seconds rather than the default two minutes.

索引モデルが issue の本文から書いたものです。

評価

技術スタック
azure, spring, spring-boot
領域
cloud, distributed-systems
issue の種類
バグ
難易度
3/5
見積もり時間
1〜2日
活発さ
活発
明瞭さ
おおむね明確
初心者へのやさしさ
55/100

新しい issue をメールで受け取る

初心者向けの GitHub issue を短くまとめたダイジェスト。