Skip to content

Commit fd29edb

Browse files
authoredApr 29, 2024··
WIP serializeImagePulls*
1 parent 6fd6d7a commit fd29edb

File tree

1 file changed

+25
-0
lines changed

1 file changed

+25
-0
lines changed
 

‎kubernetes/README.md

+25
Original file line numberDiff line numberDiff line change
@@ -81,3 +81,28 @@ So, for the user with UID=999 you can do it in your pod definition [like that](h
8181
securityContext:
8282
runAsUser: 999
8383
```
84+
85+
## 4. 问题4 - 现象是:kuberntes节点上拉取镜像被hang住,等一段时间后,镜像又可以正常被拉取下来
86+
87+
* 问题描述
88+
如标题,场景是在一次性通过helm的方式部署了很多应用,因为业务服务依赖于TiDB,观察发现TiDB的服务在拉取镜像时消耗了很多时间(10mins+, 这个具体的环境有关系),但等了一段时间后
89+
TiDB又正常拉取到镜像并运行起来了。
90+
91+
* 问题分析
92+
93+
我们知道镜像的拉取动作属于kubelet组件的职责(不知道的哥哥们可以看下,一个pod是如何在kubernetes上运行起来的详细介绍类的文章),那我们就先去看下对应节点Kubelet的日志,
94+
95+
* 问题原因
96+
默认情况下`serializeImagePulls=true`,
97+
98+
* 解决方案
99+
100+
修改`serializeImagePulls=false`,
101+
102+
* 更多
103+
104+
```sh
105+
kubectl get --raw "/api/v1/nodes/<nodename>/proxy/configz" | jq
106+
```
107+
Just make sure you replace`<nodename>` with your node name. And if you don't have `jq` installed, leave out the `| jq` part as that's only for formatting.
108+

0 commit comments

Comments
 (0)
Please sign in to comment.