cczz12345 opened a new issue #5431:
URL: https://github.com/apache/dolphinscheduler/issues/5431
**For better global communication, Please describe it in English. If you
feel the description in English is not clear, then you can append description
in Chinese(just for Mandarin(CN)), thx! **
**Describe the bug**
(I found a bug when I test v1.3.6, which did not happen in v1.3.5.I need to
used host network dns policy on k8s, so I add config in
templates/statefulset-dolphinscheduler-worker.yaml)
我在测试刚发布的v1.3.6的时候，发现一个关于worker的命名的bug，在之前的1.3.5中并未出现。
起因：因为spark的一些特性，我需要在k8s的worker上使用本地网络，因此我在templates/statefulset-dolphinscheduler-worker.yaml添加了：
hostNetwork: true
dnsPolicy: ClusterFirstWithHostNet
于是我发现1.3.6在worker中命名规则与1.3.5并不一致，可以看到1.3.5中：
(but I found there is a different behavior between 1.3.5 and 1.3.6, we can
see in v1.3.5:)
而在1.3.6中:
(but in v1.3.6:)
可以看出1.3.6里worker的hostname把后面的.local截掉了。导致工作流运作起来的时候，master找不到对应的worker主机。
(we can see v1.3.6 the hostname of workers is cut, which makes the workflow
doesn't work. because the Master can't find the correct worker dns address.)
**Expected behavior**
roll back to v1.3.5，make full hostname of the worker。
**Screenshots**
If applicable, add screenshots to help explain your problem.
**Which version of Dolphin Scheduler:**
-[v1.3.6]
**Additional context**
Add any other context about the problem here.
**Requirement or improvement**
- Please describe about your requirements or improvement suggestions.
--
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.
For queries about this service, please contact Infrastructure at:
[email protected]
