Skip to content

Conversation

@avijayanhwx
Copy link
Contributor

What changes were proposed in this pull request?

  • Refactor upgrade finalizer classes.
  • Allow prepare command to succeed even if majority of OMs are prepared.
  • Other minor fixes

What is the link to the Apache JIRA

https://issues.apache.org/jira/browse/HDDS-5108

How was this patch tested?

Manually tested.
Added unit tests.

@avijayanhwx
Copy link
Contributor Author

cc @errose28 for review.

@fapifta
Copy link
Contributor

fapifta commented Apr 20, 2021

Hi @avijayanhwx,

thank you for taking on this, I think it is a great improvement in the code and I like the approach. I did a quick scroll though only so far, would like to take a deeper look, but if I don't get back to you in time, and others do not see anything that should be addressed, feel free to push it and don't wait for me, as I can not promise to get to it in reasonable time.

Copy link
Contributor

@errose28 errose28 left a comment

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Thanks for working on this @avijayanhwx, I like that more upgrade code has been moved out of each component class, and the reduced code duplication for finalizing each component. Just a few inline comments/questions.

@avijayanhwx
Copy link
Contributor Author

Thank you for the review @errose28. I have addressed your comments.

@errose28
Copy link
Contributor

Thanks for the updates. LGTM +1.

Copy link
Contributor

@fapifta fapifta left a comment

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Hi @avijayanhwx,

thank you for your patience, I went through the changes, and found a few minor things, I am sharing inline.

Looking at the bigger picture, the refactoring seems to be successful.

One thing I would like to note, I am not sure if you have any plans with UpgradeActions, and UpgradeActionType, and how it is defined and used, but that might be an other are for improvements with a similar approach, but this is certainly out of scope here to do.

@avijayanhwx
Copy link
Contributor Author

Thanks for the review @fapifta. I have addressed your comments.

@fapifta
Copy link
Contributor

fapifta commented Apr 22, 2021

Hi @avijayanhwx,

thank you for addressing all the things, I think the changes are good.
I just saw one last missing NL at the end of DefaultUpgradeFinalizationExecutor, if you would like to fix it, with or without that, I think we are good to go.

fapifta
fapifta approved these changes Apr 22, 2021
@avijayanhwx
Copy link
Contributor Author

Merging without CI since last commit just added newline.

Thanks for keeping me honest @errose28 & @fapifta!

@avijayanhwx avijayanhwx merged commit 2823320 into apache:HDDS-3698-nonrolling-upgrade Apr 22, 2021
errose28 added a commit to errose28/ozone that referenced this pull request Apr 22, 2021
* HDDS-3698-nonrolling-upgrade:
  HDDS-5108. Attempt to remove state from *UpgradeFinalizer classes. (apache#2160)
  Update upgrade-dev-primer.md
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

3 participants