[ 
https://issues.apache.org/jira/browse/RANGER-4038?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=17911892#comment-17911892
 ] 

AlexVazquez edited comment on RANGER-4038 at 1/13/25 8:44 AM:
--------------------------------------------------------------

Hi [~bpatel] ,

I'm glad to hear you consider this work much-needed, so thank you.

Yes, it's fully operational, and all services are working. I would say it's in 
the testing phase. I’ve tried creating, editing, and deleting entities from the 
interface, reviewed logs for all services, and not much-more. Since my 
knowledge of Ranger's application and deployment is somewhat limited, it would 
be great if someone else could also test it.

Regarding your thoughts:
 # The JDK17 PR is clear to me, and some of the latest commits could be split 
into separate PRs. However, there will still be a huge PR because the 
spring/jakarta change is very deep and forces the rest of the libraries to be 
changed. II don’t see how to make several small PRs without leaving the branch 
and/or the application inconsistent.
 # As I mentioned, using a Maven plugin would likely be a better solution. 
However, from my understanding, implementing it would require adding the plugin 
to the Hadoop project and using a classifier. I opted for my current approach 
because it’s simpler and independent of the Hadoop project.

The process I followed is straightforward—though I’m unsure if it’s the best 
solution—but it works:
 # {{{}download-archives.sh{}}}:
 ** Performs its original tasks.
 ** Downloads the two libraries (hadoop-common, hadoop-auth).
 ** Runs the migration tool.
 ** Installs both libraries into the local Maven repository.
 # The build process incorporates them into the WAR/JAR files, and the 
{{ranger-distro}} process packages the tar.gz with the shaded libraries.

What do you think?


was (Author: JIRAUSER307575):
Hi [~bpatel] ,

I'm glad to hear you consider this work much-needed, so thank you.

Yes, it's fully operational, and all services are working. I would say it's in 
the testing phase. I’ve tried creating, editing, and deleting entities from the 
interface, reviewed logs for all services, and not much-more. Since my 
knowledge of Ranger's application and deployment is somewhat limited, it would 
be great if someone else could also test it.

Regarding your thoughts:
 # The JDK17 PR is clear to me, and some of the latest commits could be split 
into separate PRs. However, there will still be a huge PR because the 
spring/jakarta change is very deep and forces the rest of the libraries to be 
changed. II don’t see how to make several small PRs without leaving the branch 
and/or the application inconsistent.

 # As I mentioned, using a Maven plugin would likely be a better solution. 
However, from my understanding, implementing it would require adding the plugin 
to the Hadoop project and using a classifier. I opted for my current approach 
because it’s simpler and independent of the Hadoop project.

The process I followed is straightforward—though I’m unsure if it’s the best 
solution—but it works:
 # {{{}download-archives.sh{}}}:

 ** Performs its original tasks.
 ** Downloads the two libraries (hadoop-common, hadoop-auth).
 ** Runs the migration tool.
 ** Installs both libraries into the local Maven repository.
 # The build process incorporates them into the WAR/JAR files, and the 
{{ranger-distro}} process packages the tar.gz with the shaded libraries.

What do you think?

> Upgrade spring framework and spring security versions
> -----------------------------------------------------
>
>                 Key: RANGER-4038
>                 URL: https://issues.apache.org/jira/browse/RANGER-4038
>             Project: Ranger
>          Issue Type: Bug
>          Components: Ranger
>            Reporter: Himanshu Maurya
>            Priority: Major
>
> Pivotal Spring Framework up to (excluding) 6.0.0 suffers from a potential 
> remote code execution (RCE) issue if used for Java deserialization of 
> untrusted data. Depending on how the library is implemented within a product, 
> this issue may or not occur, and authentication may be required.



--
This message was sent by Atlassian Jira
(v8.20.10#820010)

Reply via email to