This patch completely removes legacy replication:

Mate this is amazing! I'm all for code cleanup's like this.

Only concerns I have:

• I don't think we can remove the attrs from schema. Can we update their descriptions instead to say "unused"?
• will this affects @lkrispen 's replication patches he has in flight? I don't expect so, but probably good to check.

Anyway, you have my ack otherwise :)

Mate this is amazing! I'm all for code cleanup's like this.
Only concerns I have:

I don't think we can remove the attrs from schema. Can we update their descriptions instead to say "unused"?

I was on the fence about this, but since we remove the plugin itself I thought it was okay. I am fine leaving it though.

will this affects @lkrispen 's replication patches he has in flight? I don't expect so, but probably good to check.

It should not impact anything, and the CI tests do pass, but I also would like to hear from @lkrispen on this. :-)

Hi, you did the work again I already had done in #48545 :-)
Interestingly enough, I had picked up my patch yesterday to prepare it for review. I will compare both patches and comment.

Just a few comments about this great/big patch @mreynolds.

• about 'cn=Legacy Replication Plugin,cn=plugins,cn=config' entry, the update script is deleting this entry. In case the entry does not exist, is it going to break update ?
• I think REPL_PROTOCOL_40 is no longer used
• Same for OP_FLAG_LEGACY_REPLICATION_DN that could be removed from slapi-private.h

Hi, you did the work again I already had done in #48545 :-)

This is exactly why we all need to look through the tickets, and triage them. I can't do it all by myself (although I am trying), and I am sure there are more duplicate tickets out there.

Interestingly enough, I had picked up my patch yesterday to prepare it for review. I will compare both patches and comment.

Thanks Ludwig!

@mreynolds looks like your patch is more complete than mine, and we should commit it
but Thierry had some comments to be addresed

My repl patches will cope with it, I will rebase them and fix if needed

about the attributes, do you mean in the schema ? I think we can remove them, but izt doesn't hurt to keep them.

@mreynolds looks like your patch is more complete than mine, and we should commit it
but Thierry had some comments to be addresed

I'll update my patch and send it back our for review...

My repl patches will cope with it, I will rebase them and fix if needed
about the attributes, do you mean in the schema ? I think we can remove them, but izt doesn't hurt to keep them.

Lets leave them as is - less schema churn is better.

Okay, so a new patch will be coming out soon...

Thanks,
Mark

Just a few comments about this great/big patch @mreynolds.

about 'cn=Legacy Replication Plugin,cn=plugins,cn=config' entry, the update script is deleting this entry. In case the entry does not exist, is it going to break update ?

It works fine - I ran many offline and online upgrades back to back without issue

I think REPL_PROTOCOL_40 is no longer used
Same for OP_FLAG_LEGACY_REPLICATION_DN that could be removed from slapi-private.h

Thanks! This uncovered a lot of other code that needed to be removed!

New patch attached:

The patch is looking good to me. ACK

c8370b8..b95f5f2 master -> master

Found regression from legacy cleanup that only impacted 389-console

59d1556..cc87118 master -> master

The last patch reported error 68 at every startup. Instead of adding it during the init, just add it to the dse.ldif by default.

56ed382..16aeef6 master -> master

Fix cleanup script precedence

0ea523c..1fe2c76 master -> master

87c118c..6d28c75 389-ds-base-1.3.7 -> 389-ds-base-1.3.7

389-ds-base is moving from Pagure to Github. This means that new issues and pull requests
will be accepted only in 389-ds-base's github repository.

This issue has been cloned to Github and is available here:
- https://github.com/389ds/389-ds-base/issues/2097

If you want to receive further updates on the issue, please navigate to the github issue
and click on subscribe button.

Thank you for understanding. We apologize for all inconvenience.