PR - Ticket 49576 - Update ds-replcheck for new conflict entries

[389-ds-bot]

Cloned from Pagure Pull-Request: https://pagure.io/389-ds-base/pull-request/49700

• Created at 2018-05-18 13:31:22 by mreynolds (@mreynolds389)
• Merged at 2018-05-18 18:32:01

---

Description: This patch addresses the recvent changes to conflict
entries and tombstones.

Resolves: #2635

Reviewed by: ?

[389-ds-bot]

Comment from tbordaz (@tbordaz) at 2018-05-18 17:07:50

';deleted' can match a deleted value but a present attribute. for 'removed_attr' I would recommend to test ''deletedattribute"

[389-ds-bot]

Comment from mreynolds (@mreynolds389) at 2018-05-18 17:13:01

Are you suggesting replacing ";deleted" with ";deletedattribute"? I think I had to do this to get the "correct" results. If I am understanding you correctly, ";deleted" is not sufficient, a present value could still have this state info (;deleted in it)? If that's the case should the condition test be something like: ";deleted" and not "present"?

[389-ds-bot]

Comment from tbordaz (@tbordaz) at 2018-05-18 17:18:00

shouldn't ignore_list also contains modifiersname.
If a replicated update triggers internal modify (mep, memberof,...) the modifersname will differ

[389-ds-bot]

Comment from tbordaz (@tbordaz) at 2018-05-18 17:27:33

If the entry is a glue entry, why setting the flag subentry as well. IIRC a glue entry is not systematically a subentry

[389-ds-bot]

Comment from tbordaz (@tbordaz) at 2018-05-18 17:31:57

typo contlfiocty

[389-ds-bot]

Comment from mreynolds (@mreynolds389) at 2018-05-18 17:33:15

It's treating it as a subentry for the purpose of the report. Its used down on line 377 in the script to meet a special condition.

[389-ds-bot]

Comment from tbordaz (@tbordaz) at 2018-05-18 17:33:50

Note sure.
Could it be replaced with (rresult['entry'] or rresult['glue'] or
rresult['conflict'] or rresult['tombstone']) ?

[389-ds-bot]

Comment from tbordaz (@tbordaz) at 2018-05-18 17:46:07

Okay. That is fine thanks

[389-ds-bot]

Comment from tbordaz (@tbordaz) at 2018-05-18 17:55:23

Yes a present attribute can have '';deleted" values. So the attribute exists and has at least more than one value, but its stateinfo can contain deleted values ';deleted'

If you want to test deleted attribute in that function, it should be tested with ';deletedattribute'

[389-ds-bot]

Comment from mreynolds (@mreynolds389) at 2018-05-18 17:58:40

Its a python thing. You are only supposed to use that format if the value is True or False, not if it's None or "something"

[389-ds-bot]

Comment from mreynolds (@mreynolds389) at 2018-05-18 18:31:32

rebased onto 53e58cd

[389-ds-bot]

Comment from mreynolds (@mreynolds389) at 2018-05-18 18:32:02

Pull-Request has been merged by mreynolds389

[389-ds-bot]

Patch
49700.patch