ldb 1.1.13 now exposes some symbols from libreplace and internal symbols from ldb_map. This shouldn't be the case as its abi_match restricts to ldb_*. This is likely a regression from 9c3e294400234ebdf9b98031bae583524fd0b0ac
Should we backport that to v4-0-test ?
(In reply to comment #1) > Should we backport that to v4-0-test ? Only if we're going to be doing ldb releases from the v4-0-test branch.
(In reply to comment #2) > (In reply to comment #1) > > Should we backport that to v4-0-test ? > > Only if we're going to be doing ldb releases from the v4-0-test branch. I don't think this is specific to ldb, what about all the non-standalone libraries, which are part of the Samba release e.g. libwbclient, libsmbclient...
(In reply to comment #3) > (In reply to comment #2) > > (In reply to comment #1) > > > Should we backport that to v4-0-test ? > > > > Only if we're going to be doing ldb releases from the v4-0-test branch. > > I don't think this is specific to ldb, what about all the non-standalone > libraries, which are part of the Samba release e.g. libwbclient, > libsmbclient... It only occurs if you use inclusion and exclusion in abi_match. ldb is the only library that does that, last I checked. That said, I don't think it's a bad idea to backport these three commits. Feel free to do so.
Created attachment 8163 [details] Patches for v4-0-test
Is there a chance to get review until Sunday? I would like to add this one for rc5. Thanks!
(In reply to comment #6) > Is there a chance to get review until Sunday? > I would like to add this one for rc5. > > Thanks! Sorry, Metze just told me that Jelmer granted review on the bug (I was looking at the patch). Pushing to autobuild.
Pushed to v4-0-test. Closing out bug report. Thanks!