Apr 2023

5 commits
Further cleanup of autoconf output files for GSSAPI changes.
17 Apr 2023Tom Lanemaster16beta1REL_19_STABLE16beta1REL_18_STABLE16beta1REL_17_STABLE16beta1REL_16_STABLE16beta1
LLM analysis, 95% surewritten from the discussion, no matching patch1 of 1 reviewer saw the final version

The commit applies Tom Lane's inline autoheader output from #47, adding two HAVE_GSSAPI extension-header declarations to src/include/pg_config.h.in. Although identical generated declarations appeared in #8, the explicit landing provenance is this newly identified omission and inline fix, not the unrelated configure attachment in #46.

  • Stephen Frostsaw the final versionLLM analysis, 87% sureStephen acknowledged the missing autoheader update and explicitly approved Tom pushing the inline fix in #48.
Update Kerberos/GSSAPI configure/meson check
17 Apr 2023Stephen Frostmaster16beta1REL_19_STABLE16beta1REL_18_STABLE16beta1REL_17_STABLE16beta1REL_16_STABLE16beta1
LLM analysis, 100% surelanded as postedfrom patchset #411 of 1 reviewer saw the final version

Both configure.ac and meson.build diffs match #41 exactly apart from diff metadata, replacing the function probe and its error text with gss_store_cred_into.

  • Tom Lanesaw the final versionLLM analysis, 100% sureTom reported the unsupported-library failure in #40 and accepted the attached function-check fix in #42.
Explicitly require MIT Kerberos for GSSAPI
13 Apr 2023Stephen Frostmaster16beta1REL_19_STABLE16beta1REL_18_STABLE16beta1REL_17_STABLE16beta1REL_16_STABLE16beta1
LLM analysis, 100% surepart of a posted patchsetlight changesfrom patchset #363 of 4 reviewers saw the final version

Exactly matches the second component of #36, including configure, configure.ac, meson.build, MIT Kerberos documentation and the two backend comment updates. It is partial only because the delegation restoration was committed separately.

  • Justin Pryzbysaw an earlier versionLLM analysis, 98% sureJustin identified the missing Meson equivalent in #25, after which a new header-detection conditional was added and corrected.
  • Jonathan S. Katzsaw the final versionLLM analysis, 99% sureJonathan requested precise MIT Kerberos wording in #30 and approved the revised documentation and build checks in #33, matching the misspelled reviewer credit in the commit message.
  • Daniel Gustafssonsaw the final versionLLM analysis, 96% sureDaniel identified the stale Heimdal comment in #31, prompting the two comment updates included in #36.
  • Tom Lanesaw the final versionLLM analysis, 95% sureTom requested the reviewable restoration-plus-delta split in #28 and approved the resulting series in #35, with only documentation, comment and commit-message polishing afterward.
De-Revert "Add support for Kerberos credential delegation"
13 Apr 2023Stephen Frostmaster16beta1REL_19_STABLE16beta1REL_18_STABLE16beta1REL_17_STABLE16beta1REL_16_STABLE16beta1
LLM analysis, 100% surepart of a posted patchsetlight changesfrom patchset #362 of 2 reviewers saw the final version

Matches the first component of #36 across all 36 files, with only a shifted hunk location in doc/src/sgml/monitoring.sgml. The separate MIT Kerberos requirements component was committed independently.

  • Tom Lanesaw the final versionLLM analysis, 98% sureTom proposed restoring delegation while dropping Heimdal support in #18, requested a straight restoration for review in #28, and approved the series in #35.
  • Thomas Munrosaw the final versionLLM analysis, 85% sureThomas verified and reported in #23 that FreeBSD CI had actually exercised the Kerberos tests with MIT Kerberos on the original implementation subsequently restored by #36.
Revert "Add support for Kerberos credential delegation"
8 Apr 2023Stephen Frostmaster16beta1REL_19_STABLE16beta1REL_18_STABLE16beta1REL_17_STABLE16beta1REL_16_STABLE16beta1revert
LLM analysis, 100% surewritten from the discussion, no matching patch1 of 1 reviewer saw the final version

This removes delegation following Tom Lane's NetBSD portability objection in #9, rather than applying either attached configure fix. Stephen explicitly confirms the revert in #11; no attached revert patch exists in the thread.

  • Tom Lanesaw the final versionLLM analysis, 98% sureTom identified current NetBSD releases lacking the required header and objected to imposing the requirement near feature freeze in #9, directly motivating the discussion-only revert.

Credits across 5 commits

Author
Stephen FrostLLM3
Committer
Stephen Frost4
Tom Lane1

N LLM means N of the credits come from LLM analysis, not from commit trailers.