Nov 2024

1 commit
Fix a possibility of logical replication slot's restart_lsn going backwards.
16 Nov 2024Masahiko Sawadamaster18beta1REL_19_STABLE18beta1REL_18_STABLE18beta1REL_17_STABLE17.2REL_16_STABLE16.6REL_15_STABLE15.10REL_14_STABLE14.15REL_13_STABLE13.18Also in: logical replication: restart_lsn can go backwards (and more), seems broken since 9.4 (landed as posted)
LLM analysis, 100% surewritten from the discussion, no matching patch1 of 3 reviewers saw the final version

e5ed873b1b changes the independent candidate-update condition to else-if and adds two spinlock releases in LogicalIncreaseRestartDecodingForSlot(), implementing the inline proposal in #47 reiterated in #54. Neither attachment contains this fix: #18 adds checkpoint diagnostics, while #50 clears candidate fields on release and supplies reproduction instrumentation.

  • Tomas Vondrano review foundtrailerTomas is credited by e5ed873b1b, but no review activity by him appears in the supplied thread.
  • Ashutosh Bapatno review foundtrailerAshutosh is credited by e5ed873b1b, but no review activity by him appears in the supplied thread.
  • Amit Kapilasaw the final versiontrailerAmit analyzed the restart sequence and proposed the committed else-if correction inline in #47, which Masahiko endorsed in #54; this contribution did not review an attached patchset.

Credits across 1 commit

Author
Masahiko SawadaLLM1
Reviewer
Amit Kapila1
Ashutosh Bapat1
Tomas Vondra1
Committer
Masahiko Sawada1

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