PXC-5267: Code refresh for - PXC 8.4.11-11 - #29
Draft
jaideepkarande wants to merge 5 commits into
Draft
Conversation
…er lockup Issue: When an applier commit fails (e.g. commit order could not be entered because the node is leaving the primary component), the caller skipped after_apply() and left the transaction active. It is then carried to client_state::close(), which runs the local-only after_statement() and asserts client_state_.mode() == m_local in debug builds. Solution: On commit failure roll the applier transaction back and always call after_apply() so it is terminated on the error path, mirroring the apply-error branch.
The .gitmodules file uses the old Codership Github repo link. This redirects to the mariadb-corporation repo, so it isn't a functional problem, but it should still be updated to the correct repo.
Noticed in failed to build on clang-24.
…1a00) (conflicts solved)
Merge remote-tracking branch 'maria_galera/master' into PXC_5267
Conflicts:
.gitmodules
….4.11-11 wsrep-API/v26: 7fe08ed333419c353438ed0d1068b9ea72fa5d50 -> f80a23b6642c2d7d4d2e32ad4ba9d6fea4bdfeba
This was referenced Aug 20, 2026
kamil-holubicki
approved these changes
Aug 28, 2026
kamil-holubicki
approved these changes
Aug 28, 2026
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
5 participants
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
wsrep-lib refresh for the PXC 8.4.11-11 release (PXC-5267).
What this merges
mariadb-corporation/wsrep-libmaster@cf331a00— three commits:e55d2f2c0a3cf331a00c74std::back_inserterneeds iterator include header — build fixa54893ae9ecThe one manual conflict — and what was deliberately not taken
a54893ae9ecrewrites.gitmodulesto point wsrep-API atmariadb-corporation. That conflicted, and the resolution was keep ours:PXC consumes Percona's wsrep-API fork, not MariaDB's, so adopting upstream's URL would have silently repointed the submodule at a different project. This is the only hand-resolution in the PR, and it is the reason the merge commit is marked
(conflicts solved).Recording it here because a pointer or URL left unchanged is invisible in a diff — the next release will hit the same conflict, and this is the answer to why it resolves this way.
Submodule pointer
wsrep-API/v267fe08ed3334→f80a23b6642That carries
mariadb/master@cf9c0a2cinto wsrep-API. Its five commits are all upstreamv26up-merges carrying nothing of their own, so wsrep-API's tree is unchanged and the bump records provenance only.Merge order — this matters
The pointer above references
f80a23b6642, which currently exists only onjaideepkarande/wsrep-API. Until the wsrep-API PR merges, a clone ofpercona/wsrep-libat this branch cannotgit submodule update.Validation
wsrep-lib is not built standalone; it is exercised through the PXC release runs of
pxc-8.x-pipeline-parallel-mtr, which build this exact sha as a submodule:UNSTABLEis the normal outcome of that job; the gate is no new failures against the paired baseline. Worth noting for this PR specifically: the release's one crash — every Galera TOI apply aborting the applier — ran throughwsrep-lib/src/server_state.cppandwsrep_provider_v26.cppin its stack, but the defect was not in wsrep-lib. It was a PXC-side predicate (Relay_log_info::belongs_to_client()) reporting the wsrep applier as a client session, tripping an assertion new in PS 8.4.11-11. Fixed on the PXC branch; nothing here required a change. Full analysis on #2349.Related
Before this leaves draft
percona/wsrep-API.