Files
rippled/.github
Pratik Mankawde 92e988c8b8 Merge branch 'pratik/otel-phase10-workload-validation' into pratik/otel-sync-diagnostics
Brings phase-10 up to c531ac569b: the workload trigger now keys on changed paths
rather than branch name, and eleven span hierarchies gained assertions.

One conflict, in parent_child_relationships, and it was an append-both: each
branch added entries to the same array, so the resolution is the union of the
two. Nothing was chosen over anything. Both sides' final object was unclosed
because the conflict boundary cut mid-entry, with the shared closing brace after
the marker -- the first attempt asserted the wrong shape and failed loudly rather
than producing malformed JSON, which is why the assertion was there.

The two sets are disjoint by design and by ownership. phase-10 added the eleven
whose spans it owns: rpc.http_request -> rpc.process, three txq parentings and
seven under consensus.round and consensus.establish. This branch added the three
ledger-acquire phase parentings, which could not go on phase-10 because
ledger.acquire.header and ledger.acquire.txtree do not exist there.

Net effect on this branch: 21 relationships declared, 17 of them asserted, up
from 8 declared and 5 asserted. The four still skipped are the two wildcard
rpc.command.* families, which the validator cannot match because it resolves a
wildcard child to a single literal probe, and pathfind.compute, whose child never
fires under this workload.

Verification: no conflict markers repo-wide, no unmerged entries, two parents;
JSON parses; 21 relationships with no duplicates and every non-wildcard child
resolving to a declared span entry; counters still 48 span types and 74 unique
attributes; otel-naming exits 0; the workflow YAML parses and no longer carries a
branches key; pre-commit clean. Two C++ files in this worktree carry another
party's uncommitted work and were deliberately left alone -- only
expected_spans.json was staged, and both remain modified after the commit.
2026-08-26 16:23:03 +01:00
..