github-actions[bot] commented on code in PR #63771:
URL: https://github.com/apache/doris/pull/63771#discussion_r3345651868
##########
be/src/service/http/action/compaction_action.cpp:
##########
@@ -141,15 +140,6 @@ Status
CompactionAction::_handle_run_compaction(HttpRequest* req, std::string* j
return Status::NotSupported("The compaction type '{}' is not
supported", compaction_type);
}
- // "remote" = "true" means tablet should do single replica compaction to
fetch rowset from peer
- bool fetch_from_remote = false;
- std::string param_remote = req->param(PARAM_COMPACTION_REMOTE);
- if (param_remote == "true") {
- fetch_from_remote = true;
- } else if (!param_remote.empty() && param_remote != "false") {
- return Status::NotSupported("The remote = '{}' is not supported",
param_remote);
- }
-
// "force" = "true" means skip permit limiter when submitting full
compaction to thread pool
bool force = false;
std::string param_force = req->param(PARAM_COMPACTION_FORCE);
Review Comment:
After removing `SingleReplicaCompaction`, this endpoint no longer parses
`remote`, so existing calls like
`/api/compaction/run?tablet_id=...&compact_type=cumulative&remote=true` now
fall through and run a local cumulative compaction while returning success.
Before this PR, `remote=true` selected the peer-fetch compaction path and
invalid values were rejected. Since that behavior no longer exists, please
explicitly reject any non-empty `remote` parameter (or at least `remote=true`)
and keep a negative test; otherwise operators/automation can believe they
triggered a remote/single-replica repair when the BE actually did a different
local compaction.
--
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.
To unsubscribe, e-mail: [email protected]
For queries about this service, please contact Infrastructure at:
[email protected]
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
