[Chore](load) print error url to the log when encountering data quali…

…ty issues during load (apache#45212) There has been an issue where load does not print the error URL. Printing the error URL in the log is convenient for locating the problem.
CalvinKirs · Dec 11, 2024 · 02e8c8c · 02e8c8c
1 parent 1ee8d40
commit 02e8c8c
Show file tree

Hide file tree

Showing 4 changed files with 17 additions and 5 deletions.
diff --git a/be/src/http/action/stream_load.cpp b/be/src/http/action/stream_load.cpp
@@ -145,7 +145,8 @@ void StreamLoadAction::handle(HttpRequest* req) {
               << ctx->commit_and_publish_txn_cost_nanos / 1000000
               << ", number_total_rows=" << ctx->number_total_rows
               << ", number_loaded_rows=" << ctx->number_loaded_rows
-              << ", receive_bytes=" << ctx->receive_bytes << ", loaded_bytes=" << ctx->loaded_bytes;
+              << ", receive_bytes=" << ctx->receive_bytes << ", loaded_bytes=" << ctx->loaded_bytes
+              << ", error_url=" << ctx->error_url;
 
     // update statistics
     streaming_load_requests_total->increment(1);

diff --git a/be/src/runtime/fragment_mgr.cpp b/be/src/runtime/fragment_mgr.cpp
@@ -390,12 +390,20 @@ void FragmentMgr::coordinator_callback(const ReportStatusRequest& req) {
     params.load_counters.emplace(s_unselected_rows, std::to_string(num_rows_load_unselected));
 
     if (!req.runtime_state->get_error_log_file_path().empty()) {
-        params.__set_tracking_url(
-                to_load_error_http_path(req.runtime_state->get_error_log_file_path()));
+        std::string error_log_url =
+                to_load_error_http_path(req.runtime_state->get_error_log_file_path());
+        LOG(INFO) << "error log file path: " << error_log_url
+                  << ", query id: " << print_id(req.query_id)
+                  << ", fragment instance id: " << print_id(req.fragment_instance_id);
+        params.__set_tracking_url(error_log_url);
     } else if (!req.runtime_states.empty()) {
         for (auto* rs : req.runtime_states) {
             if (!rs->get_error_log_file_path().empty()) {
-                params.__set_tracking_url(to_load_error_http_path(rs->get_error_log_file_path()));
+                std::string error_log_url = to_load_error_http_path(rs->get_error_log_file_path());
+                LOG(INFO) << "error log file path: " << error_log_url
+                          << ", query id: " << print_id(req.query_id)
+                          << ", fragment instance id: " << print_id(rs->fragment_instance_id());
+                params.__set_tracking_url(error_log_url);
             }
             if (rs->wal_id() > 0) {
                 params.__set_txn_id(rs->wal_id());

diff --git a/be/src/runtime/runtime_state.cpp b/be/src/runtime/runtime_state.cpp
@@ -341,7 +341,9 @@ Status RuntimeState::create_error_log_file() {
         LOG(WARNING) << error_msg.str();
         return Status::InternalError(error_msg.str());
     }
-    VLOG_FILE << "create error log file: " << _error_log_file_path;
+    LOG(INFO) << "create error log file: " << _error_log_file_path
+              << ", query id: " << print_id(_query_id)
+              << ", fragment instance id: " << print_id(_fragment_instance_id);
 
     return Status::OK();
 }

diff --git a/fe/fe-core/src/main/java/org/apache/doris/qe/Coordinator.java b/fe/fe-core/src/main/java/org/apache/doris/qe/Coordinator.java
@@ -2409,6 +2409,7 @@ public void updateFragmentExecStatus(TReportExecStatusParams params) {
             updateLoadCounters(params.getLoadCounters());
         }
         if (params.isSetTrackingUrl()) {
+            LOG.info("query_id={} tracking_url: {}", DebugUtil.printId(queryId), params.getTrackingUrl());
             trackingUrl = params.getTrackingUrl();
         }
         if (params.isSetTxnId()) {