From 1ecc62802ddf0a2fedbe4ab66ac121bcce446d2f Mon Sep 17 00:00:00 2001
From: Guangming Lu <71873108+LuGuangming@users.noreply.github.com>
Date: Tue, 26 Dec 2023 21:59:47 +0800
Subject: [PATCH] [Bug](security) BE download_files function exists log print
 sensitive msg #28592 (#28594)

---
 be/src/olap/task/engine_clone_task.cpp | 15 +++++++++++----
 be/src/olap/task/engine_clone_task.h   |  2 ++
 2 files changed, 13 insertions(+), 4 deletions(-)

diff --git a/be/src/olap/task/engine_clone_task.cpp b/be/src/olap/task/engine_clone_task.cpp
index 581cd3515c204a1..068e87233547d1b 100644
--- a/be/src/olap/task/engine_clone_task.cpp
+++ b/be/src/olap/task/engine_clone_task.cpp
@@ -31,6 +31,7 @@
 #include <memory>
 #include <mutex>
 #include <ostream>
+#include <regex>
 #include <set>
 #include <shared_mutex>
 #include <system_error>
@@ -390,7 +391,7 @@ Status EngineCloneTask::_make_and_download_snapshots(DataDir& data_dir,
         status = _download_files(&data_dir, remote_url_prefix, local_data_path);
         if (!status.ok()) [[unlikely]] {
             LOG_WARNING("failed to download snapshot from remote BE")
-                    .tag("url", remote_url_prefix)
+                    .tag("url", _mask_token(remote_url_prefix))
                     .error(status);
             continue; // Try another BE
         }
@@ -528,11 +529,11 @@ Status EngineCloneTask::_download_files(DataDir* data_dir, const std::string& re
 
         std::string local_file_path = local_path + "/" + file_name;
 
-        LOG(INFO) << "clone begin to download file from: " << remote_file_url
+        LOG(INFO) << "clone begin to download file from: " << _mask_token(remote_file_url)
                   << " to: " << local_file_path << ". size(B): " << file_size
                   << ", timeout(s): " << estimate_timeout;
 
-        auto download_cb = [&remote_file_url, estimate_timeout, &local_file_path,
+        auto download_cb = [this, &remote_file_url, estimate_timeout, &local_file_path,
                             file_size](HttpClient* client) {
             RETURN_IF_ERROR(client->init(remote_file_url));
             client->set_timeout_ms(estimate_timeout * 1000);
@@ -548,7 +549,8 @@ Status EngineCloneTask::_download_files(DataDir* data_dir, const std::string& re
             }
             if (local_file_size != file_size) {
                 LOG(WARNING) << "download file length error"
-                             << ", remote_path=" << remote_file_url << ", file_size=" << file_size
+                             << ", remote_path=" << _mask_token(remote_file_url)
+                             << ", file_size=" << file_size
                              << ", local_file_size=" << local_file_size;
                 return Status::InternalError("downloaded file size is not equal");
             }
@@ -829,4 +831,9 @@ Status EngineCloneTask::_finish_full_clone(Tablet* tablet,
     // TODO(plat1ko): write cooldown meta to remote if this replica is cooldown replica
 }
 
+std::string EngineCloneTask::_mask_token(const std::string& str) {
+    std::regex pattern("token=[\\w|-]+");
+    return regex_replace(str, pattern, "token=******");
+}
+
 } // namespace doris
diff --git a/be/src/olap/task/engine_clone_task.h b/be/src/olap/task/engine_clone_task.h
index 4f502ea50fd87f5..dd33f263e185077 100644
--- a/be/src/olap/task/engine_clone_task.h
+++ b/be/src/olap/task/engine_clone_task.h
@@ -86,6 +86,8 @@ class EngineCloneTask : public EngineTask {
 
     Status _release_snapshot(const std::string& ip, int port, const std::string& snapshot_path);
 
+    std::string _mask_token(const std::string& str);
+
 private:
     const TCloneReq& _clone_req;
     vector<TTabletInfo>* _tablet_infos = nullptr;