fix: retry 12 seconds if lock file failed

source/dnode/mgmt/node_util/src/dmFile.c

@@ -127,14 +127,24 @@ TdFilePtr dmCheckRunning(const char *dataDir) {
     return NULL;
   }
 
-  int32_t ret = taosLockFile(pFile);
-  if (ret != 0) {
+  int32_t retryTimes = 0;
+  int32_t ret = 0;
+  do {
+    ret = taosLockFile(pFile);
+    if (ret == 0) break;
+    terrno = TAOS_SYSTEM_ERROR(errno);
+    taosMsleep(100);
+    retryTimes++;
+    dError("failed to lock file:%s since %s, retryTimes:%d", filepath, terrstr(), retryTimes);
+  } while (retryTimes < 120);
+
+  if (ret < 0) {
     terrno = TAOS_SYSTEM_ERROR(errno);
-    dError("failed to lock file:%s since %s", filepath, terrstr());
     taosCloseFile(&pFile);
     return NULL;
   }
 
+  terrno = 0;
   dDebug("lock file:%s to prevent repeated starts", filepath);
   return pFile;
 }

tests/script/tsim/mnode/basic5.sim

@@ -233,7 +233,6 @@ if $data(1)[3] != dropping then
 endi
 
 print =============== step9: start mnode1 and wait it dropped
-sleep 3000
 system sh/exec.sh -n dnode1 -s start
 
 $x = 0