[hadoop] branch branch-3.2 updated: YARN-8587. Added retries for fetching docker exit code. Contributed by Charo Zhang

[ey...@apache.org]

This is an automated email from the ASF dual-hosted git repository.

eyang pushed a commit to branch branch-3.2
in repository https://gitbox.apache.org/repos/asf/hadoop.git


The following commit(s) were added to refs/heads/branch-3.2 by this push:
     new ac85aa8  YARN-8587. Added retries for fetching docker exit code.            Contributed by Charo Zhang
ac85aa8 is described below

commit ac85aa80d973d56452bc678637b5868ab43bee4f
Author: Eric Yang <ey...@apache.org>
AuthorDate: Wed Oct 24 17:28:23 2018 -0400

    YARN-8587. Added retries for fetching docker exit code.
               Contributed by Charo Zhang
    
    (cherry picked from commit c16c49b8c3b8e2e42c00e79a50e7ae029ebe98e2)
---
 .../container-executor/impl/container-executor.c   | 32 ++++++++++++----------
 1 file changed, 17 insertions(+), 15 deletions(-)

diff --git a/hadoop-yarn-project/hadoop-yarn/hadoop-yarn-server/hadoop-yarn-server-nodemanager/src/main/native/container-executor/impl/container-executor.c b/hadoop-yarn-project/hadoop-yarn/hadoop-yarn-server/hadoop-yarn-server-nodemanager/src/main/native/container-executor/impl/container-executor.c
index 7765308..7e86e88 100644
--- a/hadoop-yarn-project/hadoop-yarn/hadoop-yarn-server/hadoop-yarn-server-nodemanager/src/main/native/container-executor/impl/container-executor.c
+++ b/hadoop-yarn-project/hadoop-yarn/hadoop-yarn-server/hadoop-yarn-server-nodemanager/src/main/native/container-executor/impl/container-executor.c
@@ -1765,20 +1765,22 @@ int launch_docker_container_as_user(const char * user, const char *app_id,
   docker_binary, container_id);
   fprintf(LOGFILE, "Obtaining the exit code...\n");
   fprintf(LOGFILE, "Docker inspect command: %s\n", docker_inspect_exitcode_command);
-  FILE* inspect_exitcode_docker = popen(docker_inspect_exitcode_command, "r");
-  if(inspect_exitcode_docker == NULL) {
-    fprintf(ERRORFILE, "Done with inspect_exitcode, inspect_exitcode_docker is null\n");
-    fflush(ERRORFILE);
-    exit_code = -1;
-    goto cleanup;
-  }
-  res = fscanf (inspect_exitcode_docker, "%d", &exit_code);
-  if (pclose (inspect_exitcode_docker) != 0 || res <= 0) {
-  fprintf (ERRORFILE,
-   "Could not inspect docker to get exitcode:  %s.\n", docker_inspect_exitcode_command);
-    fflush(ERRORFILE);
-    exit_code = -1;
-    goto cleanup;
+  int count = 0;
+  int max_retries = get_max_retries(&CFG);
+  while (count < max_retries) {
+    FILE* inspect_exitcode_docker = popen(docker_inspect_exitcode_command, "r");
+    res = fscanf (inspect_exitcode_docker, "%d", &exit_code);
+    if (pclose (inspect_exitcode_docker) != 0 || res <= 0) {
+      fprintf (ERRORFILE, "Could not inspect docker to get Exit code %s.\n", docker_inspect_exitcode_command);
+      fflush(ERRORFILE);
+      exit_code = -1;
+    } else {
+      if (exit_code != 0) {
+        break;
+      }
+    }
+    sleep(3);
+    count++;
   }
   fprintf(LOGFILE, "Exit code from docker inspect: %d\n", exit_code);
 
@@ -2708,4 +2710,4 @@ int remove_docker_container(char**argv, int argc) {
     exit_code = clean_docker_cgroups(yarn_hierarchy, container_id);
   }
   return exit_code;
-}
\ No newline at end of file
+}


---------------------------------------------------------------------
To unsubscribe, e-mail: common-commits-unsubscribe@hadoop.apache.org
For additional commands, e-mail: common-commits-help@hadoop.apache.org