[PPML] Fix tracker bind issue happens in xgboost examples with SGX enabled

docs/readthedocs/source/doc/PPML/Overview/ppml.md

@@ -583,7 +583,7 @@ The result should look something like this:
 
 This example shows how to run trusted Spark XGBoost Regressor.
 
-First, make sure that `Boston_Housing.csv` is under `work/data` directory or the same path in the `start-spark-local-xgboost-regressor-sgx.sh`. Replace the value of `RABIT_TRACKER_IP` with your own IP address in the script.
+First, make sure that `Boston_Housing.csv` is under `work/data` directory or the same path in the `start-spark-local-xgboost-regressor-sgx.sh`.
 
 Run the script to run trusted Spark XGBoost Regressor and it would take some time to show the final results:
 
@@ -643,7 +643,7 @@ The result should look something like this:
 
 This example shows how to run trusted Spark XGBoost Classifier.
 
-Before running the example, download the sample dataset from [pima-indians-diabetes](https://raw.githubusercontent.com/jbrownlee/Datasets/master/pima-indians-diabetes.data.csv) dataset. After downloading the dataset, make sure that `pima-indians-diabetes.data.csv` is under `work/data` directory or the same path in the `start-spark-local-xgboost-classifier-sgx.sh`. Replace `path_of_pima_indians_diabetes_csv` with your path of `pima-indians-diabetes.data.csv`  and the value of `RABIT_TRACKER_IP` with your own IP address in the script.
+Before running the example, download the sample dataset from [pima-indians-diabetes](https://raw.githubusercontent.com/jbrownlee/Datasets/master/pima-indians-diabetes.data.csv) dataset. After downloading the dataset, make sure that `pima-indians-diabetes.data.csv` is under `work/data` directory or the same path in the `start-spark-local-xgboost-classifier-sgx.sh`. Replace `path_of_pima_indians_diabetes_csv` with your path of `pima-indians-diabetes.data.csv`.
 
 Run the script to run trusted Spark XGBoost Classifier and it would take some time to show the final results:
 

ppml/docs/examples.md

@@ -307,7 +307,7 @@ bash bigdl-ppml-submit.sh \
 <details><summary>This example shows how to run trusted Spark XGBoost Regressor.</summary>
 
 
-First, make sure that `Boston_Housing.csv` is under `work/data` directory or the same path in the `start-spark-local-xgboost-regressor-sgx.sh`. Replace the value of `RABIT_TRACKER_IP` with your own IP address in the script.
+First, make sure that `Boston_Housing.csv` is under `work/data` directory or the same path in the `start-spark-local-xgboost-regressor-sgx.sh`.
 
 Run the script to run trusted Spark XGBoost Regressor and it would take some time to show the final results:
 
@@ -368,7 +368,7 @@ The result should look something like this:
 <details><summary>This example shows how to run trusted Spark XGBoost Classifier.</summary>
 
 
-Before running the example, download the sample dataset from [pima-indians-diabetes](https://raw.githubusercontent.com/jbrownlee/Datasets/master/pima-indians-diabetes.data.csv) dataset. After downloading the dataset, make sure that `pima-indians-diabetes.data.csv` is under `work/data` directory or the same path in the `start-spark-local-xgboost-classifier-sgx.sh`. Replace `path_of_pima_indians_diabetes_csv` with your path of `pima-indians-diabetes.data.csv`  and the value of `RABIT_TRACKER_IP` with your own IP address in the script.
+Before running the example, download the sample dataset from [pima-indians-diabetes](https://raw.githubusercontent.com/jbrownlee/Datasets/master/pima-indians-diabetes.data.csv) dataset. After downloading the dataset, make sure that `pima-indians-diabetes.data.csv` is under `work/data` directory or the same path in the `start-spark-local-xgboost-classifier-sgx.sh`. Replace `path_of_pima_indians_diabetes_csv` with your path of `pima-indians-diabetes.data.csv`.
 
 Run the script to run trusted Spark XGBoost Classifier and it would take some time to show the final results:
 

ppml/trusted-big-data-ml/python/docker-gramine/tracker.py

@@ -455,16 +455,12 @@ def start_rabit_tracker(args):
 
 def main():
     """Main function if tracker is executed in standalone mode."""
-    host_ip = os.environ['RABIT_TRACKER_IP']
-    if host_ip == None:
-        sys.stdout.write("###PYTHONWARN### RABIT_TRACKER_IP not set in env")
-
     parser = argparse.ArgumentParser(description='Rabit Tracker start.')
     parser.add_argument('--num-workers', required=True, type=int,
                         help='Number of worker proccess to be launched.')
     parser.add_argument('--num-servers', default=0, type=int,
                         help='Number of server process to be launched. Only used in PS jobs.')
-    parser.add_argument('--host-ip', default=host_ip, type=str,
+    parser.add_argument('--host-ip', default=None, type=str,
                         help=('Host IP addressed, this is only needed ' +
                               'if the host IP cannot be automatically guessed.'))
     parser.add_argument('--log-level', default='INFO', type=str,
@@ -473,6 +469,10 @@ def main():
     args = parser.parse_args()
     sys.stdout.write("###PYTHONWARN### args for tracker: " + str(args))
 
+    # Open a file and prepare to set the hostname
+    with open("/etc/hostname", 'r') as f:
+        hostname = f.readline().strip()
+        socket.sethostname(hostname)
     fmt = '%(asctime)s %(levelname)s %(message)s'
     if args.log_level == 'INFO':
         level = logging.INFO

ppml/trusted-big-data-ml/python/docker-graphene/README.md

@@ -438,7 +438,7 @@ After changing:
 > 0.00632,18,2.31,**0**,0.538,6.575,65.2,4.09,1,296,15.3,396.9,4.98,24
 
 ```bash
-/graphene/Tools/argv_serializer bash -c "export RABIT_TRACKER_IP=your_IP_address && /opt/jdk8/bin/java -cp \
+/graphene/Tools/argv_serializer bash -c "/opt/jdk8/bin/java -cp \
     '/ppml/trusted-big-data-ml/work/bigdl-2.1.0-SNAPSHOT/jars/*:/ppml/trusted-big-data-ml/work/spark-3.1.2/conf/:/ppml/trusted-big-data-ml/work/spark-3.1.2/jars/*' \
   -Xmx2g \
   org.apache.spark.deploy.SparkSubmit \
@@ -514,7 +514,7 @@ wget https://raw.githubusercontent.com/jbrownlee/Datasets/master/pima-indians-di
 After downloading the dataset, make sure that `pima-indians-diabetes.data.csv` is under `work/data` directory or the same path in the command. Run the example with SGX spark local mode with the following command in the terminal. Replace `your_IP_address` with your IP address and `path_of_pima_indians_diabetes_csv` with your path of `pima-indians-diabetes.data.csv`.
 
 ```bash
-/graphene/Tools/argv_serializer bash -c "export RABIT_TRACKER_IP=your_IP_address && /opt/jdk8/bin/java -cp \
+/graphene/Tools/argv_serializer bash -c "/opt/jdk8/bin/java -cp \
   '/ppml/trusted-big-data-ml/work/bigdl-2.1.0-SNAPSHOT/jars/*:/ppml/trusted-big-data-ml/work/spark-3.1.2/conf/:/ppml/trusted-big-data-ml/work/spark-3.1.2/jars/*' \
   -Xmx2g \
   org.apache.spark.deploy.SparkSubmit \

ppml/trusted-big-data-ml/python/docker-graphene/start-scripts/start-spark-local-xgboost-classifier-sgx.sh

@@ -1,5 +1,5 @@
 #!/bin/bash
-SGX=1 ./pal_loader bash -c "export RABIT_TRACKER_IP=your_IP_address && /opt/jdk8/bin/java -cp \
+SGX=1 ./pal_loader bash -c "/opt/jdk8/bin/java -cp \
   '/ppml/trusted-big-data-ml/work/bigdl-2.1.0-SNAPSHOT/jars/*:/ppml/trusted-big-data-ml/work/spark-3.1.2/conf/:/ppml/trusted-big-data-ml/work/spark-3.1.2/jars/*' \
   -Xmx2g \
   org.apache.spark.deploy.SparkSubmit \

ppml/trusted-big-data-ml/python/docker-graphene/start-scripts/start-spark-local-xgboost-regressor-sgx.sh

@@ -1,5 +1,5 @@
 #!/bin/bash
-SGX=1 ./pal_loader bash -c "export RABIT_TRACKER_IP=your_IP_address && /opt/jdk8/bin/java -cp \
+SGX=1 ./pal_loader bash -c "/opt/jdk8/bin/java -cp \
     '/ppml/trusted-big-data-ml/work/bigdl-2.1.0-SNAPSHOT/jars/*:/ppml/trusted-big-data-ml/work/spark-3.1.2/conf/:/ppml/trusted-big-data-ml/work/spark-3.1.2/jars/*' \
   -Xmx2g \
   org.apache.spark.deploy.SparkSubmit \

ppml/trusted-big-data-ml/python/docker-graphene/tracker.py

@@ -455,16 +455,12 @@ def start_rabit_tracker(args):
 
 def main():
     """Main function if tracker is executed in standalone mode."""
-    host_ip = os.environ.get("RABIT_TRACKER_IP")
-    if host_ip == None:
-        sys.stdout.write("###PYTHONWARN### RABIT_TRACKER_IP not set in env")
-
     parser = argparse.ArgumentParser(description='Rabit Tracker start.')
     parser.add_argument('--num-workers', required=True, type=int,
                         help='Number of worker proccess to be launched.')
     parser.add_argument('--num-servers', default=0, type=int,
                         help='Number of server process to be launched. Only used in PS jobs.')
-    parser.add_argument('--host-ip', default=host_ip, type=str,
+    parser.add_argument('--host-ip', default=None, type=str,
                         help=('Host IP addressed, this is only needed ' +
                               'if the host IP cannot be automatically guessed.'))
     parser.add_argument('--log-level', default='INFO', type=str,
@@ -473,6 +469,10 @@ def main():
     args = parser.parse_args()
     sys.stdout.write("###PYTHONWARN### args for tracker: " + str(args))
 
+    # Open a file and prepare to set the hostname
+    with open("/etc/hostname", 'r') as f:
+        hostname = f.readline().strip()
+        socket.sethostname(hostname)
     fmt = '%(asctime)s %(levelname)s %(message)s'
     if args.log_level == 'INFO':
         level = logging.INFO

python/dllib/examples/nnframes/xgboost/xgboost_classifier.py

@@ -38,6 +38,7 @@
 
 def process(filepath, demo):
     sparkConf = init_spark_conf().setAppName("testXGBClassifier")
+    sparkConf = sparkConf.set("xgboost.spark.ignoreSsl", True)
     sc = init_nncontext(sparkConf)
     sqlContext = SQLContext(sc)
     if demo:

python/dllib/examples/nnframes/xgboost/xgboost_example.py

@@ -39,11 +39,12 @@
 
 def Processdata(filepath, demo):
     '''
-         preProcess the data read from filepath
+    preProcess the data read from filepath
     :param filepath:
     :return: assembledf:
     '''
     sparkConf = init_spark_conf().setAppName("testNNClassifer")
+    sparkConf = sparkConf.set("xgboost.spark.ignoreSsl", True)
     sc = init_nncontext(sparkConf)
     sqlContext = SQLContext(sc)
     if demo:

python/dllib/examples/nnframes/xgboost/xgboost_regressor.py

@@ -38,11 +38,12 @@
 
 def Processdata(filepath, demo):
     '''
-         preProcess the data read from filepath
+    preProcess the data read from filepath
     :param filepath:
     :return: assembledf:
     '''
     sparkConf = init_spark_conf().setAppName("testNNClassifer")
+    sparkConf = sparkConf.set("xgboost.spark.ignoreSsl", True)
     sc = init_nncontext(sparkConf)
     sqlContext = SQLContext(sc)
     if demo: