Added retry to deployer. The VM we use timesout all the time

drip-api/src/main/java/nl/uva/sne/drip/api/v1/rest/LogController.java

@@ -17,31 +17,20 @@ package nl.uva.sne.drip.api.v1.rest;
 
 import com.webcohesion.enunciate.metadata.rs.ResponseCode;
 import com.webcohesion.enunciate.metadata.rs.StatusCodes;
-import nl.uva.sne.drip.drip.commons.data.v1.external.ToscaRepresentation;
 import java.io.IOException;
-import java.util.ArrayList;
-import java.util.List;
 import java.util.concurrent.TimeoutException;
 import java.util.logging.Level;
-import java.util.logging.LogRecord;
 import java.util.logging.Logger;
 import javax.annotation.security.RolesAllowed;
-import org.json.JSONException;
 import org.springframework.beans.factory.annotation.Autowired;
 import org.springframework.stereotype.Component;
 import org.springframework.web.bind.annotation.RequestMapping;
 import org.springframework.web.bind.annotation.RequestMethod;
-import org.springframework.web.bind.annotation.RequestParam;
 import org.springframework.web.bind.annotation.ResponseBody;
 import org.springframework.web.bind.annotation.RestController;
-import org.springframework.web.multipart.MultipartFile;
-import org.springframework.web.bind.annotation.PathVariable;
-import nl.uva.sne.drip.api.exception.BadRequestException;
 import nl.uva.sne.drip.api.service.DRIPLogService;
-import nl.uva.sne.drip.api.service.ToscaService;
 import nl.uva.sne.drip.api.service.UserService;
 import nl.uva.sne.drip.drip.commons.data.v1.external.DRIPLogRecord;
-import org.springframework.web.bind.annotation.RequestBody;
 
 /**
  * This controller is responsible for storing TOSCA descriptions that can be
@@ -60,7 +49,6 @@ public class LogController {
     @Autowired
     private DRIPLogService logService;
 
-    @RequestMapping(method = RequestMethod.GET)
     @RolesAllowed({UserService.USER, UserService.ADMIN})
     public @ResponseBody
     DRIPLogRecord get() {

drip-api/src/main/java/nl/uva/sne/drip/api/v1/rest/MonitorringMessageController.java

@@ -25,7 +25,6 @@ import org.springframework.beans.factory.annotation.Autowired;
 import org.springframework.stereotype.Component;
 import org.springframework.web.bind.annotation.RequestMapping;
 import org.springframework.web.bind.annotation.RequestMethod;
-import org.springframework.web.bind.annotation.RequestParam;
 import org.springframework.web.bind.annotation.ResponseBody;
 import org.springframework.web.bind.annotation.RestController;
 import org.springframework.web.bind.annotation.PathVariable;

drip-deployer/docker_check.py

@@ -37,6 +37,9 @@ if not getattr(logger, 'handler_set', None):
     logger.handler_set = True
     
 
+retry=0    
+
+
 def get_resp_line(line):
     line = line.encode('utf-8').strip('\n').encode('string_escape')
     return json.dumps(line)
@@ -124,6 +127,7 @@ def docker_check(vm, compose_name):
         
         logger.info("Finished docker info services on: "+vm.ip)                
     except Exception as e:
+        global retry
         exc_type, exc_obj, tb = sys.exc_info()
         f = tb.tb_frame
         lineno = tb.tb_lineno

drip-deployer/docker_compose.py

@@ -33,6 +33,8 @@ if not getattr(logger, 'handler_set', None):
     logger.handler_set = True
 
 
+retry=0
+
 def deploy_compose(vm, compose_file, compose_name):
     try:
         logger.info("Starting docker compose deployment on: "+vm.ip)        
@@ -47,9 +49,14 @@ def deploy_compose(vm, compose_file, compose_name):
         stdout.read()
         logger.info("Finished docker compose deployment on: "+vm.ip)        
     except Exception as e:
+        global retry
+        if 'Connection timed out' in str(e) and retry < 3:
+            retry+=1
+            deploy_compose(vm)               
         logger.error(vm.ip + " " + str(e))
         return "ERROR:" + vm.ip + " " + str(e)
     ssh.close()
+    retry=0
     return "SUCCESS"
 
 

drip-deployer/docker_engine.py

@@ -32,6 +32,7 @@ if not getattr(logger, 'handler_set', None):
     logger.handler_set = True
         
 
+retry=0
 
 def install_engine(vm):
 	try:
@@ -54,9 +55,16 @@ def install_engine(vm):
 		stdout.read()
 		logger.info("Finised docker engine installation on: "+(vm.ip))
 	except Exception as e:
+                global retry
+                if 'Connection timed out' in str(e) and retry < 3:
+                    logger.warning(vm.ip + " " + str(e)+". Retiring")
+                    retry+=1
+                    install_engine(vm)
+                
 		logger.error(vm.ip + " " + str(e))
 		return "ERROR:"+vm.ip+" "+str(e)
 	ssh.close()
+	retry=0
 	return "SUCCESS"
 
 def run(vm_list,rabbitmq_host,owner):

drip-deployer/docker_kubernetes.py

@@ -20,6 +20,21 @@ import paramiko, os
 from vm_info import VmInfo
 import linecache
 import sys
+import logging
+from drip_logging.drip_logging_handler import *
+
+
+logger = logging.getLogger(__name__)
+if not getattr(logger, 'handler_set', None):
+    logger.setLevel(logging.INFO)
+    h = logging.StreamHandler()
+    formatter = logging.Formatter('%(asctime)s - %(name)s - %(levelname)s - %(message)s')
+    h.setFormatter(formatter)
+    logger.addHandler(h)
+    logger.handler_set = True
+    
+
+retry=0    
 
 def PrintException():
     exc_type, exc_obj, tb = sys.exc_info()
@@ -34,7 +49,7 @@ def PrintException():
 
 def install_manager(vm):
 	try:
-		print "%s: ====== Start Kubernetes Master Installing ======" % (vm.ip)
+		logger.info("Starting kubernetes master installation on: "+(vm.ip))
                 parentDir = os.path.dirname(os.path.abspath(vm.key))
                 os.chmod(parentDir, 0o700)
                 os.chmod(vm.key, 0o600)
@@ -65,9 +80,11 @@ def install_manager(vm):
 		stdin, stdout, stderr = ssh.exec_command("sudo chgrp %s /tmp/admin.conf" % (vm.user))
 		stdout.read()
 		sftp.get("/tmp/admin.conf", file_path+"/admin.conf")
-		print "%s: ========= Kubernetes Master Installed =========" % (vm.ip)
+		logger.info("Finished kubernetes master installation on: "+(vm.ip))
 	except Exception as e:
-		print '%s: %s' % (vm.ip, e)
+                global retry
+		#print '%s: %s' % (vm.ip, e)
+		logger.error(vm.ip + " " + str(e))
 		PrintException()
 		return "ERROR:"+vm.ip+" "+str(e)
 	ssh.close()
@@ -75,7 +92,7 @@ def install_manager(vm):
 
 def install_worker(join_cmd, vm):
 	try:
-		print "%s: ====== Start Kubernetes Slave Installing ======" % (vm.ip)
+		logger.info("Starting kubernetes slave installation on: "+(vm.ip))
 		ssh = paramiko.SSHClient()
 		ssh.set_missing_host_key_policy(paramiko.AutoAddPolicy())
 		ssh.connect(vm.ip, username=vm.user, key_filename=vm.key)
@@ -97,14 +114,17 @@ def install_worker(join_cmd, vm):
 		stdout.read()
 		stdin, stdout, stderr = ssh.exec_command("sudo %s" % (join_cmd))
 		stdout.read()
-		print "%s: ========= Kubernetes Slave Installed =========" % (vm.ip)
+		logger.info("Finished kubernetes slave installation on: "+(vm.ip))
 	except Exception as e:
-		print '%s: %s' % (vm.ip, e)
+		#print '%s: %s' % (vm.ip, e)
+		logger.error(vm.ip + " " + str(e))
 		return "ERROR:"+vm.ip+" "+str(e)
 	ssh.close()
 	return "SUCCESS"
 
-def run(vm_list):
+def run(vm_list,rabbitmq_host,owner):
+        rabbit = DRIPLoggingHandler(host=rabbitmq_host, port=5672,user=owner)
+        logger.addHandler(rabbit)    
 	for i in vm_list:
 		if i.role == "master": 
 			join_cmd = install_manager(i)

drip-deployer/docker_service.py

@@ -31,6 +31,8 @@ if not getattr(logger, 'handler_set', None):
     logger.handler_set = True
 
 
+retry=0
+
 def scale_service(vm, application_name, service_name, service_num):
     try:
         logger.info("Starting docker service scaling on: "+vm.ip)
@@ -42,6 +44,7 @@ def scale_service(vm, application_name, service_name, service_num):
         stdout.read()
         logger.info("Finished docker service scaling on: "+vm.ip)        
     except Exception as e:
+        global retry
         logger.error(vm.ip + " " + str(e))
         return "ERROR:" + vm.ip + " " + str(e)
     ssh.close()

drip-deployer/docker_swarm.py

@@ -33,6 +33,8 @@ if not getattr(logger, 'handler_set', None):
     logger.handler_set = True
 
 
+retry=0
+
 def install_manager(vm):
 	try:
 		logger.info("Starting swarm manager installation on: "+(vm.ip))
@@ -58,9 +60,14 @@ def install_manager(vm):
 		ret = retstr[2].encode()
 		logger.info("Finished swarm manager installation on: "+(vm.ip))
 	except Exception as e:
+                global retry
+                if 'Connection timed out' in str(e) and retry < 3:
+                    retry+=1
+                    install_manager(vm)
 		logger.error(vm.ip + " " + str(e))
 		return "ERROR:" + vm.ip + " " + str(e)
 	ssh.close()
+	retry=0
 	return ret
 
 def install_worker(join_cmd, vm):
@@ -79,9 +86,13 @@ def install_worker(join_cmd, vm):
 		stdout.read()
 		logger.info("Finished swarm worker installation on: "+(vm.ip))
 	except Exception as e:
+                if 'Connection timed out' in str(e) and retry < 3:
+                    retry+=1
+                    install_worker(vm)            
 		logger.error(vm.ip + " " + str(e))
 		return "ERROR:" + vm.ip + " " + str(e)
 	ssh.close()
+	retry=0
 	return "SUCCESS"
 
 def run(vm_list,rabbitmq_host,owner):

drip-deployer/rpc_server.py

@@ -110,7 +110,7 @@ def handleDelivery(message):
             
 
     if manager_type == "kubernetes":
-        ret = docker_kubernetes.run(vm_list)
+        ret = docker_kubernetes.run(vm_list,rabbitmq_host,owner)
         return ret
     elif manager_type == "swarm":
         ret = docker_engine.run(vm_list,rabbitmq_host,owner)