added log recovery with --gatherlogs

CLnode.py

@@ -92,6 +92,7 @@ def copy(self):
 	# Then extracts it and runs the setup script
 	def deploy(self,payload,sshKey,launch=False):
 		# TODO: error handeling
+		# remove SIG with ssh-keygen -f "/home/madmaze/.ssh/known_hosts" -R ec2-184-73-46-186.compute-1.amazonaws.com
 		print "\n============================================"
 		print "Deploing",self.instID,"/",self.instName
 		print "====="
@@ -123,3 +124,37 @@ def deploy(self,payload,sshKey,launch=False):
 					return -1
 		else:
 			print "Master node: No need to deploy!"
+
+	def gather(self,logsDir,sshKey):
+		print "\n============================================"
+		print "Gathering logs: ",self.instID,"/",self.instName
+		print "====="
+		logsDir+='/'+self.instID+'-'+self.instName
+
+		try:
+			res=GF.run("mkdir -p "+logsDir)
+			res=GF.run("scp -r -o StrictHostKeyChecking=no -i "+sshKey+" ubuntu@"+self.url+":/var/tmp/log "+logsDir)
+			print "scp -r -o StrictHostKeyChecking=no -i "+sshKey+" ubuntu@"+self.url+":/var/tmp/log "+logsDir
+		except Exception as x:
+			print x, "\n", res
+			return -1
+
+		'''
+		# EXTRACT payload
+		try:
+			res=GF.run("ssh -o StrictHostKeyChecking=no -i "+sshKey+" ubuntu@"+self.url+" 'tar xvf ~/bundle.tar;'")
+			print "ssh -o StrictHostKeyChecking=no -i "+sshKey+" ubuntu@"+self.url+" 'tar xvf ~/bundle.tar;'"
+		except Exception as x:
+			print x, "\n", res
+			return -1
+		
+		if launch is True:
+			# LAUNCH Payload
+			try:
+				res=GF.run("ssh -o StrictHostKeyChecking=no -i "+sshKey+" ubuntu@"+self.url+" 'python ~/payload/setup.py'")
+				print "ssh -o StrictHostKeyChecking=no -i "+sshKey+" ubuntu@"+self.url+" 'python ~/payload/setup.py'"
+				self.deployed = True
+			except Exception as x:
+				print x, "\n", res
+				return -1
+		'''

masterSpore.py

@@ -237,12 +237,13 @@ def monitor(n, timeout):
 	sshKey='~/.ec2/pkey'
 	rebuildBundle=True
 	payload='./bundle.tar'
+	logsDir='./logs'
 	payloadDir='./payload'
 	print "Cluster manager v0.2";
 	argc=0
 	try:
 		opts, args = getopt.getopt(sys.argv[1:], "dilh", ["help", "debug", "info", "list", "listspots", "listblock", 
-		"launch=", "shutdown", "deployall", "killall", "kill=","deploy=","master="])
+		"launch=", "shutdown", "deployall","gatherlogs", "killall", "kill=","deploy=","master="])
         except getopt.GetoptError, err:
 		# print help information and exit:
 		print str(err) # will print something like "option -a not recognized"
@@ -386,6 +387,16 @@ def monitor(n, timeout):
 				if n.status=='running':
 					n.deploy(payload,sshKey,True)
 			saveState()
+		elif o in ("--gatherlogs"):
+			lt = time.localtime(time.time())	
+			#should be ddmmyyyy-hhmmss AKA should not allow single digits in time
+			timestamp=str(lt[2])+str(lt[1])+str(lt[0])+"-"+str(lt[3])+":"+str(lt[4])+":"+str(lt[5])
+			if GF.confirmQuestion("!!This will gather up all logs from agents and master!\nAre you sure you want to continue?") is False:
+				sys.exit()
+			for n in GF.nodes:
+				if n.status=='running':
+					n.gather(logsDir+"/"+timestamp,sshKey)
+			saveState()
 		elif o in ("--help"):
 			print 	"Help:\n"
 			print	"./masterSporte.py --help - help menue"

payload/setup.py

@@ -10,5 +10,7 @@ def run(cmd):
 
 try:
 	res=run("touch ~/TEST")
+	res=run("mkdir -p /var/tmp/log")
+	res=run("ifconfig > /var/tmp/log/conifg")
 except Exception as x:
 	print x, "\n", res