From: Stephen Soltesz <soltesz@cs.princeton.edu>
Date: Wed, 9 Apr 2008 17:13:58 +0000 (+0000)
Subject: - added a checked time value
X-Git-Tag: Monitor-1.0-1~12
X-Git-Url: http://git.onelab.eu/?a=commitdiff_plain;h=0cb968415dbfb271022d23f497d82f0472e2f5f7;p=monitor.git

- added a checked time value
- added new kernel version.  need a better way to do this.
---

diff --git a/findbad.py b/findbad.py
index d1ee263..7c0a289 100755
--- a/findbad.py
+++ b/findbad.py
@@ -54,6 +54,9 @@ def collectPingAndSSH(nodename, cohash):
 	else:
 		values['ping'] = "PING"
 
+	#uptime = soltesz.SSH('root', nodename)
+	#(oval,eval) = uptime.run_noexcept("uptime | awk '{print $3,$4}' | tr , ' '")
+
 	### RUN SSH ######################
 	b_getbootcd_id = True
 	ssh = soltesz.SSH('root', nodename)
@@ -61,13 +64,12 @@ def collectPingAndSSH(nodename, cohash):
 	eval = ""
 	(oval, eval) = ssh.run_noexcept('echo `uname -a ; ls /tmp/bm.log`')
 	val = oval
-	if "2.6.17" in oval or "2.6.20" in oval:
+	if "2.6.17" in oval or "2.6.2" in oval:
 		values['ssh'] = 'SSH'
+		values['category'] = 'ALPHA'
 		if "bm.log" in oval:
-			values['category'] = 'ALPHA'
 			values['state'] = 'DEBUG'
 		else:
-			values['category'] = 'ALPHA'
 			values['state'] = 'BOOT'
 	elif "2.6.12" in oval or "2.6.10" in oval:
 		values['ssh'] = 'SSH'
@@ -104,7 +106,9 @@ def collectPingAndSSH(nodename, cohash):
 		val = oval
 		if "BootCD" in val:
 			values['bootcd'] = val
-			if "v2" in val and nodename is not "planetlab1.cs.unc.edu":
+			if "v2" in val and \
+				( nodename is not "planetlab1.cs.unc.edu" and \
+				  nodename is not "planetlab2.cs.unc.edu" ):
 				values['category'] = 'OLDBOOTCD'
 		else:
 			values['bootcd'] = ""
@@ -127,8 +131,10 @@ def collectPingAndSSH(nodename, cohash):
 	plc_lock.acquire()
 
 	try:
-		d_node = plc.getNodes({'hostname': nodename}, ['pcu_ids', 'site_id', 'last_contact', 'boot_state'])
+		d_node = plc.getNodes({'hostname': nodename}, ['pcu_ids', 'site_id', 'last_contact', 'boot_state', 'nodegroup_ids'])
 	except:
+		b_except = True
+		import traceback
 		b_except = True
 		import traceback
 		traceback.print_exc()
@@ -145,10 +151,12 @@ def collectPingAndSSH(nodename, cohash):
 			values['pcu'] = "NOPCU"
 		site_id = d_node[0]['site_id']
 		last_contact = d_node[0]['last_contact']
+		nodegroups = d_node[0]['nodegroup_ids']
 		values['plcnode'] = {'status' : 'SUCCESS', 
 							'pcu_ids': pcu, 
 							'boot_state' : d_node[0]['boot_state'],
 							'site_id': site_id,
+							'nodegroups' : nodegroups,
 							'last_contact': last_contact}
 	else:
 		values['pcu']     = "UNKNOWN"
@@ -183,6 +191,8 @@ def collectPingAndSSH(nodename, cohash):
 	else:
 		values['plcsite'] = {'status' : "GS_FAILED"}
 
+	values['checked'] = time.time()
+
 	return (nodename, values)
 
 def recordPingAndSSH(request, result):
@@ -261,13 +271,11 @@ def main():
 
 	# history information for all nodes
 	cohash = cotop.coget(cotop_url)
-
-	if config.filename == "":
-		l_nodes = syncplcdb.create_plcdb()
-		l_nodes = [node['hostname'] for node in l_nodes]
+	l_nodes = syncplcdb.create_plcdb()
+	l_nodes = [node['hostname'] for node in l_nodes]
 		#l_nodes = cohash.keys()
-	else:
-		l_nodes = config.getListFromFile(config.filename)
+#	else:
+#		l_nodes = config.getListFromFile(config.filename)
 
 	checkAndRecordState(l_nodes, cohash)