def getLoads(dl):
result = 0
loads = {}
cmd = "uptime"
rval, lines = core.system(cmd, 3 <= dl)
if rval == 0 and len(lines) == 1:
parts = lines[0].split("load average: ")
if len(parts) == 2:
load_parts = parts[1].split(', ')
if len(load_parts) == 3:
loads = {
"1min": load_parts[0],
"5min": load_parts[1],
"15min": load_parts[2]
}
cmd = "nproc"
rval, lines = core.system(cmd, 3 <= dl)
if rval == 0 and len(lines) == 1:
loads["ncore"] = lines[0]
else:
loads["ncore"] = 0
return result, loads
def run (self):
cmd = "cd " + self.dir + "; " + self.cmd
rval = system_piped (cmd, self.pipe, self.dl <= DL)
if rval == 0:
rval2, lines = system ("touch " + self.dir + "/obs.finished")
else:
rval2, lines = system ("touch " + self.dir + "/obs.failed")
return rval
def run(self):
cmd = "cd " + self.dir + "; " + self.cmd
rval = system_piped(cmd, self.pipe, self.dl <= DL)
if rval == 0:
rval2, lines = system("touch " + self.dir + "/obs.finished")
else:
rval2, lines = system("touch " + self.dir + "/obs.failed")
return rval
def run(self):
cmd = "cd " + self.dir + "; " + self.cmd
self.script.log(2, "UWBProcThread::run " + self.cmd)
rval = self.script.system_piped(cmd, self.pipe, 2, self.env)
self.script.log(2, "UWBProcThread::run rval=" + str(rval))
if rval == 0:
rval2, lines = system("touch " + self.dir + "/obs.finished")
else:
rval2, lines = system("touch " + self.dir + "/obs.failed")
return rval
def run (self):
cmd = "cd " + self.dir + "; " + self.cmd
self.script.log (2, "UWBProcThread::run " + self.cmd)
rval = self.script.system_piped (cmd, self.pipe, 2, self.env)
self.script.log (2, "UWBProcThread::run rval=" + str(rval))
if rval == 0:
rval2, lines = system ("touch " + self.dir + "/obs.finished")
else:
rval2, lines = system ("touch " + self.dir + "/obs.failed")
return rval
def test_fluxcal(target, fluxcal_on_file, fluxcal_off_file):
# check if the target matches the fluxcal off file
cmd = "grep " + target + " " + fluxcal_on_file + " | wc -l"
rval, lines = system(cmd)
if rval == 0 and len(lines) == 1 and int(lines[0]) > 0:
return (True, "")
# check if the target matches the fluxcal off file
cmd = "grep " + target + " " + fluxcal_on_file + " | wc -l"
rval, lines = system(cmd)
if rval == 0 and len(lines) == 1 and int(lines[0]) > 0:
return (True, "")
return (False, target + " did not exist in fluxcal files")
def test_fluxcal (target, fluxcal_on_file, fluxcal_off_file):
target = clean_target(target)
# check if the target matches the fluxcal off file
cmd = "grep " + target + " " + fluxcal_on_file + " | wc -l"
rval, lines = system (cmd)
if rval == 0 and len(lines) == 1 and int(lines[0]) > 0:
return (True, "valid fluxcal on")
# check if the target matches the fluxcal off file
cmd = "grep " + target + " " + fluxcal_off_file + " | wc -l"
rval, lines = system (cmd)
if rval == 0 and len(lines) == 1 and int(lines[0]) > 0:
return (True, "valid fluxcal off")
return (False, target + " did not exist in fluxcal files")
def getDBState (self, key):
cmd = "dada_dbmetric -k " + key
rval, lines = system (cmd, False)
if rval == 0:
a = lines[0].split(',')
dat = {'nbufs':a[0], 'full':a[1], 'clear':a[2], 'written':a[3],'read':a[4]}
hdr = {'nbufs':a[5], 'full':a[6], 'clear':a[7], 'written':a[8],'read':a[9]}
return 0, hdr, dat
else:
return 1, {}, {}
def get_parfile_param (target, param):
target = clean_target(target)
(ok, message) = get_par_file(target)
if not ok:
return (False, message)
par_file = message
cmd = "grep -i " + param + " " + par_file + " | awk '{print $2}'"
rval, lines = system (cmd)
if rval != 0 or len(lines) <= 0:
return (False, "could not read parfile")
return (True, lines[0])
def get_psrcat_param(target, param):
cmd = "psrcat -all " + target + " -c " + param + " -nohead -o short"
rval, lines = system(cmd)
if rval != 0 or len(lines) <= 0:
return ("fail", "could not use psrcat")
if lines[0].startswith("WARNING"):
return ("fail",
"pulsar " + target + " did not exist in catalog " + lines[0])
parts = lines[0].split()
if len(parts) == 2 and parts[0] == "1":
return ("ok", parts[1])
def getLoads (dl):
result = 0
loads = {}
cmd = "uptime"
rval, lines = core.system (cmd, 3 <= dl)
if rval == 0 and len(lines) == 1:
parts = lines[0].split("load average: ")
if len(parts) == 2:
load_parts = parts[1].split(', ')
if len(load_parts) == 3:
loads = {"1min" : load_parts[0], "5min": load_parts[1], "15min": load_parts[2]}
cmd = "nproc"
rval, lines = core.system (cmd, 3 <= dl)
if rval == 0 and len(lines) == 1:
loads["ncore"] = lines[0]
else:
loads["ncore"] = 0
return result, loads
def getDiskCapacity (dirs, dl):
result = 0
disks = {}
for dir in dirs:
cmd = "df " + dir + " -B 1048576 -P | tail -n 1 | awk '{print $2,$3,$4}'"
rval, lines = core.system (cmd, 3 <= dl)
result += rval
if rval == 0 and len(lines) == 1:
parts = lines[0].split()
disks[dir] = {"size": parts[0], "used": parts[1], "available": parts[2]}
return result, disks
def get_psrcat_param (target, param):
target = clean_target(target)
cmd = "psrcat -all " + target + " -c " + param + " -nohead -o short"
rval, lines = system (cmd)
if rval != 0 or len(lines) <= 0:
return (False, "could not use psrcat")
if lines[0].startswith("WARNING"):
return (False, "pulsar " + target + " did not exist in catalog")
parts = lines[0].split()
if len(parts) == 2 and parts[0] == "1":
return (True, parts[1])
else:
return (False, "pulsar " + target + " did not match only 1 entry")
def get_psrcat_param(target, param):
# remove the _R suffix
if target.endswith('_R'):
target = target[:-2]
cmd = "psrcat -all " + target + " -c " + param + " -nohead -o short"
rval, lines = system(cmd)
if rval != 0 or len(lines) <= 0:
return (False, "could not use psrcat")
if lines[0].startswith("WARNING"):
return (False, "pulsar " + target + " did not exist in catalog")
parts = lines[0].split()
if len(parts) == 2 and parts[0] == "1":
return (True, parts[1])
else:
return (False, "pulsar " + target + " did not match only 1 entry")
def getNTPSynced (dl):
result = 0
if os.path.isfile ("/usr/sbin/ntpq"):
cmd = "/usr/sbin/ntpq -c 'rv 0 offset'"
elif os.path.isfile ("/usr/bin/ntpq"):
cmd = "/usr/bin/ntpq -c 'rv 0 offset'"
else:
return (0, True)
rval, lines = core.system (cmd, 3 <= dl)
if rval == 0 and len(lines) == 1:
if lines[0] != "/usr/sbin/ntpq: read: Connection refused":
parts = lines[0].split("=")
offset = float(parts[1])
# offset in milliseconds
if offset < 5:
return (result, True)
return (result, False)
def getDBState(self, key):
cmd = "dada_dbmetric -k " + key
rval, lines = system(cmd, False)
if rval == 0:
a = lines[0].split(',')
dat = {
'nbufs': a[0],
'full': a[1],
'clear': a[2],
'written': a[3],
'read': a[4]
}
hdr = {
'nbufs': a[5],
'full': a[6],
'clear': a[7],
'written': a[8],
'read': a[9]
}
return 0, hdr, dat
else:
return 1, {}, {}
def getIPMISensors(dl):
sensors = {}
cmd = "ipmitool sensor"
(rval, lines) = core.system(cmd, 3 <= dl)
if rval == 0 and len(lines) > 0:
for line in lines:
data = line.split("|")
metric_name = data[0].strip().lower().replace("+", "").replace(
" ", "_")
value = data[1].strip()
lower = data[6].strip()
upper = data[7].strip()
# Skip missing sensors
if value == 'na':
continue
state = "na"
metric_units = "na"
metric_value = "na"
# if sensor is a boolean value
if re.search("(0x)", value):
if value == "0x1":
metric_value = "ok"
metric_units = "na"
state = "ok"
else:
metric_value = "fail"
metric_units = "na"
state = "fail"
else:
# Extract out a float value
vmatch = re.search("([0-9.]+)", value)
if not vmatch:
continue
metric_value = vmatch.group(1)
metric_units = data[2].strip().replace("degrees C", "C")
state = data[3]
#lnr = float(data[4].strip())
#lcr = float(data[5].strip())
#lnc = float(data[6].strip())
#unc = float(data[7].strip())
#ucr = float(data[8].strip())
#unr = float(data[9].strip())
#state = "ok"
#if value <= lnr or value >= unr:
# state = "non recoverable"
#if value <= lcr or value >= ucr:
# state = "critical"
#if value <= lnc or value >= unc:
# state = "non critical"
sensors[metric_name] = {
"value": metric_value,
"units": metric_units,
"state": state,
"lower": lower,
"upper": upper
}
return rval, sensors
def getIPMISensors (dl):
sensors = {}
return (0, sensors)
cmd = "ipmitool sensor"
(rval, lines) = core.system (cmd, 3 <= dl)
if rval == 0 and len(lines) > 0:
for line in lines:
data = line.split("|")
metric_name = data[0].strip().lower().replace("+", "").replace(" ", "_")
value = data[1].strip()
lower = data[6].strip()
upper = data[7].strip()
# Skip missing sensors
if value == 'na':
continue
state = "na"
metric_units = "na"
metric_value = "na"
# if sensor is a boolean value
if re.search("(0x)", value):
if value == "0x1":
metric_value = "ok"
metric_units = "na"
state = "ok"
else:
metric_value = "fail"
metric_units = "na"
state = "fail"
else:
# Extract out a float value
vmatch = re.search("([0-9.]+)", value)
if not vmatch:
continue
metric_value = vmatch.group(1)
metric_units = data[2].strip().replace("degrees C", "C")
state = data[3]
#lnr = float(data[4].strip())
#lcr = float(data[5].strip())
#lnc = float(data[6].strip())
#unc = float(data[7].strip())
#ucr = float(data[8].strip())
#unr = float(data[9].strip())
#state = "ok"
#if value <= lnr or value >= unr:
# state = "non recoverable"
#if value <= lcr or value >= ucr:
# state = "critical"
#if value <= lnc or value >= unc:
# state = "non critical"
sensors[metric_name] = { "value": metric_value, "units": metric_units, "state": state, "lower":lower, "upper":upper }
return rval, sensors
Daemon.conclude(self)
###############################################################################
#
if __name__ == "__main__":
if len(sys.argv) != 2:
print "ERROR: 1 command line argument expected"
sys.exit(1)
cfg = sys.argv[1]
cfg_dir = os.environ.get('SPIP_ROOT') + "/share"
cmd = "cp " + cfg_dir + "/" + cfg + "/*.cfg " + cfg_dir + "/"
system(cmd, False)
hostname = getHostNameShort()
script = LMCDaemon("spip_lmc", hostname)
state = script.configure(DAEMONIZE, DL, "lmc", "lmc")
if state != 0:
sys.exit(state)
script.log(1, "STARTING SCRIPT")
try:
script.main()
except: