def saveCfgInfo(addr, data):
try:
succFlag = 0
# 解析参数数据
str = "INSERT INTO OP_MCH_DT_CFG(OP_MCH_HD_ID, GATHERDATE, CPUCOREQTY, DISKSIZE,SWAPSIZE,MEMSIZE, SYSTEMVERSION, EFFFLAG ,MEMO) " \
"VALUES (%s, %s, %s, %s, %s, %s, %s, %s, %s)"
val = (
db.query_node_id(addr),
# data['gatherDate'],
time.strftime("%Y%m%d%H%M%S",
time.strptime(data['gatherDate'], "%Y%m%d%H%M")),
int(data['cpuCoreQty']),
data['diskSize'],
data['swapSize'],
data['memSize'],
data['systemVersion'],
1,
"")
succFlag = db.insert_data(sql=str, val=val)
except Exception:
# logging.exception('保存配置信息 saveCfgInfo error')
setLog('保存配置信息 saveCfgInfo error:{0}'.format(traceback.format_exc()))
# setLog( '保存配置信息 saveCfgInfo.data:{0}'.format(data))
succFlag = 0
return succFlag
def insert_table_batch(hdSql, hdVal, dtSql, dtValList):
con = connect_op_db()
cur = con.cursor()
valList = []
succFlag = 0
try:
cur.execute(hdSql, hdVal)
hdKey = cur.lastrowid
for i in dtValList:
i = (hdKey, ) + i
valList.append(i)
cur.executemany(dtSql, valList)
assert cur.rowcount == len(dtValList), 'my error message'
con.commit()
succFlag = 1
except Exception:
setLog('insert_table_batch() Insert operation error:{0}'.format(
traceback.format_exc()))
con.rollback()
succFlag = 0
# logging.exception('Insert operation error')
return 0
finally:
cur.close()
con.close()
return succFlag
def testConn(sleeptime):
while True:
print('testConn')
try:
# print('Waiting for connection...')
testConnect()
except Exception as e:
setLog('连接测试失败:{0}'.format(e))
exit()
time.sleep(sleeptime)
def agentMonitor(sleeptime):
while True:
print('agentMonitor')
try:
# print('Waiting for connection...')
monitor()
except Exception as e:
setLog('代理进程监控失败:{0}'.format(e))
exit()
time.sleep(sleeptime)
def RecvData(sock, addr, buffsize):
try:
data = ''
sock.send(b'Welcome!')
# 用字符串数组缓存数据
buffer = []
while True:
# 一次接收1024字节的数据
try:
recv = sock.recv(buffsize).decode('utf-8')
except Exception as e:
logging.INFO(e)
# 休眠1s
time.sleep(0.1)
# 如果还能接收到数据，则继续接收，否则中断接收
if recv:
buffer.append(recv)
else:
break
# 字符数组组合成字符串
data = data + ''.join(buffer)
buffer = []
# data最前面10位为信息内容长度，10到23位为发送的时间戳（毫秒级），后面的为实际内容
# 以上内容Demo格式，具体格式需要统一定义
if len(data) >= int(data[0:10]):
# lens = data[0:10]
sendtime = data[10:23]
hostname = re.findall(r'({.*?})', data[23:])[0]
data = data[23 + len(hostname):]
break
# print('lens：' + str(int(lens))) # 信息内容长度
# print('sendtime：' + sendtime) # 发送的时间戳（毫秒级）
# print('hostname:' + hostname) # hostname
# print('data:' + data) # 实际内容
# 调用数据处理路由，处理数据
# datarouter(data)
# sock.close()
lens = str(len(data))
if datarouter(addr[0], data):
# add by cwt 20181022 返回接收数据后的处理数据结果
# 如果成功则，code=200
srecvmsg = '{"code":"200","lens":"' + lens + '","msg":"传输成功"}'
sock.send(srecvmsg.encode('utf-8'))
else:
srecvmsg = '{"code":"998","lens":"' + lens + '","msg":"数据库数据写入失败"}'
sock.send(srecvmsg.encode('utf-8'))
except Exception as e:
srecvmsg = '{"code":"999","lens":"' + lens + '","msg":"{0}"{1}'.format(
e, '}')
sock.send(srecvmsg.encode('utf-8'))
setLog(traceback.format_exc())
finally:
sock.close()
def receiveData():
print('进入进程')
while True:
print('receiveData')
try:
# print('Waiting for connection...')
clientSocket, addr = serverSocket.accept()
print('addr =', addr)
RecvData(clientSocket, addr, CON_RECV['BUFFSIZE'])
except Exception as e:
setLog('数据接收失败:{0}'.format(e))
exit()
def select_table(sql):
con = connect_op_db()
cur = con.cursor()
try:
cur.execute(sql)
results = cur.fetchall()
except Exception:
setLog('select_table() select operation error:{0}'.format(
traceback.format_exc()))
con.rollback()
return 0
finally:
cur.close()
con.close()
return results
def query_node_id(ipAddr):
sql_str = ("SELECT id" + " FROM OP_MCH_HD" + " WHERE ipAddr='%s'" %
(ipAddr))
# logging.info(sql_str)
con = connect_op_db()
cur = con.cursor()
cur.execute(sql_str)
row = cur.fetchone()
cur.close()
con.close()
if not (row):
setLog('Fatal error: ipAddr:{0} does not exists!'.format(ipAddr))
else:
return row[0]
def update_table(sql):
con = connect_op_db()
cur = con.cursor()
succFlag = 1
try:
cur.execute(sql)
con.commit()
except Exception:
succFlag = 0
setLog('update_table() update operation error:{0}'.format(
traceback.format_exc()))
con.rollback()
finally:
cur.close()
con.close()
return succFlag
def testConnect():
# startTime = time.strftime("%Y%m%d%H%M%S", time.localtime())
sqlStr = (
"SELECT H.id,H.ipaddr,break_time,run_time" +
" FROM OP_MCH_HD H,OP_MCH_SET_INFO I WHERE I.mch_id=H.id AND I.normal_flag=1"
)
# logging.info(sql_str)
con = connect_op_db()
cur = con.cursor()
cur.execute(sqlStr)
row = cur.fetchall()
mchList = []
start = 0
for i in range(0, 7):
end = start + ceil(len(row) / 7)
mchList.append(row[start:end])
start = end
try:
thread1 = createThread(1, "thread1 ", connMch, (mchList[0], ))
thread2 = createThread(2, "thread2 ", connMch, (mchList[1], ))
thread3 = createThread(3, "thread3 ", connMch, (mchList[2], ))
thread4 = createThread(4, "thread4 ", connMch, (mchList[3], ))
thread5 = createThread(5, "thread5 ", connMch, (mchList[4], ))
thread6 = createThread(6, "thread6 ", connMch, (mchList[5], ))
thread7 = createThread(7, "thread7 ", connMch, (mchList[6], ))
thread1.start()
thread2.start()
thread3.start()
thread4.start()
thread5.start()
thread6.start()
thread7.start()
thread1.join()
thread2.join()
thread3.join()
thread4.join()
thread5.join()
thread6.join()
thread7.join()
except:
# raise
setLog('多线程主机连接测试失败：{0}'.format(traceback.format_exc()))
finally:
cur.close()
con.close()
def insert_data_batch(sql, val):
con = connect_op_db()
cur = con.cursor()
succFlag = 0
try:
cur.executemany(sql, val)
assert cur.rowcount == len(val), 'my error message'
con.commit()
succFlag = 1
except Exception:
con.rollback()
setLog('insert_data_batch() Insert operation error:{0}'.format(
traceback.format_exc()))
succFlag = 0
finally:
cur.close()
con.close()
return succFlag
def insert_data(sql, val):
con = connect_op_db()
cur = con.cursor()
succFlag = 0
try:
cur.execute(sql, val)
# lastrowid = cur.lastrowid
con.commit()
succFlag = 1
except Exception:
con.rollback()
setLog('insert_data() Insert operation error:{0}'.format(
traceback.format_exc()))
raise
succFlag = 0
finally:
cur.close()
con.close()
return succFlag
def connMch(mchList):
try:
con = connect_op_db()
cur = con.cursor()
selectDict = (
"SELECT id,dictcode from stm_dict where dictcode in ('D_HOST_NORMAL','D_HOST_CLOSE') "
)
cur.execute(selectDict)
dictRow = cur.fetchall()
for dictInfo in dictRow:
if dictInfo[1] == 'D_HOST_NORMAL':
normalKey = dictInfo[0]
elif dictInfo[1] == 'D_HOST_CLOSE':
closeKey = dictInfo[0]
for info in mchList:
mchId = info[0]
ipAddr = info[1]
result = subprocess.call('ping -c 1 ' + ipAddr,
shell=True,
stdout=subprocess.PIPE,
stderr=subprocess.PIPE)
modifyTime = time.strftime("%Y%m%d%H%M%S", time.localtime())
if result == 1:
if not info[2]:
sqlUpdateStr = "update OP_MCH_HD set break_time={0},mch_status_id={1},run_time=NULL,amdlog={0} where id={2}".format(
modifyTime, closeKey, mchId)
cur.execute(sqlUpdateStr)
else:
if not info[3]:
sqlUpdateStr = "update OP_MCH_HD set run_time={0},mch_status_id={1},break_time=NULL,amdlog={0} where id={2}".format(
modifyTime, normalKey, mchId)
cur.execute(sqlUpdateStr)
con.commit()
except:
raise
setLog('主机测试连接失败：{0}'.format(traceback.format_exc()))
finally:
cur.close()
con.close()
def SendData(addr, buffersize, buffer):
try:
hostname = '{' + socket.gethostname() + '}'
# 定义一个套接字
sock = socket.socket(socket.AF_INET, socket.SOCK_STREAM)
# 连接服务，指定主机和端口
sock.connect(addr)
sock.recv(buffersize).decode('utf-8')
# buffer最前面10位加上信息内容长度，10到23位为发送的时间戳（毫秒级）
buffer = str('%010d' % (len(buffer) + 23 + len(hostname))) + str(
int(round(time.time() * 1000))) + hostname + buffer
sock.send(buffer.encode('utf-8'))
srecvmsg = sock.recv(buffersize).decode('utf-8')
# 日志直接写返回json
logging.info('recvtime:' + srecvmsg + ',data:' +
buffer[23 + len(hostname):])
sock.close()
except Exception as e:
setLog('信息发送失败:{0}'.format(traceback.format_exc()))
srecvmsg = '{0}"code":999,"msg":"发送失败{1}"{2}'.format('{', e, '}')
return srecvmsg
def monitor():
try:
sqlStr = (
"SELECT H.id,H.ipaddr,h.break_time,h.run_time" +
" FROM OP_MCH_HD H,OP_MCH_SET_INFO I WHERE I.mch_id=H.id AND I.agent_process_flag=1"
)
# sqlStr = ("SELECT H.id,H.ipaddr"
# + " FROM OP_MCH_HD H")
con = connect_op_db()
cur = con.cursor()
cur.execute(sqlStr)
row = cur.fetchall()
for mchInfo in row:
mchId = mchInfo[0]
bufferStr = '{0}{1}'.format(
mchId, time.strftime("%Y%m%d%H%M%S", time.localtime()))
ipAddr = (mchInfo[1], MCH_REC_PORT)
srecvmsg = json.loads(SendData(ipAddr, 1024, bufferStr))
curTime = time.strftime("%Y%m%d%H%M%S", time.localtime())
if int(srecvmsg['code']) != 200:
sqlUpdateStr = "INSERT INTO OP_MCH_DT_AGENT( OP_MCH_HD_ID, AGENT_NAME, END_TIME,MEMO, CRTUSR, CRTLOG ,AMDUSR,AMDLOG) " \
'VALUES (' + str(
mchId) + ',' + '\'' + '数据采集代理程序' + '\'' + ',' + curTime + ',' + '\'' + '数据采集代理程序失败' + '\'' + ',' + '1' + ',' + curTime + ',' + '1' + ',' + curTime + ')'
succFlag = db.insert_data_one(sqlUpdateStr)
if succFlag != 1:
setLog('主机代理进程监控失败：{0}'.format(traceback.format_exc()))
# else:
# sqlUpdateStr = "INSERT INTO OP_MCH_DT_AGENT( OP_MCH_HD_ID, AGENT_NAME, END_TIME,MEMO, CRTUSR, CRTLOG ,AMDUSR,AMDLOG) " \
# 'VALUES (' + str(
# mchId) + ',' + '\'' + '数据采集代理程序' + '\'' + ',' + curTime + ',' + '\'' + '数据采集代理程序' + '\'' + ',' + '1' + ',' + curTime + ',' + '1' + ',' + curTime + ')'
# succFlag = db.insert_data_one(sqlUpdateStr)
except:
# raise
setLog('主机代理进程监控失败：{0}'.format(traceback.format_exc()))
finally:
cur.close()
con.close()
def saveSeverInfo(addr, data):
try:
succFlag = 1
mchhdid = db.query_node_id(addr)
serverList = data['serverList']
updatesql = None
updateflag = 0
# 拼接超负荷设置查询sql语句
getMchSetSql = 'select ' \
' o.server_flag ' \
' from op_mch_set_info o ' \
'where o.effflag = 1 ' \
' and o.mch_id =' + str(db.query_node_id(addr))
getMchSetResults = db.select_table(getMchSetSql)
if getMchSetResults:
# 获取第一条信息
setRow = getMchSetResults[0]
# 判断是否需要写入进程
if setRow[0] == 1:
getsql = 'select ds.id, d.dictname,ds.last_begin_time,ds.last_close_time,ds.pid ' \
' from op_mch_dt_server s' \
' ,op_mch_dtt_server ds ' \
' ,stm_dict d ' \
' where s.id = ds.op_mch_dt_server_id ' \
' and ds.node_id = d.id ' \
' and s.op_mch_hd_id =' + str(mchhdid)
results = db.select_table(getsql)
for row in results:
nodeName = row[1]
for server in serverList:
if server['nodeName'] == nodeName:
# 无开始时间为第一次获取 更新开始时间
if not row[2]:
updatesql = 'update op_mch_dtt_server ds ' \
' set ds.last_begin_time =' + 'str_to_date' + '(' + data[
'gatherDate'] + ',' + '\'%Y%m%d%H%i\')' + \
' ,ds.pid = ' + server['pid'] + \
' where ds.id =' + str(row[0])
# 之前已有结束时间则认为已结束 此次检测到则重置开始时间 和 结束时间
elif row[3]:
updatesql = 'update op_mch_dtt_server ds ' \
' set ds.last_begin_time =' + 'str_to_date' + '(' + data[
'gatherDate'] + ',' + '\'%Y%m%d%H%i\')' \
' ,ds.last_close_time =' + 'null' \
' where ds.id =' + str(
row[0])
updateflag = 1
else:
pass
# 如果数据表有数据 但是轮询采集的进程无匹配数据 则认为进程已关闭 且当前结束时间为空 开始更新结束时间
if updateflag == 0 and not row[3]:
updatesql = 'update op_mch_dtt_server ds ' \
' set ds.last_close_time =' + 'str_to_date' + '(' + data[
'gatherDate'] + ',' + '\'%Y%m%d%H%i\')' \
' ,ds.pid = null' \
' where ds.id =' + str(row[0])
if updatesql:
succFlag = db.update_table(updatesql)
except Exception as e:
succFlag = 0
results = []
# logging.exception('保存服务进程信息 saveSeverInfo error')
setLog('保存服务进程信息 saveSeverInfo error:{0}'.format(
traceback.format_exc()))
# setLog( '保存服务进程信息 saveSeverInfo.data:{0}'.format(data))
return succFlag
except Exception as e:
setLog('代理进程监控失败:{0}'.format(e))
exit()
time.sleep(sleeptime)
if __name__ == '__main__':
try:
# 只在服务启动的时候执行一次，创建日志文件
initialize()
connSleepTime = sleepTime(0, 1, 0)
monitorSleepTime = sleepTime(0, 10, 0)
# 启用多线程处理消息接收
receiveData = createThread(1, "receiveData ", '', receiveData, (''))
testconn = createThread(2, "testConn ", connSleepTime, testConn,
(connSleepTime, ))
agentmonitor = createThread(3, "agentMonitor ", monitorSleepTime,
agentMonitor, (monitorSleepTime, ))
receiveData.start()
testconn.start()
agentmonitor.start()
receiveData.join()
testconn.join()
agentmonitor.join()
except Exception as e:
# print('traceback.format_exc()=', traceback.format_exc())
setLog(traceback.format_exc())
serverSocket.close()
raise
def initialize():
createLogFolder(FILE_PATH)
log.setLog("文件创建成功")
def saveLoadInfo(addr, data):
try:
alarm = {}
alarmList = []
mchAlarmCode = 'D_HOST_NORMAL'
# 解析参数数据
infoSql = "INSERT INTO OP_MCH_DT_INFO(" \
"OP_MCH_HD_ID,GATHERDATE,CPUUS,CPUSY,CPUID,CPUWA,MEMUSED,MEMBUFFCACHE,SWAPUSED" \
",LOADAVGFIVE,LOADAVGTEN,LOADAVGFIFTEEN,TOTALDISKREAD,TOTALDISKWRITE,NETINCOMINGAVG,NETOUTGOINGAVG,EFFFLAG,MEMO) " \
"VALUES (%s, %s, %s, %s, %s, %s, %s, %s, %s, %s, %s, %s, %s, %s, %s, %s, %s, %s)"
# 保留三位小数
infoVal = (
db.query_node_id(addr),
# data['gatherDate'],
time.strftime("%Y%m%d%H%M%S",
time.strptime(data['gatherDate'], "%Y%m%d%H%M")),
data['cpuUs'],
data['cpuSy'],
data['cpuId'],
data['cpuWa'],
float('%.3f' % data['memUsed']),
float('%.3f' % data['memBuffCache']),
float('%.3f' % data['swapUsed']),
data['loadAvgFive'],
data['loadAvgTen'],
data['loadAvgFifteen'],
float('%.3f' % data['totalDiskRead']),
float('%.3f' % data['totalDiskWrite']),
float('%.3f' % data['netIncomingAvg']),
float('%.3f' % data['netOutgoingAvg']),
1,
"")
dirSql = "INSERT INTO OP_MCH_DT_DIR( DIRNAME, USEDSIZE,FILESYSTEM,DIRSIZE,USEDPER,MCH_ID, EFFFLAG ,MEMO,AMDLOG,AMDUSR,CRTLOG,CRTUSR) " \
"VALUES (%s, %s, %s, %s, %s,%s, %s, %s, %s, %s, %s, %s)"
succFlag = db.insert_data(infoSql, infoVal)
if succFlag == 1:
if len(data['dirInfo']) > 0:
# 清除旧的数据
deletesql = 'delete from op_mch_dt_dir where mch_id =' + str(
db.query_node_id(addr))
succFlag = db.delete_data(deletesql)
if succFlag == 1:
for i in data['dirInfo']:
val = tuple(i.values()) + (
db.query_node_id(addr), 1, "",
time.strftime("%Y%m%d%H%M%S", time.localtime()), 1,
time.strftime("%Y%m%d%H%M%S", time.localtime()), 1)
succFlag = db.insert_data(dirSql, val)
# 拼接主机配置查询语句
getMchCfgSql = 'select ' \
' c.disksize' \
', c.memsize' \
', c.swapsize ' \
' from op_mch_dt_cfg c' \
' where c.op_mch_hd_id = ' + str(db.query_node_id(addr)) + \
' order by c.gatherdate desc' \
' limit 1'
getMchCfgResults = db.select_table(getMchCfgSql)
if getMchCfgResults:
cfgRow = getMchCfgResults[0]
totalMem = cfgRow[1]
totalSwap = cfgRow[2]
# 拼接超负荷设置查询sql语句
getMchSetSql = 'select ' \
' o.cpu_flag ' \
', o.cpu_warning ' \
', o.cpu_serious' \
', o.memory_flag' \
', o.memory_warning' \
', o.memory_serious' \
', o.file_sys_dir_flag' \
', o.file_sys_dir_warning' \
', o.file_sys_dir_serious' \
', o.disk_io_flag' \
', o.disk_io_warning' \
', (select s.dictcode from stm_dict s where s.id = o.disk_io_warning_unit) disk_io_warning_code' \
', o.disk_io_serious' \
', (select s.dictcode from stm_dict s where s.id = o.disk_io_serious_unit) disk_io_serious_code' \
', o.swap_flag' \
', o.swap_warning' \
', o.swap_serious' \
', o.net_io_flag' \
', o.net_io_warning' \
', (select s.dictcode from stm_dict s where s.id = o.net_io_warning_unit) net_io_warning_unit_code' \
', o.net_io_serious' \
', (select s.dictcode from stm_dict s where s.id = o.net_io_serious_unit) net_io_serious_unit_code' \
', o.load_flag' \
', o.load_warning' \
', o.load_serious' \
', o.time_check_flag' \
', o.time_check_num' \
' from op_mch_set_info o ' \
'where o.effflag = 1 ' \
' and o.mch_id =' + str(db.query_node_id(addr))
getMchSetResults = db.select_table(getMchSetSql)
if getMchSetResults:
row = getMchSetResults[0]
# 判断cpu是否超负载
if row[0] == 1:
alarm = {}
if Decimal.from_float(data['cpuUs']) >= row[2]:
alarm['alarmtypecode'] = 'D_ALARM_DANGER'
alarm['alarminfo'] = 'cpu使用率' + str(
data['cpuUs'] * 100) + '%' + '负荷严重！'
elif Decimal.from_float(data['cpuUs']) >= row[1]:
alarm['alarmtypecode'] = 'D_ALARM_WARM'
alarm['alarminfo'] = 'cpu使用率' + str(
data['cpuUs'] * 100) + '%' + '负荷告警！'
else:
pass
if len(alarm) > 0:
alarmList.append(alarm)
# 判断内存是否超负载
if row[3] == 1:
if totalMem:
alarm = {}
if Decimal.from_float(
data['memUsed']) / totalMem >= row[5]:
alarm['alarmtypecode'] = 'D_ALARM_DANGER'
alarm['alarminfo'] = '内存使用达' + str(
data['memUsed']) + 'M' + '使用严重！'
elif Decimal.from_float(
data['memUsed']) / totalMem >= row[4]:
alarm['alarmtypecode'] = 'D_ALARM_WARM'
alarm['alarminfo'] = '内存使用达' + str(
data['memUsed']) + 'M' + '使用告警！'
else:
pass
if len(alarm) > 0:
alarmList.append(alarm)
# 判断文件系统使用状况
if row[6] == 1:
if len(data['dirInfo']) > 0:
for i in data['dirInfo']:
alarm = {}
if Decimal.from_float(i['usedPer']) >= row[8]:
alarm['alarmtypecode'] = 'D_ALARM_DANGER'
alarm['alarminfo'] = '文件系统 挂载点:' + i[
'dirName'] + ' ' + str(
i['usedPer'] * 100) + '%' + '使用严重！'
elif Decimal.from_float(
i['usedPer']) >= row[7]:
alarm['alarmtypecode'] = 'D_ALARM_WARM'
alarm['alarminfo'] = '文件系统 挂载点:' + i[
'dirName'] + ' ' + str(
i['usedPer'] * 100) + '%' + '使用告警！'
else:
pass
if len(alarm) > 0:
alarmList.append(alarm)
# 判断磁盘io负载
if row[9] == 1:
# 更新警告阀值为kb
warnIo = row[10]
if row[11] == 'D_IOUNIT_MB':
warnIo = row[10] * 1024
elif row[11] == 'D_IOUNIT_GB':
warnIo = row[10] * 1024 * 1024
else:
pass
# 更新危险阀值为kb
serIo = row[12]
if row[13] == 'D_IOUNIT_MB':
serIo = row[12] * 1024
elif row[13] == 'D_IOUNIT_GB':
serIo = row[12] * 1024 * 1024
else:
pass
# 重置告警信息
alarm = {}
if Decimal.from_float(data['totalDiskRead']) >= serIo:
alarm['alarmtypecode'] = 'D_ALARM_DANGER'
alarm['alarminfo'] = '磁盘读取io 为' + str(
data['totalDiskRead']) + 'kb/s' + '负载严重！'
elif Decimal.from_float(
data['totalDiskRead']) >= warnIo:
alarm['alarmtypecode'] = 'D_ALARM_WARM'
alarm['alarminfo'] = '磁盘读取io 为' + str(
data['totalDiskRead']) + 'kb/s' + '负载告警！'
if len(alarm) > 0:
alarmList.append(alarm)
# 重置告警信息
alarm = {}
if Decimal.from_float(data['totalDiskWrite']) >= serIo:
alarm['alarmtypecode'] = 'D_ALARM_DANGER'
alarm['alarminfo'] = '磁盘写入io 为' + str(
data['totalDiskWrite']) + 'kb/s' + '负载严重！'
elif Decimal.from_float(
data['totalDiskWrite']) >= warnIo:
alarm['alarmtypecode'] = 'D_ALARM_WARM'
alarm['alarminfo'] = '磁盘写入io 为' + str(
data['totalDiskWrite']) + 'kb/s' + '负载告警！'
if len(alarm) > 0:
alarmList.append(alarm)
# 判断Swap是否超负载
if row[14] == 1:
if totalSwap:
# 重置告警信息
alarm = {}
if Decimal.from_float(
data['swapUsed']) / totalSwap >= row[16]:
alarm['alarmtypecode'] = 'D_ALARM_DANGER'
alarm['alarminfo'] = 'Swap使用达为' + str(
data['swapUsed']) + 'M' + '使用严重！'
elif Decimal.from_float(
data['swapUsed']) / totalSwap >= row[15]:
alarm['alarmtypecode'] = 'D_ALARM_WARM'
alarm['alarminfo'] = 'Swap使用达为' + str(
data['swapUsed']) + 'M' + '使用告警！'
else:
pass
if len(alarm) > 0:
alarmList.append(alarm)
# 判断网络io负载
if row[17] == 1:
# 更新警告阀值为kb
warnnetIo = row[18]
if row[19] == 'D_IOUNIT_MB':
warnnetIo = row[18] * 1024
elif row[19] == 'D_IOUNIT_GB':
warnnetIo = row[18] * 1024 * 1024
else:
pass
# 更新危险阀值为kb
sernetIo = row[20]
if row[21] == 'D_IOUNIT_MB':
sernetIo = row[20] * 1024
elif row[21] == 'D_IOUNIT_GB':
sernetIo = row[20] * 1024 * 1024
else:
pass
# 重置告警信息
alarm = {}
if Decimal.from_float(
data['netIncomingAvg']) >= sernetIo:
alarm['alarmtypecode'] = 'D_ALARM_DANGER'
alarm['alarminfo'] = '网络平均下载io 为' + str(
data['netIncomingAvg']) + 'kb/s' + '负载严重！'
elif Decimal.from_float(
data['totalDiskRead']) >= warnnetIo:
alarm['alarmtypecode'] = 'D_ALARM_WARM'
alarm['alarminfo'] = '网络平均下载io 为' + str(
data['netIncomingAvg']) + 'kb/s' + '负载告警！'
if len(alarm) > 0:
alarmList.append(alarm)
# 重置告警信息
alarm = {}
if Decimal.from_float(
data['netOutgoingAvg']) >= sernetIo:
alarm['alarmtypecode'] = 'D_ALARM_DANGER'
alarm['alarminfo'] = '网络平均上传io 为' + str(
data['netOutgoingAvg']) + 'kb/s' + '负载严重！'
elif Decimal.from_float(
data['totalDiskWrite']) >= warnnetIo:
alarm['alarmtypecode'] = 'D_ALARM_WARM'
alarm['alarminfo'] = '网络平均上传io 为' + str(
data['netOutgoingAvg']) + 'kb/s' + '负载告警！'
if len(alarm) > 0:
alarmList.append(alarm)
# 判断负载均衡数值
if row[22] == 1:
alarm = {}
if Decimal.from_float(data['loadAvgFive']) >= row[24]:
alarm['alarmtypecode'] = 'D_ALARM_DANGER'
alarm['alarminfo'] = '五分钟负载均衡达' + str(
data['loadAvgFive']) + '负荷严重！'
elif Decimal.from_float(
data['loadAvgFive']) >= row[23]:
alarm['alarmtypecode'] = 'D_ALARM_WARM'
alarm['alarminfo'] = '五分钟负载均衡达' + str(
data['loadAvgFive']) + '负荷告警！'
else:
pass
if len(alarm) > 0:
alarmList.append(alarm)
alarm = {}
if Decimal.from_float(data['loadAvgTen']) >= row[24]:
alarm['alarmtypecode'] = 'D_ALARM_DANGER'
alarm['alarminfo'] = '十分钟分钟负载均衡达' + str(
data['loadAvgTen']) + '负荷严重！'
elif Decimal.from_float(
data['loadAvgFive']) >= row[23]:
alarm['alarmtypecode'] = 'D_ALARM_WARM'
alarm['alarminfo'] = '十分钟负载均衡达' + str(
data['loadAvgTen']) + '负荷告警！'
else:
pass
if len(alarm) > 0:
alarmList.append(alarm)
alarm = {}
if Decimal.from_float(
data['loadAvgFifteen']) >= row[24]:
alarm['alarmtypecode'] = 'D_ALARM_DANGER'
alarm['alarminfo'] = '十五分钟负载均衡达' + str(
data['loadAvgFifteen']) + '负荷严重！'
elif Decimal.from_float(
data['loadAvgFive']) >= row[23]:
alarm['alarmtypecode'] = 'D_ALARM_WARM'
alarm['alarminfo'] = '十五分钟负载均衡达' + str(
data['loadAvgFifteen']) + '负荷告警！'
else:
pass
if len(alarm) > 0:
alarmList.append(alarm)
# 写入报警信息表
if alarmList:
# 存在告警信息代表主机状态至少是警告状态
mchAlarmCode = 'D_HOST_WARNING'
for alarm in alarmList:
alarm_type_id = getDictId(alarm['alarmtypecode'])
alarm_info = alarm['alarminfo']
# 拼接select语句
alarmSql = "INSERT INTO op_mch_dt_alarm(" \
"OP_MCH_HD_ID,GATHERDATE,ALARM_TYPE_ID,ALARM_INFO,EFFFLAG) "
# 拼接value语句
alarmVal = 'values' + '(' \
+ str(db.query_node_id(addr)) + ',' \
+ '\'' + time.strftime("%Y%m%d%H%M%S",
time.strptime(data['gatherDate'], "%Y%m%d%H%M")) + '\'' + ',' \
+ ('null' if not alarm_type_id else str(alarm_type_id)) + ',' \
+ '\'' + alarm_info + '\'' + ',' \
+ str(1) + ')'
if alarmSql:
succFlag = db.insert_data_one(alarmSql + alarmVal)
# 如果存在危险信息 则更新主机状态为危险
if alarm[
'alarmtypecode'] == 'D_ALARM_DANGER' and mchAlarmCode != 'D_HOST_ERROR':
mchAlarmCode = 'D_HOST_ERROR'
else:
# 无告警信息代表正常
mchAlarmCode = 'D_HOST_NORMAL'
updateMchSql = 'update op_mch_hd h ' \
' set h.mch_status_id = ' + str(getDictId(mchAlarmCode)) + \
' where h.id =' + str(db.query_node_id(addr))
if updateMchSql:
db.update_table(updateMchSql)
except Exception as e:
setLog('保存负载信息 saveLoadInfo error:{0}'.format(traceback.format_exc()))
# setLog( '保存配置信息 saveCfgInfo.data:{0}'.format(data))
succFlag = 0
return succFlag
