完善相关命令, 增加diagnose命令.

29 files changed, 869 insertions, 162 deletions

diff --git a/.gitignore b/.gitignore
index a4fb0cc..17451b0 100644
--- a/.gitignore
+++ b/.gitignore
@@ -1 +1,2 @@
 py_bin/*

+py_temp/*

diff --git a/deploy_doc/deploy.md b/deploy_doc/deploy.md
deleted file mode 100644
index aa7c573..0000000
--- a/deploy_doc/deploy.md
+++ /dev/null
@@ -1,22 +0,0 @@
-1.交换板文件

-  /opt/tsg/bin

-  /opt/tsg/etc

-  

-  更新telegraf.conf的global tags, 本机sn.

-

-2.计算板文件(4台)

-

-

-3.服务

-启动交换板crontab服务:

-service crond start

-chkconfig crond on

-crontab -e, 输入以下内容:

-* * * * * /opt/tsg/bin/tsg_update_tags

-* * * * * sleep 10; /opt/tsg/bin/tsg_update_tags

-* * * * * sleep 20; /opt/tsg/bin/tsg_update_tags

-* * * * * sleep 30; /opt/tsg/bin/tsg_update_tags

-* * * * * sleep 40; /opt/tsg/bin/tsg_update_tags

-* * * * * sleep 50; /opt/tsg/bin/tsg_update_tags

-

-修改后要执行: service crond restart
\ No newline at end of file
diff --git a/deploy_doc/cli部署.txt b/deploy_doc/oam_cli_deply.md
index 5e4a4d6..5e4a4d6 100644
--- a/deploy_doc/cli部署.txt
+++ b/deploy_doc/oam_cli_deply.md
diff --git a/deploy_doc/tsg_cli_deploy.md b/deploy_doc/tsg_cli_deploy.md
new file mode 100644
index 0000000..cacaf5e
--- /dev/null
+++ b/deploy_doc/tsg_cli_deploy.md
@@ -0,0 +1,25 @@
+1.交换板文件

+  /opt/tsg/bin

+  /opt/tsg/tools

+  /opt/tsg/etc

+  

+  更新telegraf.conf的global tags, 本机sn.

+

+2.计算板文件(4台)

+  

+

+3.服务

+启动交换板crontab服务:

+service crond start

+chkconfig crond on

+crontab -e, 输入以下内容:

+* * * * * /opt/tsg/tools/tsg_update_tags

+* * * * * /opt/tsg/tools/tsg_monit_interface

+* * * * * /opt/tsg/tools/tsg_monit_stream

+* * * * * sleep 10; /opt/tsg/tools/tsg_update_tags

+* * * * * sleep 20; /opt/tsg/tools/tsg_update_tags

+* * * * * sleep 30; /opt/tsg/tools/tsg_update_tags

+* * * * * sleep 40; /opt/tsg/tools/tsg_update_tags

+* * * * * sleep 50; /opt/tsg/tools/tsg_update_tags

+

+修改后要执行: service crond restart
\ No newline at end of file
diff --git a/deploy_etc/tsg_module_deploy.json b/deploy_etc/tsg_module_deploy.json
new file mode 100644
index 0000000..0afcdb3
--- /dev/null
+++ b/deploy_etc/tsg_module_deploy.json
@@ -0,0 +1,12 @@
+{
+	"modules_deploy": {
+		"mcn0": ["kni", "a.out"],
+		"mcn1": ["tfe", "a1.out"],
+		"mcn2": ["tfe", "a2.out"],
+		"mcn3": ["tfe", "a3.out"]
+	},
+	"modules_operator": {
+		"kni": ["sapp", "r3 sapp", "/home/tsg/kni", "r2", "killall", "exec", "ps"],
+		"telegraf": ["teleraf", "#", "#", "#", "systemctl_stop", "systemctl_start", "systemctl_status"]
+	}
+}
\ No newline at end of file
diff --git a/deploy_etc/tsg_modules_operator.json.bak b/deploy_etc/tsg_modules_operator.json.bak
new file mode 100644
index 0000000..273e9fa
--- /dev/null
+++ b/deploy_etc/tsg_modules_operator.json.bak
@@ -0,0 +1,6 @@
+{
+	"modules_operator": {
+		"sapp": ["sapp", "r3", "/home/tsg/kni", "r2", "killall", "exec", "ps"],
+		"telegraf": ["teleraf", "#", "#", "#", "systemctl_stop", "systemctl_start", "systemctl_status"]
+	}
+}
\ No newline at end of file
diff --git a/py_cmd/.gitignore b/py_cmd/.gitignore
new file mode 100644
index 0000000..bee8a64
--- /dev/null
+++ b/py_cmd/.gitignore
@@ -0,0 +1 @@
+__pycache__
diff --git a/py_cmd/common_modules_deploy.json b/py_cmd/common_modules_deploy.json
deleted file mode 100644
index 786644f..0000000
--- a/py_cmd/common_modules_deploy.json
+++ /dev/null
@@ -1,12 +0,0 @@
-#此文件定义四块计算板分别运行什么模块

-#先根据tsg_chassis_ip.json, 根据当前设备的ip地址, 知道当前设备的sled名称

-#然后根据设备类型名称, 找到当前运行的所有模块(应用)名称

-#然后根据模块名称, 到common_modules_operator.json查找每个模块的操作方法

-{

-    "modules_dispatch":

-     

-       "mcn0": ["kni"]

-       "mcn1": ["tfe"]

-        "mcn2": ["tfe"]

-        "mcn3": ["tfe"]

-}
\ No newline at end of file
diff --git a/py_cmd/common_modules_operator.json b/py_cmd/common_modules_operator.json
deleted file mode 100644
index 1bd6a16..0000000
--- a/py_cmd/common_modules_operator.json
+++ /dev/null
@@ -1,26 +0,0 @@
-#此文件定义, 每个模块的启动、停止、检查运行状态的方法

-{

-   "modules":[

-   {sapp:

-   start: 

-       "cwd"

-       "exe"

-   stop:

-       "killall -9 r3 sapp"

-   check healthy:

-        "ps -ef | grep sapp "

-   }

-   

-    {telegraf:

-   start: 

-       "cwd" : ""

-       "exe" : "service influxdb restart"

-   stop:

-       "service influxdb stop"

-   check healthy:

-        "service influxdb status | grep Active "

-   }  

-   

-   ]

-

-}
\ No newline at end of file
diff --git a/py_cmd/tsg_monit_device.py b/py_cmd/tsg_monit_device.py.bak
index e8d0e63..e8d0e63 100644
--- a/py_cmd/tsg_monit_device.py
+++ b/py_cmd/tsg_monit_device.py.bak
diff --git a/py_cmd/tsg_monit_device_by_telegraf.py b/py_cmd/tsg_monit_device_by_telegraf.py.bak
index f2366a2..f2366a2 100644
--- a/py_cmd/tsg_monit_device_by_telegraf.py
+++ b/py_cmd/tsg_monit_device_by_telegraf.py.bak
diff --git a/py_cmd/tsg_monit_stream.py b/py_cmd/tsg_monit_stream.py.bak
index c34f217..c34f217 100644
--- a/py_cmd/tsg_monit_stream.py
+++ b/py_cmd/tsg_monit_stream.py.bak
diff --git a/py_cmd/tsg_software_reboot.py b/py_cmd/tsg_software_reboot.py
index efb592c..6320e83 100644
--- a/py_cmd/tsg_software_reboot.py
+++ b/py_cmd/tsg_software_reboot.py
@@ -5,6 +5,11 @@ import syslog
 import subprocess
 import time
 import re
+from sys import path
+path.append(r'../py_common') #将存放module的路径添加进来
+from common_modules_operator import *
+from common_whoami import *
+from common_modules_deploy import *
 
 ##define KERN_EMERG    "<0>"  /* system is unusable               */
 ##define KERN_ALERT    "<1>"  /* action must be taken immediately */
@@ -15,7 +20,7 @@ import re
 ##define KERN_INFO     "<6>"  /* informational                    */
 ##define KERN_DEBUG    "<7>"  /* debug-level messages             */
 
-MSG_PREFIX = ['EMERG', 'ALERT', 'CRIT', 'ERR', 'WARNING', 'NOTICE', 'INFO', 'DEBUG']
+G_SYS_LOG_STRING = ['EMERG', 'ALERT', 'CRIT', 'ERR', 'WARNING', 'NOTICE', 'INFO', 'DEBUG']
 
 G_LOCAL_NODE_NAME = "TSG_MXN"
 
@@ -23,18 +28,18 @@ class CommandException(Exception):
     pass
     
 def tsg_restart_err_log(error_num, user_msg):
-    msg = "[%s] %s" %(MSG_PREFIX[syslog.LOG_ERR], user_msg)
+    msg = "[%s] %s" %(G_SYS_LOG_STRING[syslog.LOG_ERR], user_msg)
     syslog.syslog(syslog.LOG_ERR, msg) 
     print (msg)
     
-    msg = "[%s] %s" %(MSG_PREFIX[syslog.LOG_ERR], "tsg software reboot error")
+    msg = "[%s] %s" %(G_SYS_LOG_STRING[syslog.LOG_ERR], "tsg software reboot error")
     syslog.syslog(syslog.LOG_ERR, msg)
     print (msg)
     
     sys.exit(error_num)
 
 def tsg_restart_succ_log():
-    msg = "[%s] %s" %(MSG_PREFIX[syslog.LOG_NOTICE], "tsg software reboot success")
+    msg = "[%s] %s" %(G_SYS_LOG_STRING[syslog.LOG_NOTICE], "tsg software reboot success")
     syslog.syslog(syslog.LOG_NOTICE, msg)
     print (msg)
     sys.exit(0)
@@ -97,36 +102,20 @@ def tsg_kill_app_process_by_killall(module_name, extra_progs):
     return 0    
 
 def tsg_kill_app_process_by_systemctl_stop(module_name):
-    print("TODO")
+    print("tsg_kill_app_process_by_systemctl_stop  TODO!!!")
     sys.exit(1)
     return 0
     
-def tsg_stop_app_process(module_name, extra_progs, stop_method, check_method):
+def tsg_stop_app_process(module_name, extra_progs, stop_func, check_func):
     res_code = 0
     running_flag = 0
     
-    if check_method == 'ps':
-        check_func = tsg_check_process_health_by_ps
-    elif check_method == 'systemctl_status':
-        check_func = tsg_check_process_health_by_systemctl_status
-    else:
-        print("not support check method:%s" %(check_method))
-        sys.exit(1)
-    
     running_flag = check_func(module_name)    
         
     if running_flag == 0:
-        #print("%s is not running, start it..." %(module_name))
+        logger.debug("%s is not running, start it..." %(module_name))
         return 0
     #ready to stop progcess,  retry for 3 times
-
-    if stop_method == 'killall':
-        stop_func = tsg_kill_app_process_by_killall
-    elif stop_method == 'systemctl_stop':
-        stop_func = tsg_kill_app_process_by_systemctl_stop
-    else:
-        print("not support stop method:%s" %(check_method))
-        sys.exit(1)
     
     for times in range(3):
         stop_func(module_name, extra_progs)
@@ -140,55 +129,42 @@ def tsg_stop_app_process(module_name, extra_progs, stop_method, check_method):
             
     if res_code != 0: 
         errmsg = "can't stop process %s" %(module_name)
+        logger.error(errmsg)
         tsg_restart_err_log(res_code, errmsg)
     
     return res_code
 
-def tsg_start_app_process_by_exec_call(module_name, module_cwd, module_exe, check_method):
+def tsg_start_app_process_by_exec_call(module_name, module_cwd, module_exe):
+    logger.debug("try cd to dir:%s" %(module_cwd))
     try:
         os.chdir(module_cwd)
     except Exception as e: 
         print("%s" %(e))
         return 1
     
+    logger.debug("try call exec :%s" %(module_exe))
     cmd_str = "./%s" %(module_exe)
     ret_code, output = system_cmd_run(cmd_str)
     if ret_code != 0:
         errmsg = "start program %s error, call %s/%s failed" %(module_name, module_cwd, module_exe)
         tsg_restart_err_log(ret_code, errmsg)
-    
-    if check_method == 'ps':
-        check_func = tsg_check_process_health_by_ps
-    elif check_method == 'systemctl_status':
-        check_func = tsg_check_process_health_by_systemctl_status
-    else:
-        print("not support check method:%s" %(check_method))
-        return 1
-    
-    running_flag = check_func(module_name)    
-    if running_flag == 0:
-        errmsg = "start process %s error" %(module_name)
-        tsg_restart_err_log(ret_code, errmsg)
- 
+
     return 0
     
 def tsg_start_app_process_by_systemctl_start(module_name, module_cwd, module_exe, check_method):
-    print("TODO")
+    print("tsg_start_app_process_by_systemctl_start  TODO!!!!!")
     sys.exit(1)
 
-def tsg_start_app_process(module_name, module_cwd, module_exe, start_method, check_method):
-    if start_method == 'exec_call':
-        start_func = tsg_start_app_process_by_exec_call
-    elif start_method == 'systemctl_start':
-        start_func = tsg_start_app_process_by_systemctl_start
-    else:
-        print("not support start method:%s" %(start_method))
-        sys.exit(1)
-
-    ret = start_func(module_name, module_cwd, module_exe, check_method)        
+def tsg_start_app_process(module_name, module_cwd, module_exe, start_func, check_func):
+    ret = start_func(module_name, module_cwd, module_exe)        
     if ret != 0:
-        sys.exit(1) 
+        return 1
 
+    running_flag = check_func(module_name)    
+    if running_flag == 0:
+        errmsg = "start process %s error" %(module_name)
+        return 1
+ 
     return 0
     
 #参数说明:
@@ -201,25 +177,82 @@ def tsg_start_app_process(module_name, module_cwd, module_exe, start_method, che
 # start_method: 启动应用方法  
 # check_method: 检测应用是否运行方法  
 #  
-def tsg_restart_app_process(module_name, extra_progs, module_cwd, module_exe, stop_method, start_method, check_method):
+def tsg_restart_app_process(module_name, extra_progs, module_cwd, module_exe, start_func, stop_func, check_func):
     res_code = 0
     
-    res_code = tsg_stop_app_process(module_name, extra_progs, stop_method, check_method)
+    res_code = tsg_stop_app_process(module_name, extra_progs, stop_func, check_func)
     if res_code != 0:
         return res_code
         
-    res_code = tsg_start_app_process(module_name, module_cwd, module_exe, start_method, check_method)
+    res_code = tsg_start_app_process(module_name, module_cwd, module_exe, start_func, check_func)
     if res_code != 0:
         return res_code
     
     return 0
+
+#根据配置文件的参数, 选择用那种操作函数继续下一步
+#返回值, 函数指针:
+#ret_code, start_fun, stop_fun, check_health_fun    
+def tsg_get_operator_by_config(module_operator):    
+    if module_operator[TSG_OP_MODULE_START_INDEX] == 'exec':
+        start_func = tsg_start_app_process_by_exec_call
+    elif module_operator[TSG_OP_MODULE_START_INDEX] == 'systemctl_start':
+        start_func = tsg_start_app_process_by_systemctl_start
+    else:
+        errmsg = "not support start method:%s, only be [exec, systemctl_start]" %(module_operator[TSG_OP_MODULE_START_INDEX])
+        tsg_restart_err_log(1, errmsg)
+
+    if module_operator[TSG_OP_MODULE_STOP_INDEX] == 'killall':
+        stop_func = tsg_kill_app_process_by_killall
+    elif module_operator[TSG_OP_MODULE_STOP_INDEX] == 'systemctl_stop':
+        stop_func = tsg_kill_app_process_by_systemctl_stop
+    else:
+        errmsg = "not support stop method:%s, only be [killall, systemctl_stop]" %(module_operator[TSG_OP_MODULE_STOP_INDEX])
+        tsg_restart_err_log(1, errmsg)
+
+    if module_operator[TSG_OP_MODULE_STATUS_INDEX] == 'ps':
+        check_func = tsg_check_process_health_by_ps
+    elif module_operator[TSG_OP_MODULE_STATUS_INDEX] == 'systemctl_status':
+        check_func = tsg_check_process_health_by_systemctl_status
+    else:
+        errmsg = "not support check method:%s, only be [ps, systemctl_status]" %(module_operator[TSG_OP_MODULE_STATUS_INDEX])
+        tsg_restart_err_log(1, errmsg)
+    
+    return 0, start_func, stop_func, check_func
     
 def tsg_software_reboot():
     #G_LOCAL_NODE_NAME = get_local_node_name()
     
     log_handle = syslog.openlog(G_LOCAL_NODE_NAME)
-
-    tsg_restart_app_process("sapp", "r3", "/home/tsg/kni", "r2", "killall", "exec_call", "ps")
+    
+    sled_type,sled_id,sled_name = tsg_whoami()
+    if sled_name == "":
+        tsg_restart_err_log(1, "can't get local sled name")
+        sys.exit(1)
+    
+    module_array = tsg_get_local_sled_modules(sled_name)    
+    if len(module_array) <= 0:
+        tsg_restart_err_log(1, "can't get local sled modules")
+        sys.exit(1)
+    
+    logger.debug("len(module_array) = %d" %(len(module_array)))  
+    for module_name in module_array:
+        module_operator = tsg_get_module_opertor(module_name)
+        if len(module_operator) <= 0:
+            tsg_restart_err_log(1, "can't get local sled module operator for %s" %(module_name))
+            sys.exit(1)
+        
+        ret, start_func, stop_func, check_func =  tsg_get_operator_by_config(module_operator)    
+        if ret != 0:
+            tsg_restart_err_log(1, "can't get operator for %s" %(module_operator[TSG_OP_MODULE_NAME_INDEX]))
+            sys.exit(1)
+        
+        tsg_restart_app_process(module_operator[TSG_OP_MODULE_NAME_INDEX], module_operator[TSG_OP_MODULE_EXTRA_INDEX], 
+                                module_operator[TSG_OP_MODULE_CWD_INDEX], 
+                                module_operator[TSG_OP_MODULE_EXE_INDEX],   
+                                start_func, stop_func, check_func)
+        
+    #tsg_restart_app_process("sapp", "r3", "/home/tsg/kni", "r2", "killall", "exec_call", "ps")
     #tsg_restart_app_process("telegraf", "systemctl_stop", "systemctl_start", "systemctl_status")
     #tsg_restart_app_process("marsio", "systemctl_stop", "systemctl_start", "systemctl_status")
     #tsg_restart_app_process("influxd", "systemctl_stop", "systemctl_start", "systemctl_status")
@@ -227,4 +260,11 @@ def tsg_software_reboot():
     tsg_restart_succ_log()
 
 if __name__ == '__main__':
+    global logger
+    
+    if len(sys.argv) >= 2 and sys.argv[1] == "debug":
+        logger = logger_init(logging.DEBUG)
+    else:
+        logger = logger_init(logging.CRITICAL)
+        
     tsg_software_reboot()
diff --git a/py_cmd/tsg_software_reboot.py.bak b/py_cmd/tsg_software_reboot.py.bak
new file mode 100644
index 0000000..efb592c
--- /dev/null
+++ b/py_cmd/tsg_software_reboot.py.bak
@@ -0,0 +1,230 @@
+#coding=utf-8
+import os
+import sys
+import syslog
+import subprocess
+import time
+import re
+
+##define KERN_EMERG    "<0>"  /* system is unusable               */
+##define KERN_ALERT    "<1>"  /* action must be taken immediately */
+##define KERN_CRIT     "<2>"  /* critical conditions              */
+##define KERN_ERR      "<3>"  /* error conditions                 */
+##define KERN_WARNING  "<4>"  /* warning conditions               */
+##define KERN_NOTICE   "<5>"  /* normal but significant condition */
+##define KERN_INFO     "<6>"  /* informational                    */
+##define KERN_DEBUG    "<7>"  /* debug-level messages             */
+
+MSG_PREFIX = ['EMERG', 'ALERT', 'CRIT', 'ERR', 'WARNING', 'NOTICE', 'INFO', 'DEBUG']
+
+G_LOCAL_NODE_NAME = "TSG_MXN"
+
+class CommandException(Exception):
+    pass
+    
+def tsg_restart_err_log(error_num, user_msg):
+    msg = "[%s] %s" %(MSG_PREFIX[syslog.LOG_ERR], user_msg)
+    syslog.syslog(syslog.LOG_ERR, msg) 
+    print (msg)
+    
+    msg = "[%s] %s" %(MSG_PREFIX[syslog.LOG_ERR], "tsg software reboot error")
+    syslog.syslog(syslog.LOG_ERR, msg)
+    print (msg)
+    
+    sys.exit(error_num)
+
+def tsg_restart_succ_log():
+    msg = "[%s] %s" %(MSG_PREFIX[syslog.LOG_NOTICE], "tsg software reboot success")
+    syslog.syslog(syslog.LOG_NOTICE, msg)
+    print (msg)
+    sys.exit(0)
+
+#return exitcode value + output message:
+#    0: succ
+#    1: error
+def system_cmd_run(cmd_str):
+    dangerous_cmd = {"rm", "mv", "poweroff", "shutdown"}
+    
+    for cmd in dangerous_cmd:
+        pattern = "\s*%s" %(cmd)
+        match_str = re.match(pattern, cmd_str)
+        if not match_str is None:
+            print("can't run this cmd:%s" %(cmd_str))
+            sys.exit(1)
+            
+    try:
+        exitcode, output = subprocess.getstatusoutput(cmd_str)
+    except Exception as e:  
+        print(e)    
+        print("###### %s" %(e.message))
+        #if exitcode != 0:
+        #    output = ""
+        return 1, e.message
+
+    return exitcode, output
+    
+#return value:
+# 1: progcess of prog_name is exist
+# 0: progcess of prog_name is not exist  
+def tsg_check_process_health_by_ps(module_name):
+    cmd_str = "ps -afx | grep %s | grep -v grep" %(module_name)
+    exitcode, output = system_cmd_run(cmd_str)
+    if exitcode == 0:
+        return 1
+    
+    return 0
+
+#return value:
+# 1: progcess of prog_name is exist
+# 0: progcess of prog_name is not exist    
+def tsg_check_process_health_by_systemctl_status(module_name):
+    print("systemctl_status check method TODO!")
+    sys.exit(1)
+    return 0
+
+def tsg_kill_app_process_by_killall(module_name, extra_progs):
+    #todo , stop sapp, xxx, check process exist or not, maybe zombie, maybe very slow
+    command = "killall -9 %s %s" %(module_name, extra_progs)
+    try:    
+        exitcode, output = subprocess.getstatusoutput(command)
+        #print("%d" %(exitcode))
+    except Exception as e:
+        pass
+    
+    if exitcode != 0:
+        return 1
+        
+    return 0    
+
+def tsg_kill_app_process_by_systemctl_stop(module_name):
+    print("TODO")
+    sys.exit(1)
+    return 0
+    
+def tsg_stop_app_process(module_name, extra_progs, stop_method, check_method):
+    res_code = 0
+    running_flag = 0
+    
+    if check_method == 'ps':
+        check_func = tsg_check_process_health_by_ps
+    elif check_method == 'systemctl_status':
+        check_func = tsg_check_process_health_by_systemctl_status
+    else:
+        print("not support check method:%s" %(check_method))
+        sys.exit(1)
+    
+    running_flag = check_func(module_name)    
+        
+    if running_flag == 0:
+        #print("%s is not running, start it..." %(module_name))
+        return 0
+    #ready to stop progcess,  retry for 3 times
+
+    if stop_method == 'killall':
+        stop_func = tsg_kill_app_process_by_killall
+    elif stop_method == 'systemctl_stop':
+        stop_func = tsg_kill_app_process_by_systemctl_stop
+    else:
+        print("not support stop method:%s" %(check_method))
+        sys.exit(1)
+    
+    for times in range(3):
+        stop_func(module_name, extra_progs)
+        #此处不判断stop_func的返回值, 可能程序不存在, 可能守护不存在,等原因
+        #直接用check()方法检测stop()的成功
+        res_code = tsg_check_process_health_by_ps(module_name)
+        if res_code != 0:
+            continue
+        else:
+            break    
+            
+    if res_code != 0: 
+        errmsg = "can't stop process %s" %(module_name)
+        tsg_restart_err_log(res_code, errmsg)
+    
+    return res_code
+
+def tsg_start_app_process_by_exec_call(module_name, module_cwd, module_exe, check_method):
+    try:
+        os.chdir(module_cwd)
+    except Exception as e: 
+        print("%s" %(e))
+        return 1
+    
+    cmd_str = "./%s" %(module_exe)
+    ret_code, output = system_cmd_run(cmd_str)
+    if ret_code != 0:
+        errmsg = "start program %s error, call %s/%s failed" %(module_name, module_cwd, module_exe)
+        tsg_restart_err_log(ret_code, errmsg)
+    
+    if check_method == 'ps':
+        check_func = tsg_check_process_health_by_ps
+    elif check_method == 'systemctl_status':
+        check_func = tsg_check_process_health_by_systemctl_status
+    else:
+        print("not support check method:%s" %(check_method))
+        return 1
+    
+    running_flag = check_func(module_name)    
+    if running_flag == 0:
+        errmsg = "start process %s error" %(module_name)
+        tsg_restart_err_log(ret_code, errmsg)
+ 
+    return 0
+    
+def tsg_start_app_process_by_systemctl_start(module_name, module_cwd, module_exe, check_method):
+    print("TODO")
+    sys.exit(1)
+
+def tsg_start_app_process(module_name, module_cwd, module_exe, start_method, check_method):
+    if start_method == 'exec_call':
+        start_func = tsg_start_app_process_by_exec_call
+    elif start_method == 'systemctl_start':
+        start_func = tsg_start_app_process_by_systemctl_start
+    else:
+        print("not support start method:%s" %(start_method))
+        sys.exit(1)
+
+    ret = start_func(module_name, module_cwd, module_exe, check_method)        
+    if ret != 0:
+        sys.exit(1) 
+
+    return 0
+    
+#参数说明:
+#
+# module_name: 模块名称
+# extra_progs: 其他需要kill的附加程序, 如sapp的r3守护, 需要杀掉, 否则后台可能会重复启动sapp
+# module_cwd:  应用的绝对路径
+# module_exe: 启动应用的名称, 可能跟module_name不一样, 比如用r2启动sapp
+# stop_method: 停止应用方法  
+# start_method: 启动应用方法  
+# check_method: 检测应用是否运行方法  
+#  
+def tsg_restart_app_process(module_name, extra_progs, module_cwd, module_exe, stop_method, start_method, check_method):
+    res_code = 0
+    
+    res_code = tsg_stop_app_process(module_name, extra_progs, stop_method, check_method)
+    if res_code != 0:
+        return res_code
+        
+    res_code = tsg_start_app_process(module_name, module_cwd, module_exe, start_method, check_method)
+    if res_code != 0:
+        return res_code
+    
+    return 0
+    
+def tsg_software_reboot():
+    #G_LOCAL_NODE_NAME = get_local_node_name()
+    
+    log_handle = syslog.openlog(G_LOCAL_NODE_NAME)
+
+    tsg_restart_app_process("sapp", "r3", "/home/tsg/kni", "r2", "killall", "exec_call", "ps")
+    #tsg_restart_app_process("telegraf", "systemctl_stop", "systemctl_start", "systemctl_status")
+    #tsg_restart_app_process("marsio", "systemctl_stop", "systemctl_start", "systemctl_status")
+    #tsg_restart_app_process("influxd", "systemctl_stop", "systemctl_start", "systemctl_status")
+    
+    tsg_restart_succ_log()
+
+if __name__ == '__main__':
+    tsg_software_reboot()
diff --git a/py_tools/common_args.py b/py_common/common_args.py
index 52b93de..406a5b4 100644
--- a/py_tools/common_args.py
+++ b/py_common/common_args.py
@@ -1,16 +1,16 @@
-#coding=utf-8

-import argparse

-

-def setup_common_args():

-    parser = argparse.ArgumentParser(description='TSG OAM Argument Parser')

-

-    parser.add_argument('-g', '--debug', help = 'debug mode, default is disable',

-                        action='store_true', default = 0)

-    parser.add_argument('-l', '--log-level', help = 'debug log level, support:10,20,30,40,50, default is:30',

-                        type=int, default = 30)                       

-    parser.add_argument('--telegraf-ip', help = 'send log to telegraf ip address, default is:192.168.200.5',

-                        type=str, default = '192.168.200.5')

-    parser.add_argument('--telegraf-port', help = 'send log to telegraf port, default is:8126',

-                        type=int, default = 8126)

- 

+#coding=utf-8
+import argparse
+
+def setup_common_args():
+    parser = argparse.ArgumentParser(description='TSG OAM Argument Parser')
+
+    parser.add_argument('-g', '--debug', help = 'debug mode, default is disable',
+                        action='store_true', default = 0)
+    parser.add_argument('-l', '--log-level', help = 'debug log level, support:10,20,30,40,50, default is:30',
+                        type=int, default = 30)                       
+    parser.add_argument('--telegraf-ip', help = 'send log to telegraf ip address, default is:192.168.200.5',
+                        type=str, default = '192.168.200.5')
+    parser.add_argument('--telegraf-port', help = 'send log to telegraf port, default is:8126',
+                        type=int, default = 8126)
+ 
     return parser
\ No newline at end of file
diff --git a/py_common/common_diagnose.py b/py_common/common_diagnose.py
new file mode 100644
index 0000000..25f8234
--- /dev/null
+++ b/py_common/common_diagnose.py
@@ -0,0 +1,171 @@
+# coding: utf-8
+#diagnose公共模块, 供cli命令和backgroud后台自动运行工具调用
+import sys
+import time
+import json
+import logging
+import syslog
+import prettytable
+from sys import path
+path.append(r'../py_common') #将存放module的路径添加进来
+path.append(r'../py_cmd') #将存放module的路径添加进来
+from common_telegraf import *
+from common_logger import *
+from common_args import *
+from common_logger import *
+from common_influxdb import *
+from common_whoami import *
+from common_system_cmd import *
+from tsg_software_reboot import *
+
+#syslog 级别:
+#define KERN_EMERG    "<0>"  /* system is unusable               */
+#define KERN_ALERT    "<1>"  /* action must be taken immediately */
+#define KERN_CRIT     "<2>"  /* critical conditions              */
+#define KERN_ERR      "<3>"  /* error conditions                 */
+#define KERN_WARNING  "<4>"  /* warning conditions               */
+#define KERN_NOTICE   "<5>"  /* normal but significant condition */
+#define KERN_INFO     "<6>"  /* informational                    */
+#define KERN_DEBUG    "<7>"  /* debug-level messages             */
+#参数log_level表示高于此级别的才输出, syslog值越小优先级越高, 
+#即log_level <= diagnose_level时输出结果, 如果级别高于ERR, 同时写入syslog
+
+#如果没有任何警告和错误, 显示normal
+#所有检测项不能中途退出, 即使有错误, 显示当前检测项的错误后, 继续, 保证全检测一遍
+
+#根据优先级, 设置新的日志等级
+def tsg_set_log_level(old_level, new_level):
+    if new_level < old_level:
+        return new_level
+    else:
+        return old_level
+
+
+def tsg_diagnose_for_app(log_level):
+    cur_level = syslog.LOG_INFO
+    err_code = 0
+    
+    sled_type,sled_id,sled_name = tsg_whoami()
+    if sled_name == "":
+        print("can't get local sled name")
+        return 1
+    
+    module_array = tsg_get_local_sled_modules(sled_name)    
+    if len(module_array) <= 0:
+        cur_level = tsg_set_log_level(cur_level, syslog.LOG_ERR)
+        ptable.add_row([g_local_sled_name,"common", G_SYS_LOG_STRING[cur_level],"can't get local sled modules"])
+        #print("can't get local sled modules")
+        return 1
+     
+    for module_name in module_array:
+        module_operator = tsg_get_module_opertor(module_name)
+        if len(module_operator) <= 0:
+            cur_level = tsg_set_log_level(cur_level, syslog.LOG_ERR)
+            err_msg = "can't get local sled module operator for %s" %(module_name)
+            ptable.add_row([g_local_sled_name,"common", G_SYS_LOG_STRING[cur_level], err_msg])
+            #print(err_msg)
+            err_code += 1
+        
+        ret, start_func, stop_func, check_func =  tsg_get_operator_by_config(module_operator)    
+        if ret != 0:
+            cur_level = tsg_set_log_level(cur_level, syslog.LOG_ERR)
+            err_msg = "can't get operator for %s" %(module_operator[TSG_OP_MODULE_NAME_INDEX])
+            ptable.add_row([g_local_sled_name,"common", G_SYS_LOG_STRING[cur_level],err_msg])
+            #print(err_msg)
+            err_code += 1
+        
+        running_flag = check_func(module_operator[TSG_OP_MODULE_NAME_INDEX])    
+        if running_flag == 0:
+            cur_level = tsg_set_log_level(cur_level, syslog.LOG_EMERG)
+            err_msg = "app module %s is not running" %(module_name)
+            ptable.add_row([g_local_sled_name,"app", G_SYS_LOG_STRING[cur_level],err_msg])
+            err_code += 1
+    
+    #todo,  check for restart time
+    if cur_level >= syslog.LOG_INFO:
+        ptable.add_row([g_local_sled_name,"app", G_SYS_LOG_STRING[cur_level], "normal"])    
+    
+    return err_code
+
+def tsg_diagnose_for_cpu(log_level):
+    cur_level = syslog.LOG_INFO
+    ptable.add_row([g_local_sled_name,"cpu", G_SYS_LOG_STRING[cur_level],"normal"])   
+    return 0
+
+def tsg_diagnose_for_mem(log_level):
+    cur_level = syslog.LOG_INFO
+    ptable.add_row([g_local_sled_name,"memory", G_SYS_LOG_STRING[cur_level],"normal"])   
+    return 0
+
+def tsg_diagnose_for_disk(log_level):
+    cur_level = syslog.LOG_INFO
+    
+    ret, cmd_res = system_cmd_run("df | awk {'print $1,$5'}")
+    
+    print(cmd_res.split())
+    
+    print(len(cmd_res.split()))
+    print(cmd_res.split()[0])
+    print(cmd_res.split()[1])
+    print(cmd_res.split()[2])
+    
+    ptable.add_row([g_local_sled_name,"disk", G_SYS_LOG_STRING[cur_level],"normal"])  
+    return 0
+
+def tsg_diagnose_for_interface(log_level):
+    cur_level = syslog.LOG_INFO
+    sql_str = "select * from interface where PhyRXBits+PhyRXError+PhyRXMissed+PhyRXNoBUF > 0 and time > now() -5m limit 1"
+
+    ret, points, msg = tsg_influxb_query(influxdb_client, sql_str)
+    if ret == 0:
+        for point in points:
+            cur_level = tsg_set_log_level(cur_level, syslog.LOG_WARNING)
+            msg = "%s has droped %d packets" %(point['device'], int(point['PhyRXBits']) + int(point['PhyRXMissed']) + int(point['PhyRXNoBUF']) )
+            ptable.add_row([g_local_sled_name,"physical network", G_SYS_LOG_STRING[cur_level], msg])  
+ 
+    if cur_level >= syslog.LOG_INFO:
+        ptable.add_row([g_local_sled_name,"physical network", G_SYS_LOG_STRING[cur_level], "normal"])  
+        return 0
+        
+    return 1
+
+def tsg_diagnose_for_app_stream(log_level):
+    cur_level = syslog.LOG_INFO
+    ptable.add_row([g_local_sled_name,"app network", G_SYS_LOG_STRING[cur_level],"normal"])  
+    return 0
+ 
+    
+def tsg_common_diagnose(log_level):
+    #检查app进程是否存在, 是否最近n分钟内重启过
+    tsg_diagnose_for_app(log_level)
+    
+    #检查CPU占用率
+    tsg_diagnose_for_cpu(log_level)
+    
+    #检查mem占用率
+    tsg_diagnose_for_mem(log_level)
+    
+    #检查磁盘占用率
+    tsg_diagnose_for_disk(log_level)
+    
+    #检查物理网络情况
+    tsg_diagnose_for_interface(log_level)
+    
+    #检查应用流量
+    tsg_diagnose_for_app_stream(log_level)
+
+if __name__ == '__main__':
+    global ptable
+    global influxdb_client
+    global g_local_sled_name
+    type, id, g_local_sled_name = tsg_whoami()
+    
+    ptable = prettytable.PrettyTable()
+    ptable.field_names = ["Sled", "Type", "Level", "Status"]
+    
+    ret, influxdb_client, msg = tsg_influxdb_init('127.0.0.1', 8086, 'admin', 'tsg2019', 'tsg_stat')
+    if ret != 0:
+        ptable.add_row([g_local_sled_name,"common", G_SYS_LOG_STRING[syslog.LOG_ERR],"can't connect influxDB server"])
+    
+    tsg_common_diagnose(syslog.LOG_DEBUG)
+    print(ptable)
\ No newline at end of file
diff --git a/py_tools/common_get_tags.py b/py_common/common_get_tags.py
index f347171..7e3d148 100644
--- a/py_tools/common_get_tags.py
+++ b/py_common/common_get_tags.py
@@ -1,30 +1,30 @@
-import json

-

-SN_JSON_PATH = '/opt/tsg/etc/tsg_tags.json'

-

-def tsg_get_tags_json():

-    try:

-        with open(SN_JSON_PATH) as json_fp:

-            json_dict = json.load(json_fp)

-            return json_dict['tags']

-    except IOError:

-        return ""

-

-def tsg_get_tags():

-    tags = {}

-    json_fp = tsg_get_tags_json()  

-    if len(json_fp) > 0:

-        for tag_item in json_fp:

-            #print(tag_item)

-            tags[tag_item['tag']] = tag_item['value']

-    

-    return tags

-    

-def main():

-    tags = {}

-    

-    tags = tsg_get_tags()

-    print(tags)

-        

-if __name__ == '__main__':

+import json
+
+SN_JSON_PATH = '/opt/tsg/etc/tsg_tags.json'
+
+def tsg_get_tags_json():
+    try:
+        with open(SN_JSON_PATH) as json_fp:
+            json_dict = json.load(json_fp)
+            return json_dict['tags']
+    except IOError:
+        return ""
+
+def tsg_get_tags():
+    tags = {}
+    json_fp = tsg_get_tags_json()  
+    if len(json_fp) > 0:
+        for tag_item in json_fp:
+            #print(tag_item)
+            tags[tag_item['tag']] = tag_item['value']
+    
+    return tags
+    
+def main():
+    tags = {}
+    
+    tags = tsg_get_tags()
+    print(tags)
+        
+if __name__ == '__main__':
     main()
\ No newline at end of file
diff --git a/py_common/common_influxdb.py b/py_common/common_influxdb.py
new file mode 100644
index 0000000..c8ca799
--- /dev/null
+++ b/py_common/common_influxdb.py
@@ -0,0 +1,41 @@
+# coding: utf-8
+
+import sys
+from influxdb import InfluxDBClient
+
+
+#return value
+#ret, result_array, message
+def tsg_influxb_query(influxdb_client, sql_str):
+    try:
+        result = influxdb_client.query(sql_str)
+    except Exception as e:
+        return 1, {}, e
+        
+    return 0, result.get_points(), "succ"
+
+#return value:
+#ret, client, error_msg
+def tsg_influxdb_init(arg_host,arg_port,arg_username,arg_password,arg_db):
+    try:
+        influxdb_client = InfluxDBClient(host=arg_host,port=arg_port,username=arg_username,password=arg_password,database=arg_db, timeout=3) 
+    except Exception as e:
+        print("InfluxDBClient error: %s" %(e))
+        return 1, None, e
+    
+    return 0, influxdb_client, "succ"
+
+if __name__ == '__main__':
+    ret, influxdb_client, msg = tsg_influxdb_init('127.0.0.1', 8086, 'admin', 'tsg2019', 'tsg_stat')
+    if ret != 0:
+        sys.exit(1)
+        
+    ret, points,msg = tsg_influxb_query(influxdb_client, "select * from interface order by time desc limit 1")
+    if ret == 0:
+        print("--for xxx in points:-----------------")
+        for point in points:
+            print("--------%s, %s" %(point['PhyRXBits'], point['PhyRXMissed']))
+
+    else:
+        print("query error, %s" %(res))
+
diff --git a/py_common/common_json.py b/py_common/common_json.py
new file mode 100644
index 0000000..61aec60
--- /dev/null
+++ b/py_common/common_json.py
@@ -0,0 +1,20 @@
+# coding: utf-8
+
+import sys
+import psutil
+import time
+import json
+
+#return value:
+# 0, error msg, json_dict : succ
+# 1, error msg, "" : error
+def tsg_json_parse(file_name):
+    try:
+        with open(file_name) as json_fp:
+            try:
+                json_dict = json.load(json_fp)
+                return 0, "", json_dict
+            except Exception as e:
+                return 1, e, ""
+    except Exception as e:
+        return 1, e , ""
\ No newline at end of file
diff --git a/py_tools/common_logger.py b/py_common/common_logger.py
index e6e8f25..e6e8f25 100644
--- a/py_tools/common_logger.py
+++ b/py_common/common_logger.py
diff --git a/py_common/common_modules_deploy.py b/py_common/common_modules_deploy.py
new file mode 100644
index 0000000..301fae5
--- /dev/null
+++ b/py_common/common_modules_deploy.py
@@ -0,0 +1,48 @@
+#coding=utf-8
+import os
+import sys
+import syslog
+import subprocess
+import time
+import re
+import logging
+import logging.handlers
+from common_system_cmd import *
+from common_logger import *
+from common_json import *
+from common_modules_deploy import *
+
+# coding: utf-8
+#此文件定义四块计算板分别运行什么模块
+#先根据tsg_chassis_ip.json, 根据当前设备的ip地址, 知道当前设备的sled名称
+#然后根据设备类型名称, 找到当前运行的所有模块(应用)名称
+#然后根据模块名称, 到common_modules_operator.json查找每个模块的操作方法
+#{
+#	"modules_deploy": {
+#		"mcn0": ["kni", "a.out"],
+#		"mcn1": ["tfe", "a1.out"],
+#		"mcn2": ["tfe", "a2.out"],
+#		"mcn3": ["tfe", "a3.out"]
+#	}
+#}
+
+G_MODULE_DEPLOY_JSON = "/opt/tsg/etc/tsg_module_deploy.json"
+
+#返回当前板卡运行的所有模块, 数组形式
+def tsg_get_local_sled_modules(sled_name):
+    ret, err_msg, json_dict = tsg_json_parse(G_MODULE_DEPLOY_JSON)   
+    if ret != 0:
+        return {}    
+		
+    module_list = json_dict['modules_deploy']
+    if len(module_list) <= 0:
+	    return {}
+
+    return module_list[sled_name]
+	
+	
+if __name__ == '__main__':
+    global logger
+    logger = logger_init(10)
+    module_list = tsg_get_local_sled_modules("mcn0")
+    print(module_list)
\ No newline at end of file
diff --git a/py_common/common_modules_operator.py b/py_common/common_modules_operator.py
new file mode 100644
index 0000000..5d6d1b1
--- /dev/null
+++ b/py_common/common_modules_operator.py
@@ -0,0 +1,66 @@
+#coding=utf-8
+import os
+import sys
+import syslog
+import subprocess
+import time
+import re
+import logging
+import logging.handlers
+from common_system_cmd import *
+from common_logger import *
+from common_json import *
+
+# coding: utf-8
+#先根据tsg_chassis_ip.json, 根据当前设备的ip地址, 知道当前设备的sled名称
+#然后根据设备类型名称, 找到当前运行的所有模块(应用)名称
+#然后根据模块名称, 到common_modules_operator.json查找每个模块的操作方法
+#此文件定义, 每个模块的启动、停止、检查运行状态的方法
+#操作方法参数说明:
+#
+# prog_name: 实际运行进程名称, 如kni模块实际运行的是sapp
+# extra_progs: 其他需要kill的附加程序, 如sapp的r3守护, 需要杀掉, 否则后台可能会重复启动sapp
+# module_cwd:  应用的绝对路径
+# module_exe: 启动应用的名称, 可能跟module_name不一样, 比如用r2启动sapp
+# stop_method: 停止应用方法  
+# start_method: 启动应用方法  
+# check_method: 检测应用是否运行方法  
+#  
+#例如:   "kni": ["sapp", "r3", "/home/tsg/kni", "r2", "killall", "exec", "ps"],
+
+G_MODULE_DEPLOY_JSON = "/opt/tsg/etc/tsg_module_deploy.json"
+
+TSG_OP_MODULE_NAME_INDEX = 0
+TSG_OP_MODULE_EXTRA_INDEX = 1
+TSG_OP_MODULE_CWD_INDEX = 2
+TSG_OP_MODULE_EXE_INDEX = 3
+TSG_OP_MODULE_STOP_INDEX = 4
+TSG_OP_MODULE_START_INDEX = 5
+TSG_OP_MODULE_STATUS_INDEX = 6
+
+def tsg_get_module_opertor(module_name):
+    empty = []
+    
+    ret, err_msg, json_dict = tsg_json_parse(G_MODULE_DEPLOY_JSON)   
+    if ret != 0:
+        print("open or parse json file %s error, %s" %(G_MODULE_DEPLOY_JSON, err_msg))
+        return empty  
+		
+    module_oplist = json_dict['modules_operator']
+    if len(module_oplist) <= 0:
+        print("can't get modules_operator from %s" %(G_MODULE_DEPLOY_JSON))
+        return empty
+
+    for module_operator in module_oplist:
+        if module_operator == module_name:
+            #print(module_oplist[module_operator])
+            return module_oplist[module_operator]
+        
+    return empty
+	
+	
+if __name__ == '__main__':
+    global logger
+    logger = logger_init(10)
+    module_operator = tsg_get_module_opertor("kni")
+    print(module_operator)
\ No newline at end of file
diff --git a/py_common/common_system_cmd.py b/py_common/common_system_cmd.py
new file mode 100644
index 0000000..047fa40
--- /dev/null
+++ b/py_common/common_system_cmd.py
@@ -0,0 +1,28 @@
+#coding=utf-8
+import os
+import sys
+
+#return exitcode value + output message:
+#    0: succ
+#    1: error
+def system_cmd_run(cmd_str):
+    dangerous_cmd = {"rm", "mv", "poweroff", "shutdown"}
+    
+    for cmd in dangerous_cmd:
+        pattern = "\s*%s" %(cmd)
+        match_str = re.match(pattern, cmd_str)
+        if not match_str is None:
+            print("can't run this cmd:%s" %(cmd_str))
+            sys.exit(1)
+            
+    try:
+        exitcode, output = subprocess.getstatusoutput(cmd_str)
+    except Exception as e:  
+        print(e)    
+        print("###### %s" %(e.message))
+        #if exitcode != 0:
+        #    output = ""
+        return 1, e.message
+
+    return exitcode, output
+    
\ No newline at end of file
diff --git a/py_tools/common_telegraf.py b/py_common/common_telegraf.py
index e1cdc43..e3e5744 100644
--- a/py_tools/common_telegraf.py
+++ b/py_common/common_telegraf.py
@@ -6,4 +6,5 @@ def telegraf_init(arg_host, arg_port, arg_tags):
     #global telegraf_client
     #global_tags = {'host' : "lijia", 'local_ip_addr': "127.0.0.1"}
     telegraf_client = telegraf.TelegrafClient(host = arg_host, port = arg_port, tags = arg_tags)
-    return telegraf_client
\ No newline at end of file
+    return telegraf_client
+    
diff --git a/py_common/common_whoami.py b/py_common/common_whoami.py
new file mode 100644
index 0000000..786ef2a
--- /dev/null
+++ b/py_common/common_whoami.py
@@ -0,0 +1,54 @@
+#coding=utf-8
+import os
+import sys
+import syslog
+import subprocess
+import time
+import re
+import logging
+import logging.handlers
+from common_system_cmd import *
+from common_logger import *
+from common_json import *
+from common_whoami import *
+
+G_CHASSIS_IP_JSON = "/opt/tsg/etc/tsg_chassis_ip.json"
+
+
+
+#遍历所有ip地址, 判断当前的板卡类型, 
+#返回值:
+# type, id, name
+#例如"mcn", 1, "mcn1"
+def tsg_whoami():
+    ret, iplist = tsg_sys_cmd_run("hostname -I")
+    if ret != 0:
+        return "", -1, ""
+        
+    iparray = re.split(r'[:\t\s\r\n]\s*', iplist.strip())    
+
+    ret, err_msg, json_dict = tsg_json_parse(G_CHASSIS_IP_JSON)   
+    if ret != 0:
+        return "", -1, ""    
+    
+    sleds = json_dict['tsg_chassis_ip']
+    if len(sleds) <= 0:
+        logger.critical("can't get tsg_chassis_ip from json file %s!" %(G_CHASSIS_IP_JSON))
+        return "", -1, "" 
+    
+    for ipaddr in iparray:
+        #print("find ip %s in which sled..." %(ipaddr))    
+        for sled in sleds:
+            #print(sled, sled['ip'])
+            #print("diff host ip:%s with json ip:%s..." %(ipaddr), sled['ip'])     
+            if ipaddr == sled['ip']:
+                return sled['type'], sled['id'], sled['name']
+    
+    logger.critical("local ip address can't match any item in json file %s!" %(G_CHASSIS_IP_JSON))
+    return "", -1, ""     
+
+if __name__ == '__main__':
+    global logger
+    logger = logger_init(10)
+    type, id, name = tsg_whoami()
+    print(type, id, name)
\ No newline at end of file
diff --git a/py_tools/get_traffic_by_proc.py b/py_tools/get_traffic_by_proc.py
index fec12bb..ddaffd8 100644
--- a/py_tools/get_traffic_by_proc.py
+++ b/py_tools/get_traffic_by_proc.py
@@ -20,7 +20,7 @@ Transmit_packets_index = 10
 Transmit_err_index = 11

 Transmit_drop_index = 12

 

-#return exitcode value + output message:

+#return exitcode value, output message:

 #    0: succ

 #    1: error

 def system_cmd_run(cmd_str):

diff --git a/py_tools/tsg_diagnose_background.py b/py_tools/tsg_diagnose_background.py
new file mode 100644
index 0000000..bcdd037
--- /dev/null
+++ b/py_tools/tsg_diagnose_background.py
@@ -0,0 +1,23 @@
+# coding: utf-8
+#与提供给用户执行的tsg_diagnose命令区别是:
+#此文件后台周期性自动运行
+#tsg_diagnose 最低级别是warning, 
+import sys
+import time
+import json
+import logging
+from common_telegraf import *
+from common_logger import *
+from common_args import *
+from common_logger import *
+from common_influxdb import *
+
+
+
+tsg_
+
+
+
+if __name__ == '__main__':
+    
+
diff --git a/py_cmd/tsg_get_sn.py b/py_tools/tsg_get_sn.py
index e0bf4f3..e0bf4f3 100644
--- a/py_cmd/tsg_get_sn.py
+++ b/py_tools/tsg_get_sn.py
diff --git a/py_cmd/tsg_update_tags.py b/py_tools/tsg_update_tags.py
index 9650429..c56ed12 100644
--- a/py_cmd/tsg_update_tags.py
+++ b/py_tools/tsg_update_tags.py
@@ -20,7 +20,7 @@ def tsg_get_device_list_info_by_sn(sn):
 def tsg_get_tags_by_sn(sn):
     dev_list_info = tsg_get_device_list_info_by_sn(sn)   
     if len(dev_list_info) <= 0:
-        #写个空json
+        #返回空json
         return "{}"
     
     json_dict = json.loads(dev_list_info)