1.更新oam_core计算内存占用率阈值算法, buff/cache不算已占用;

2.更新tsg-monitor.sh清理/tmp/_MEI临时文件,影响其他模块正常运行的BUG;

install_config/group_vars/all.yml

@@ -52,7 +52,7 @@ oam:
     alarm_app_network_major_high: "0.0001"    
     
 tsg_cli:
-    rpm_file_name: "tsg-cli-1.1.2.93221a7-1.el7.x86_64.rpm"    
+    rpm_file_name: "tsg-cli-1.1.7.420d736-1.el7.x86_64.rpm"    
 
 ha_master:
     instance_state: MASTER

roles/oam_cli_agent/templates/oam_cli_agent.service.j2

@@ -3,6 +3,7 @@ Description=oam_cli_agent daemon
 
 [Service]
 Type=simple
+ExecStartPre=/usr/bin/mkdir -p /tmp/__tsg_monitor_tmp_dir
 Environment=OAM_DIR=/opt/tsg/cli_agent/
 ExecStart=/usr/lib/jvm/{{ java_version }}/bin/java  -jar ${OAM_DIR}oam_cli_agent.jar --spring.config.location=${OAM_DIR}application.properties,${OAM_DIR}application-centos2.properties --logging.file=${OAM_DIR}oam-cli-agent.log
 

roles/oam_cli_agent_mxn/templates/oam_cli_agent_mxn.service.j2

@@ -4,6 +4,7 @@ Description=oam_cli_agent daemon
 [Service]
 Type=simple
 Environment=OAM_DIR=/opt/tsg/cli_agent/
+ExecStartPre=/usr/bin/mkdir -p /tmp/__tsg_monitor_tmp_dir
 ExecStart=/usr/lib/jvm/{{ java_version }}/bin/java -jar ${OAM_DIR}oam_cli_agent.jar --spring.config.location=${OAM_DIR}application.properties,${OAM_DIR}application-centos2.properties --logging.file=${OAM_DIR}oam-cli-agent.log
 
 [Install]

roles/oam_core/templates/oam_core.service.j2

@@ -4,7 +4,7 @@ Description=oam_core daemon
 [Service]
 Type=simple
 Environment=OAM_DIR=/opt/tsg/oam_core/
-ExecStart=/usr/lib/jvm/{{ java_version }}/bin/java -jar ${OAM_DIR}oam_core.jar --spring.config.location=${OAM_DIR}application.properties,${OAM_DIR}application-centos2.properties --logging.file=${OAM_DIR}oam-core.log
+ExecStart=/usr/lib/jvm/{{ java_version }}/bin/java -jar ${OAM_DIR}oam-core.jar --spring.config.location=${OAM_DIR}application.properties,${OAM_DIR}application-centos2.properties --logging.file=${OAM_DIR}oam-core.log
 
 [Install]
 WantedBy=multi-user.target

roles/tsg-cli-mcn0/files/tsg-monitor.sh

@@ -13,31 +13,22 @@ killall_uncompleted_cmd(){
     killall -9 -q tsg_monit_app
     killall -9 -q tsg_monit_intercept
     killall -9 -q tsg_monit_interface
-    #killall -9 -q tsg_monit_protocol_v3
     killall -9 -q tsg_monit_protocol_v4
-    #killall -9 -q tsg_monit_stream_v3
     killall -9 -q tsg_monit_stream_v4
 
-	cmd_run=`ps -ef | grep tsg_ | grep -v grep`
+	/opt/tsg/tsg-monitor/tsg_monitor_cleanup_tmp.sh
-	if [ ${#cmd_run} -lt 1 ] ; then
-	    #在没有其他tsg_xxx命令正在执行时, 删除执行命令错误被强制kill时, python生成的临时文件
-	    rm -rf /tmp/_MEI*
-    fi	
-
 }
 
 start_background_cmd(){
+	mkdir -p  /tmp/__tsg_monitor_tmp_dir
     #后台并发运行, 保证所有命令的开始运行时间基本一样，
     #且不会因某个命令网络拥塞、执行时间长等问题，阻塞while(1)主循环
     /opt/tsg/tsg-monitor/tsg_diagnose_background > /dev/null &
     #/opt/tsg/tsg-monitor/tsg_update_tags > /dev/null &
-    
     /opt/tsg/tsg-monitor/tsg_monit_app > /dev/null &
     /opt/tsg/tsg-monitor/tsg_monit_intercept > /dev/null &
     /opt/tsg/tsg-monitor/tsg_monit_interface > /dev/null &
-    #/opt/tsg/tsg-monitor/tsg_monit_protocol_v3 > /dev/null &
     /opt/tsg/tsg-monitor/tsg_monit_protocol_v4 > /dev/null &
-    #/opt/tsg/tsg-monitor/tsg_monit_stream_v3 > /dev/null &
     /opt/tsg/tsg-monitor/tsg_monit_stream_v4 > /dev/null &    
 }
 

roles/tsg-cli-mcn0/tasks/main.yml

@@ -57,6 +57,12 @@
     src: "{{ role_path }}/files/rsyslog.conf"
     dest: "/etc/"
 
+- name: "copy tsg_monitor_cleanup_tmp.sh to destination server"
+  synchronize:
+    src: "{{ role_path }}/../tsg-common-files/tsg_monitor_cleanup_tmp.sh"
+    dest: "/opt/tsg/tsg-monitor/"
+    
+
 - name: "restart rsyslog service"
   systemd:
     name: rsyslog

roles/tsg-cli-mcn1/files/tsg-monitor.sh

@@ -12,14 +12,11 @@ killall_uncompleted_cmd(){
     killall -9 -q tsg_monit_interface
     killall -9 -q tsg_monit_intercept
 	
-	cmd_run=`ps -ef | grep tsg_ | grep -v grep`
+	/opt/tsg/tsg-monitor/tsg_monitor_cleanup_tmp.sh
-	if [ ${#cmd_run} -lt 1 ] ; then
-	    #在没有其他tsg_xxx命令正在执行时, 删除执行命令错误被强制kill时, python生成的临时文件
-	    rm -rf /tmp/_MEI*
-    fi
 }
 
 start_background_cmd(){
+	mkdir -p  /tmp/__tsg_monitor_tmp_dir
     #后台并发运行, 保证所有命令的开始运行时间基本一样，
     #且不会因某个命令网络拥塞、执行时间长等问题，阻塞while(1)主循环
     /opt/tsg/tsg-monitor/tsg_diagnose_background > /dev/null &

roles/tsg-cli-mcn1/tasks/main.yml

@@ -57,6 +57,11 @@
     src: "{{ role_path }}/files/rsyslog.conf"
     dest: "/etc/"
 
+- name: "copy tsg_monitor_cleanup_tmp.sh to destination server"
+  synchronize:
+    src: "{{ role_path }}/../tsg-common-files/tsg_monitor_cleanup_tmp.sh"
+    dest: "/opt/tsg/tsg-monitor/"
+    
 - name: "restart rsyslog service"
   systemd:
     name: rsyslog

roles/tsg-cli-mcn2/files/tsg-monitor.sh

@@ -12,14 +12,11 @@ killall_uncompleted_cmd(){
     killall -9 -q tsg_monit_interface
     killall -9 -q tsg_monit_intercept
 	
-	cmd_run=`ps -ef | grep tsg_ | grep -v grep`
+	/opt/tsg/tsg-monitor/tsg_monitor_cleanup_tmp.sh
-	if [ ${#cmd_run} -lt 1 ] ; then
-	    #在没有其他tsg_xxx命令正在执行时, 删除执行命令错误被强制kill时, python生成的临时文件
-	    rm -rf /tmp/_MEI*
-    fi
 }
 
 start_background_cmd(){
+	mkdir -p  /tmp/__tsg_monitor_tmp_dir
     #后台并发运行, 保证所有命令的开始运行时间基本一样，
     #且不会因某个命令网络拥塞、执行时间长等问题，阻塞while(1)主循环
     /opt/tsg/tsg-monitor/tsg_diagnose_background > /dev/null &

roles/tsg-cli-mcn2/tasks/main.yml

@@ -57,6 +57,11 @@
     src: "{{ role_path }}/files/rsyslog.conf"
     dest: "/etc/"
 
+- name: "copy tsg_monitor_cleanup_tmp.sh to destination server"
+  synchronize:
+    src: "{{ role_path }}/../tsg-common-files/tsg_monitor_cleanup_tmp.sh"
+    dest: "/opt/tsg/tsg-monitor/"
+    
 - name: "restart rsyslog service"
   systemd:
     name: rsyslog

roles/tsg-cli-mcn3/files/tsg-monitor.sh

@@ -12,14 +12,11 @@ killall_uncompleted_cmd(){
     killall -9 -q tsg_monit_interface
     killall -9 -q tsg_monit_intercept
 	
-	cmd_run=`ps -ef | grep tsg_ | grep -v grep`
+	/opt/tsg/tsg-monitor/tsg_monitor_cleanup_tmp.sh
-	if [ ${#cmd_run} -lt 1 ] ; then
-	    #在没有其他tsg_xxx命令正在执行时, 删除执行命令错误被强制kill时, python生成的临时文件
-	    rm -rf /tmp/_MEI*
-    fi
 }
 
 start_background_cmd(){
+	mkdir -p  /tmp/__tsg_monitor_tmp_dir
     #后台并发运行, 保证所有命令的开始运行时间基本一样，
     #且不会因某个命令网络拥塞、执行时间长等问题，阻塞while(1)主循环
     /opt/tsg/tsg-monitor/tsg_diagnose_background > /dev/null &

roles/tsg-cli-mcn3/tasks/main.yml

@@ -57,6 +57,11 @@
     src: "{{ role_path }}/files/rsyslog.conf"
     dest: "/etc/"
 
+- name: "copy tsg_monitor_cleanup_tmp.sh to destination server"
+  synchronize:
+    src: "{{ role_path }}/../tsg-common-files/tsg_monitor_cleanup_tmp.sh"
+    dest: "/opt/tsg/tsg-monitor/"
+    
 - name: "restart rsyslog service"
   systemd:
     name: rsyslog

roles/tsg-cli-mxn/files/tsg-monitor.sh

@@ -8,21 +8,17 @@
 #导致10秒内还不结束, 要强行杀掉,
 #否则长时间运行后, 会有大量后台进程运行!
 killall_uncompleted_cmd(){
-    killall -9 -q tsg_cluster_register
+    #killall -9 -q tsg_cluster_register
     #killall -9 -q tsg_diagnose_background
     #killall -9 -q tsg_update_tags
-	
+	/opt/tsg/tsg-monitor/tsg_monitor_cleanup_tmp.sh
-	cmd_run=`ps -ef | grep tsg_ | grep -v grep`
-	if [ ${#cmd_run} -lt 1 ] ; then
-	    #在没有其他tsg_xxx命令正在执行时, 删除执行命令错误被强制kill时, python生成的临时文件
-	    rm -rf /tmp/_MEI*
-    fi
 } 
 
 start_background_cmd(){
+	mkdir -p  /tmp/__tsg_monitor_tmp_dir
     #后台并发运行, 保证所有命令的开始运行时间基本一样，
     #且不会因某个命令网络拥塞、执行时间长等问题，阻塞while(1)主循环
-    /opt/tsg/tsg-monitor/tsg_cluster_register > /dev/null &
+    #/opt/tsg/tsg-monitor/tsg_cluster_register > /dev/null &
     
     #mxn板只检测cpu, mem, disk等, 前台cli命令启用diagnose,
     #后台服务依靠oam snmp模块, 无需运行tsg_diagnose_background

roles/tsg-cli-mxn/files/tsg_cluster_register.service

@@ -4,6 +4,7 @@ Requires=consul-client.service
 After=consul-client.service
 
 [Service]
+ExecStartPre=/usr/bin/mkdir -p /tmp/__tsg_monitor_tmp_dir
 ExecStart=/opt/tsg/tsg-monitor/tsg_cluster_register
 Type=oneshot
 RemainAfterExit=yes

roles/tsg-cli-mxn/tasks/main.yml

@@ -74,29 +74,44 @@
     dest: "/opt/tsg/etc/tsg_series.json"
   tags: template
 
-#2020-11-13 lijia modify
+- name: "copy tsg-monitor.service to destination server"
+  synchronize:
+    src: "{{ role_path }}/files/tsg-monitor.service"
+    dest: "/usr/lib/systemd/system"
+    
+#2020-11-13 lijia add
 - name: "copy tsg_cluster_register.service to destination server"
   synchronize:
     src: "{{ role_path }}/files/tsg_cluster_register.service"
     dest: "/usr/lib/systemd/system"
 
-# 2020-11-13 lijia close
+- name: "copy tsg-monitor.sh to destination server"
-#- name: "copy tsg-monitor.sh to destination server"
+  copy:
-#  copy:
+    src: "{{ role_path }}/files/tsg-monitor.sh"
-#    src: "{{ role_path }}/files/tsg-monitor.sh"
+    dest: "/opt/tsg/tsg-monitor/"    
-#    dest: "/opt/tsg/tsg-monitor/"    
+    mode: 0755  
-#    mode: 0755  
 
 - name: "copy rsyslog.conf to destination server"
   synchronize:
     src: "{{ role_path }}/files/rsyslog.conf"
     dest: "/etc/"    
 
+- name: "copy tsg_monitor_cleanup_tmp.sh to destination server"
+  synchronize:
+    src: "{{ role_path }}/../tsg-common-files/tsg_monitor_cleanup_tmp.sh"
+    dest: "/opt/tsg/tsg-monitor/"
+    
 - name: "restart rsyslog service"
   systemd:
     name: rsyslog
     state: restarted
 
+- name: "enable tsg-monitor service"
+  systemd:
+    name: tsg-monitor
+    enabled: yes
+    daemon_reload: yes
+    state: restarted
  
 - name: "enable tsg_cluster_register.service"
   systemd:

roles/tsg-common-files/tsg_monitor_cleanup_tmp.sh

@@ -0,0 +1,13 @@
+#!/bin/sh
+cd /tmp/__tsg_monitor_tmp_dir
+
+if [ $? != 0 ]; then
+    #echo "cd to dir error!"
+    exit 1
+fi
+
+#cur_dir=`pwd`
+#echo "cur dir is: $cur_dir"
+
+find . -ctime +1 -type d -name "_MEI*" | xargs rm -rf
+