网站运营
位置:首页>> 网站运营>> cloudstack下libvirtd服务无响应问题

cloudstack下libvirtd服务无响应问题

作者:gordonchang  发布时间:2023-07-26 05:38:44 

标签:cloudstack,libvirtd

在cloudstack4.5.2版本下,偶尔出现libvirtd服务无响应的情况,导致virsh命令无法使用,同时伴随cloudstack master丢失该slave主机连接的情况。最初怀疑是libvirtd服务或版本的问题,经过分析和排查最终确定是cloudstack-agent的问题。但是在官网上并没有找到类似的bug提交,该问题可能还存在于更高的版本,需要时间进一步从根本上分析。下面是该问题的处理过程,在此记录下,关注和使用cloudstack的朋友可以参考。

众所周知,cloudstack的社区热度远不如openstack,为什么还要选择clcoudstack?这个问题以后有机会再和大家聊。言归正传。

环境交代

宿主机操作系统:centos6.5x64(2.6.32-431.el6.x86_64)
cloudstack版本:4.5.2
libvirt版本:libvirt-0.10.2-54.el6_7.2.x86_64

问题描述

通过cloudstackapi listHosts报警信息显示:
node5.cloud.rtmap:192.168.14.20 state is Down at 2016-05-13T07:19:04+0800
#有关cloudstackapi的使用方法在其它文章中总结,不在此处说明。

登陆问题宿主服务器检查:


[root@node5 log]#virsh list --all

没有响应ctrl^c退出

这时的vm可以正常工作,但处于失控状态

尝试重启启动libvirtd服务:


[root@node5 log]# service libvirtd stop

正在关闭 libvirtd 守护进程:                               [失败]#无法关闭libvirtd服务

尝试重启启动cloudstack-agent服务:


[root@node5 libvirt]# service cloudstack-agent restart
Stopping Cloud Agent:
Starting Cloud Agent:

libvirtd故障依旧

简单维护


[root@node5 ping]# libvirtd -d -l --config /etc/libvirt/libvirtd.conf

libvirtd:错误:Unable to initialize network sockets。查看 /var/log/messages 或者运行不带 --daemon 的命令查看更多信息。


[root@node5 log]# libvirtd -d

可以执行成功,这时执行virsh list --all 可以查看和操作vm


[root@node5 log]#virsh list --all
Id  名称             状态
----------------------------------------------------
2   i-4-185-VM           running

虽然vm运行正常,现在也可以通过命令正常管理了。但是对于cloudstack平台而言,宿主机处于down状态,vm处于失控状态。

临时解决办法是在其它大的升级和维护过程中重启服务器解决,根本解决还要具体问题具体分析。

分析与排查

检查进程


[root@node5 log]# ps ax |grep libvirtd
6485 ?    R  863:37 libvirtd --daemon -l#该服务始终处于run状态


[root@node5 log]# top -p 6485
top -p 6485
top - 09:19:41 up 12 days, 22:27, 1 user, load average: 3.05, 5.07, 6.64
Tasks:  1 total,  0 running,  1 sleeping,  0 stopped,  0 zombie
Cpu(s): 4.8%us, 1.4%sy, 0.0%ni, 93.1%id, 0.6%wa, 0.0%hi, 0.1%si, 0.0%st
Mem: 264420148k total, 182040780k used, 82379368k free,  834232k buffers
Swap: 8388600k total,    92k used, 8388508k free, 100453708k cached

PID USER   PR NI VIRT RES SHR S %CPU %MEM  TIME+ COMMAND
6485 root   20  0 984m 12m 4440 R 100.2 0.0 844:22.68 libvirtd #cpu占用100%,无法释放,影响系统稳定性

杀进程


[root@node5 log]# kill -9 6485
[root@node5 log]# kill -9 6485
[root@master log]# ps ax |grep libvirtd#检查进程依然存在
6485 ?    R  863:37 libvirtd --daemon -l
[root@node5 ~]# libvirtd -d -l --config /etc/libvirt/libvirtd.conf
libvirtd:错误:Unable to initialize network sockets。查看 /var/log/messages 或者运行不带 --daemon 的命令查看更多信息。
[root@node5 ~]# netstat -antp |grep 16509
tcp    0   0 0.0.0.0:16509        0.0.0.0:*          LISTEN   3658/libvirtd    
tcp    1   0 192.168.14.25:16509     192.168.14.22:8717     CLOSE_WAIT -          
tcp    1   0 192.168.14.25:16509     192.168.14.20:5152     CLOSE_WAIT -          
tcp    1   0 192.168.14.25:16509     192.168.14.10:39359     CLOSE_WAIT -          
tcp    0   0 :::16509          :::*            LISTEN   3658/libvirtd    
tcp    39   0 ::1:16509          ::1:19715          CLOSE_WAIT -

经过上述操作,初步判断libvirtd陷入了hang死状态。

追踪进程


[root@node5 log]#strace -f libvirtd
[pid 107570] close(23058)        = -1 EBADF (Bad file descriptor)
[pid 107570] close(23059)        = -1 EBADF (Bad file descriptor)
[pid 107570] close(23060)        = -1 EBADF (Bad file descriptor)
[pid 107570] close(23061)        = -1 EBADF (Bad file descriptor)
[pid 107570] close(23062)        = -1 EBADF (Bad file descriptor)
[pid 107570] close(23063)        = -1 EBADF (Bad file descriptor)
[pid 107570] close(23064)        = -1 EBADF (Bad file descriptor)
[pid 107570] close(23065)        = -1 EBADF (Bad file descriptor)
[pid 107570] close(23066)        = -1 EBADF (Bad file descriptor)
[pid 107570] close(23067)        = -1 EBADF (Bad file descriptor)
[pid 107570] close(23068)        = -1 EBADF (Bad file descriptor)
[pid 107570] close(23069)        = -1 EBADF (Bad file descriptor)
[pid 107570] close(23070)        = -1 EBADF (Bad file descriptor)
[pid 107570] close(23071)        = -1 EBADF (Bad file descriptor)
^C[pid 107570] close(23072 <unfinished ...>
Process 107559 detached
Process 107560 detached
Process 107561 detached
Process 107562 detached
Process 107563 detached
Process 107564 detached
Process 107565 detached
Process 107566 detached
Process 107567 detached
Process 107568 detached
Process 107569 detached
Process 107570 detached

父进程6485在不断的产生和关闭子进程,并返回错误信息。Bad file descriptor的原因(如何触发的,谁触发的)? 循环为何无法退出?问题如何再现?

获得更多的线索

官方文档(libvirtd各种故障诊断记录和解决办法非常详尽)
https://access.redhat.com/documentation/en-US/Red_Hat_Enterprise_Linux/7/html/Virtualization_Deployment_and_Administration_Guide/sect-Troubleshooting-Common_libvirt_errors_and_troubleshooting.html#sect-libvirtd_failed_to_start

开启系统日志

Change libvirt's logging in /etc/libvirt/libvirtd.conf by enabling the line below. To enable the setting the line, open the /etc/libvirt/libvirtd.conf file in a text editor, remove the hash (or #) symbol from the beginning of the following line, and save the change:
log_outputs="3:syslog:libvirtd"

参照配置,重启服务器等待下次故障观察日志


......

Jun 1 12:42:26 node5 abrtd: New client connected
Jun 1 12:42:26 node5 abrtd: Directory 'pyhook-2016-06-01-12:42:26-70065' creation detected
Jun 1 12:42:26 node5 abrt-server[70066]: Saved Python crash dump of pid 70065 to /var/spool/abrt/pyhook-2016-06-01-12:42:26-70065
Jun 1 12:42:26 node5 abrtd: Package 'cloudstack-common' isn't signed with proper key
Jun 1 12:42:26 node5 abrtd: 'post-create' on '/var/spool/abrt/pyhook-2016-06-01-12:42:26-70065' exited with 1
Jun 1 12:42:26 node5 abrtd: Deleting problem directory '/var/spool/abrt/pyhook-2016-06-01-12:42:26-70065'
Jun 1 12:43:26 node5 abrt: detected unhandled Python exception in '/usr/share/cloudstack-common/scripts/vm/network/security_group.py'
......

Jun 6 10:36:21 node5 libvirtd: 102840: warning : qemuDomainObjBeginJobInternal:878 : Cannot start job (modify, none) for domain i-4-30-VM; current job is (modify, none) owned by (102925, 0)
Jun 6 10:36:21 node5 libvirtd: 102840: error : qemuDomainObjBeginJobInternal:883 : Timed out during operation: cannot acquire state change lock
Jun 6 10:39:59 node5 libvirtd: 114071: info : libvirt version: 0.10.2, package: 54.el6_7.2 (CentOS BuildSystem <http://bugs.centos.org>, 2015-11-10-10:25:08, c6b9.bsys.dev.centos.org)
Jun 6 10:39:59 node5 libvirtd: 114071: error : virNetSocketNewListenTCP:312 : Unable to bind to port: 地址已在使用
Jun 6 10:40:46 node5 libvirtd: 114147: info : libvirt version: 0.10.2, package: 54.el6_7.2 (CentOS BuildSystem <http://bugs.centos.org>, 2015-11-10-10:25:08, c6b9.bsys.dev.centos.org)
Jun 6 10:40:46 node5 libvirtd: 114147: error : virNetSocketNewListenTCP:312 : Unable to bind to port: 地址已在使用
Jun 6 10:42:15 node5 libvirtd: 114204: info : libvirt version: 0.10.2, package: 54.el6_7.2 (CentOS BuildSystem <http://bugs.centos.org>, 2015-11-10-10:25:08, c6b9.bsys.dev.centos.org)
Jun 6 10:42:15 node5 libvirtd: 114204: error : virNetSocketNewListenTCP:312 : Unable to bind to port: 地址已在使用
Jun 6 10:47:05 node5 libvirtd: 114375: info : libvirt version: 0.10.2, package: 54.el6_7.2 (CentOS BuildSystem <http://bugs.centos.org>, 2015-11-10-10:25:08, c6b9.bsys.dev.centos.org)
Jun 6 10:47:05 node5 libvirtd: 114375: error : virNetSocketNewListenTCP:312 : Unable to bind to port: 地址已在使用
Jun 6 10:47:23 node5 libvirtd: 114412: info : libvirt version: 0.10.2, package: 54.el6_7.2 (CentOS BuildSystem <http://bugs.centos.org>, 2015-11-10-10:25:08, c6b9.bsys.dev.centos.org)
Jun 6 10:47:23 node5 libvirtd: 114412: error : virNetSocketNewListenTCP:312 : Unable to bind to port: 地址已在使用
......

Jun 12 03:08:02 node5 rsyslogd: [origin software="rsyslogd" swVersion="5.8.10" x-pid="3111" x-info="http://www.rsyslog.com"] rsyslogd was HUPed
Jun 12 09:20:40 node5 libvirtd: 72575: info : libvirt version: 0.10.2, package: 54.el6_7.2 (CentOS BuildSystem <http://bugs.centos.org>, 2015-11-10-10:25:08, c6b9.bsys.dev.centos.org)
Jun 12 09:20:40 node5 libvirtd: 72575: error : virPidFileAcquirePath:410 : Failed to acquire pid file '/var/run/libvirtd.pid': 资源暂时不可用

并未获得致命错误和更多线索。(该日志配置选项还是很有必要打开的,很多问题都可以通过它来定位)

解决过程

解决思路

尝试和找到终止进程、重启服务的方法
提交bug,等待补丁升级
分析源代码,再现问题,解决问题(投入研发和时间)
由于不能再现问题,还是从简入繁吧。触发这些子进程的元凶是谁?还是cloudstack-agent的嫌疑最大,但之前重启过该服务并没有解决问题,那么agent服务是怎么一回事呢?

看下启动脚本可以基本了解,


[root@node5 libvirt]# cat /etc/rc.d/init.d/cloudstack-agent

#!/bin/bash

# chkconfig: 35 99 10
# description: Cloud Agent

# Licensed to the Apache Software Foundation (ASF) under one
# or more contributor license agreements. See the NOTICE file
# distributed with this work for additional information
# regarding copyright ownership. The ASF licenses this file
# to you under the Apache License, Version 2.0 (the
# "License"); you may not use this file except in compliance
# with the License. You may obtain a copy of the License at
#
#  http://www.apache.org/licenses/LICENSE-2.0
#
# Unless required by applicable law or agreed to in writing,
# software distributed under the License is distributed on an
# "AS IS" BASIS, WITHOUT WARRANTIES OR CONDITIONS OF ANY
# KIND, either express or implied. See the License for the
# specific language governing permissions and limitations
# under the License.

# WARNING: if this script is changed, then all other initscripts MUST BE changed to match it as well

. /etc/rc.d/init.d/functions

# set environment variables

SHORTNAME=$(basename $0 | sed -e 's/^[SK][0-9][0-9]//')
PIDFILE=/var/run/"$SHORTNAME".pid
LOCKFILE=/var/lock/subsys/"$SHORTNAME"
LOGDIR=/var/log/cloudstack/agent
LOGFILE=${LOGDIR}/agent.log
PROGNAME="Cloud Agent"
CLASS="com.cloud.agent.AgentShell"
JSVC=`which jsvc 2>/dev/null`;

# exit if we don't find jsvc
if [ -z "$JSVC" ]; then
 echo no jsvc found in path;
 exit 1;
fi

unset OPTIONS
[ -r /etc/sysconfig/"$SHORTNAME" ] && source /etc/sysconfig/"$SHORTNAME"

# The first existing directory is used for JAVA_HOME (if JAVA_HOME is not defined in $DEFAULT)
JDK_DIRS="/usr/lib/jvm/jre /usr/lib/jvm/java-7-openjdk /usr/lib/jvm/java-7-openjdk-i386 /usr/lib/jvm/java-7-openjdk-amd64 /usr/lib/jvm/java-6-openjdk /usr/lib/jvm/java-6-openjdk-i386 /usr/lib/jvm/java-6-openjdk-amd64 /usr/lib/jvm/java-6-sun"

for jdir in $JDK_DIRS; do
 if [ -r "$jdir/bin/java" -a -z "${JAVA_HOME}" ]; then
   JAVA_HOME="$jdir"
 fi
done
export JAVA_HOME

ACP=`ls /usr/share/cloudstack-agent/lib/*.jar | tr '\n' ':' | sed s'/.$//'`
* =`ls /usr/share/cloudstack-agent/plugins/*.jar 2>/dev/null | tr '\n' ':' | sed s'/.$//'`

# We need to append the JSVC daemon JAR to the classpath
# AgentShell implements the JSVC daemon methods
export CLASSPATH="/usr/share/java/commons-daemon.jar:$ACP:$ * :/etc/cloudstack/agent:/usr/share/cloudstack-common/scripts"

start() {
 echo -n $"Starting $PROGNAME: "
 if hostname --fqdn >/dev/null 2>&1 ; then
   $JSVC -Xms256m -Xmx2048m -cp "$CLASSPATH" -pidfile "$PIDFILE" \
     -errfile $LOGDIR/cloudstack-agent.err -outfile $LOGDIR/cloudstack-agent.out $CLASS
   RETVAL=$?
   echo
 else
   failure
   echo
   echo The host name does not resolve properly to an IP address. Cannot start "$PROGNAME". > /dev/stderr
   RETVAL=9
 fi
 [ $RETVAL = 0 ] && touch ${LOCKFILE}
 return $RETVAL
}

stop() {
 echo -n $"Stopping $PROGNAME: "
 $JSVC -pidfile "$PIDFILE" -stop $CLASS
 RETVAL=$?
 echo
 [ $RETVAL = 0 ] && rm -f ${LOCKFILE} ${PIDFILE}
}

case "$1" in
 start)
   start
   ;;
 stop)
   stop
   ;;
 status)
   status -p ${PIDFILE} $SHORTNAME
   RETVAL=$?
   ;;
 restart)
   stop
   sleep 3
   start
   ;;
 condrestart)
   if status -p ${PIDFILE} $SHORTNAME >&/dev/null; then
     stop
     sleep 3
     start
   fi
   ;;
 *)
 echo $"Usage: $SHORTNAME {start|stop|restart|condrestart|status|help}"
 RETVAL=3
esac

exit $RETVAL


[root@node5 libvirt]# ps ax |grep jsvc.exec
6655 ?    Ss   0:00 jsvc.exec -Xms256m -Xmx2048m -cp /usr/share/java/commons-daemon.jar:/usr/share/cloudstack-agent/lib/activation-1.1.jar:/usr/share/cloudstack-agent/lib/antisamy-1.4.3.jar:/usr/share/cloudstack-agent/lib/aopalliance-1.0.jar:/usr/share/cloudstack-agent/lib/apache-log4j-extras-1.1.jar:/usr/share/cloudstack-agent/lib/aspectjweaver-1.7.0.jar:/usr/share/cloudstack-agent/lib/aws-java-sdk-1.3.22.jar:/usr/share/cloudstack-agent/lib/batik-css-1.7.jar:/usr/share/cloudstack-agent/lib/batik-ext-1.7.jar:/usr/share/cloudstack-agent/lib/batik-util-1.7.jar:/usr/share/cloudstack-agent/lib/bcprov-jdk15-1.46.jar:/usr/share/cloudstack-agent/lib/bcprov-jdk16-1.46.jar:/usr/share/cloudstack-agent/lib/bsh-core-2.0b4.jar:/usr/share/cloudstack-agent/lib/cglib-nodep-2.2.2.jar:/usr/share/cloudstack-agent/lib/cloud-agent-4.5.2.jar:/usr/share/cloudstack-agent/lib/cloud-api-4.5.2.jar:/usr/share/cloudstack-agent/lib/cloud-core-4.5.2.jar:/usr/share/cloudstack-agent/lib/cloud-engine-api-4.5.2.jar:/usr/share/cloudstack-agent/lib/cloud-engine-components-api-4.5.2.jar:/usr/share/cloudstack-agent/lib/cloud-engine-schema-4.5.2.jar:/usr/share/cloudstack-agent/lib/cloud-framework-cluster-4.5.2.jar:/usr/share/cloudstack-agent/lib/cloud-framework-config-4.5.2.jar:/usr/share/cloudstack-agent/lib/cloud-framework-db-4.5.2.jar:/usr/share/cloudstack-agent/lib/cloud-framework-events-4.5.2.jar:/usr/share/cloudstack-agent/lib/cloud-framework-ipc-4.5.2.jar:/usr/share/cloudstack-agent/lib/cloud-framework-jobs-4.5.2.jar:/usr/share/cloudstack-agent/lib/cloud-framework-managed-context-4.5.2.jar:/usr/share/cloudstack-agent/lib/cloud-framework-rest-4.5.2.jar:/usr/share/cloudstack-agent/lib/cloud-framework-security-4.5.2.jar:/usr/share/cloudstack-agent/lib/cloud-plugin-hypervisor-kvm-4.5.2.jar:/usr/share/cloudstack-agent/lib/cloud-plugin-network-ovs-4.5.2.jar:/usr/share/cloudstack-agent/lib/cloud-server-4.5.2.jar:/usr/share/cloudstack-agent/lib/cloud-utils-4.5.2.jar:/usr/share/cloudstack-agent/lib/commons-beanutils-core-1.7.0.jar:/usr/share/cloudstack-agent/lib/commons-codec-1.6.jar:/usr/share/cloudstack-agent/lib/commons-collections-3.2.1.jar:/usr/share/cloudstack-agent/lib/commons-configuration-1.8.jar:/usr/share/cloudstack-agent/lib/commons-daemon-1.0.10.jar:/usr/share/cloudstack-agent/lib/commons-dbcp-1.4.jar:/usr/share/cloudstack-agent/lib/commons-fileupload-1.2.jar:/usr/share/cloudstack-agent/lib/commons-httpclient-3.1.jar:/usr/share/cloudstack-agent/lib/commons-io-1.4.jar:/usr/share/cloudstack-agent/lib/commons-lang-2.6.jar:/usr/share/cloudstack-agent/lib/commons-logging-1.1.3.jar:/usr/share/cloudstack-agent/lib/commons-net-3.3.jar:/usr/share/cloudstack-agent/lib/commons-pool-1.6.jar:/usr/share/cloudstack-agent/lib/cxf-bundle-jaxrs-2.7.0.jar:/usr/share/cloudstack-agent/lib/dom4j-1.6.1.jar:/usr/share/cloudstack-agent/lib/ehcache-core-2.6.6.jar:/usr/share/cloudstack-agent/lib/ejb-api-3.0.jar:/usr/share/cloudstack-agent/lib/esapi-2.0.1.jar:/usr/share/cloudstack-agent/lib/geronimo-javamail_1.4_spec-1.7.1.jar:/usr/share/cloudstack-agent/lib/geronimo-servlet_3.0_spec-1.0.jar:/usr/share/cloudstack-agent/lib/gson-1.7.2.jar:/usr/share/cloudstack-agent/lib/guava-14.0-rc1.jar:/usr/share/cloudstack-agent/lib/httpclient-4.3.6.jar:/usr/share/cloudstack-agent/lib/httpcore-4.3.3.jar:/usr/share/cloudstack-agent/lib/jackson-annotations-2.1.1.jar:/usr/share/cloudstack-agent/lib/jackson-core-2.1.1.jar:/usr/share/cloudstack-agent/lib/jackson-core-asl-1.8.9.jar:/usr/share/cloudstack-agent/lib/jackson-databind-2.1.1.jar:/usr/share/cloudstack-agent/lib/jackson-jaxrs-json-provider-2.1.1.jar:/usr/share/cloudstack-agent/lib/jackson-mapper-asl-1.8.9.jar:/usr/share/cloudstack-agent/lib/jackson-module-jaxb-annotations-2.1.1.jar:/usr/share/cloudstack-agent/lib/jasypt-1.9.0.jar:/usr/share/cloudstack-agent/lib/java-ipv6-0.10.jar:/usr/share/cloudstack-agent/lib/javassist-3.12.1.GA.jar:/usr/share/cloudstack-agent/lib/javassist-3.18.1-GA.jar:/usr/share/cloudstack-agent/lib/javax.inject-1.jar:/usr/share/cloudstack-agent/lib/javax.persistence-2.0.0.jar:/usr/share/cloudstack-agent/lib/javax.ws.rs-api-2.0-m10.jar
6657 ?    Sl   0:05 jsvc.exec -Xms256m -Xmx2048m -cp /usr/share/java/commons-daemon.jar:/usr/share/cloudstack-agent/lib/activation-1.1.jar:/usr/share/cloudstack-agent/lib/antisamy-1.4.3.jar:/usr/share/cloudstack-agent/lib/aopalliance-1.0.jar:/usr/share/cloudstack-agent/lib/apache-log4j-extras-1.1.jar:/usr/share/cloudstack-agent/lib/aspectjweaver-1.7.0.jar:/usr/share/cloudstack-agent/lib/aws-java-sdk-1.3.22.jar:/usr/share/cloudstack-agent/lib/batik-css-1.7.jar:/usr/share/cloudstack-agent/lib/batik-ext-1.7.jar:/usr/share/cloudstack-agent/lib/batik-util-1.7.jar:/usr/share/cloudstack-agent/lib/bcprov-jdk15-1.46.jar:/usr/share/cloudstack-agent/lib/bcprov-jdk16-1.46.jar:/usr/share/cloudstack-agent/lib/bsh-core-2.0b4.jar:/usr/share/cloudstack-agent/lib/cglib-nodep-2.2.2.jar:/usr/share/cloudstack-agent/lib/cloud-agent-4.5.2.jar:/usr/share/cloudstack-agent/lib/cloud-api-4.5.2.jar:/usr/share/cloudstack-agent/lib/cloud-core-4.5.2.jar:/usr/share/cloudstack-agent/lib/cloud-engine-api-4.5.2.jar:/usr/share/cloudstack-agent/lib/cloud-engine-components-api-4.5.2.jar:/usr/share/cloudstack-agent/lib/cloud-engine-schema-4.5.2.jar:/usr/share/cloudstack-agent/lib/cloud-framework-cluster-4.5.2.jar:/usr/share/cloudstack-agent/lib/cloud-framework-config-4.5.2.jar:/usr/share/cloudstack-agent/lib/cloud-framework-db-4.5.2.jar:/usr/share/cloudstack-agent/lib/cloud-framework-events-4.5.2.jar:/usr/share/cloudstack-agent/lib/cloud-framework-ipc-4.5.2.jar:/usr/share/cloudstack-agent/lib/cloud-framework-jobs-4.5.2.jar:/usr/share/cloudstack-agent/lib/cloud-framework-managed-context-4.5.2.jar:/usr/share/cloudstack-agent/lib/cloud-framework-rest-4.5.2.jar:/usr/share/cloudstack-agent/lib/cloud-framework-security-4.5.2.jar:/usr/share/cloudstack-agent/lib/cloud-plugin-hypervisor-kvm-4.5.2.jar:/usr/share/cloudstack-agent/lib/cloud-plugin-network-ovs-4.5.2.jar:/usr/share/cloudstack-agent/lib/cloud-server-4.5.2.jar:/usr/share/cloudstack-agent/lib/cloud-utils-4.5.2.jar:/usr/share/cloudstack-agent/lib/commons-beanutils-core-1.7.0.jar:/usr/share/cloudstack-agent/lib/commons-codec-1.6.jar:/usr/share/cloudstack-agent/lib/commons-collections-3.2.1.jar:/usr/share/cloudstack-agent/lib/commons-configuration-1.8.jar:/usr/share/cloudstack-agent/lib/commons-daemon-1.0.10.jar:/usr/share/cloudstack-agent/lib/commons-dbcp-1.4.jar:/usr/share/cloudstack-agent/lib/commons-fileupload-1.2.jar:/usr/share/cloudstack-agent/lib/commons-httpclient-3.1.jar:/usr/share/cloudstack-agent/lib/commons-io-1.4.jar:/usr/share/cloudstack-agent/lib/commons-lang-2.6.jar:/usr/share/cloudstack-agent/lib/commons-logging-1.1.3.jar:/usr/share/cloudstack-agent/lib/commons-net-3.3.jar:/usr/share/cloudstack-agent/lib/commons-pool-1.6.jar:/usr/share/cloudstack-agent/lib/cxf-bundle-jaxrs-2.7.0.jar:/usr/share/cloudstack-agent/lib/dom4j-1.6.1.jar:/usr/share/cloudstack-agent/lib/ehcache-core-2.6.6.jar:/usr/share/cloudstack-agent/lib/ejb-api-3.0.jar:/usr/share/cloudstack-agent/lib/esapi-2.0.1.jar:/usr/share/cloudstack-agent/lib/geronimo-javamail_1.4_spec-1.7.1.jar:/usr/share/cloudstack-agent/lib/geronimo-servlet_3.0_spec-1.0.jar:/usr/share/cloudstack-agent/lib/gson-1.7.2.jar:/usr/share/cloudstack-agent/lib/guava-14.0-rc1.jar:/usr/share/cloudstack-agent/lib/httpclient-4.3.6.jar:/usr/share/cloudstack-agent/lib/httpcore-4.3.3.jar:/usr/share/cloudstack-agent/lib/jackson-annotations-2.1.1.jar:/usr/share/cloudstack-agent/lib/jackson-core-2.1.1.jar:/usr/share/cloudstack-agent/lib/jackson-core-asl-1.8.9.jar:/usr/share/cloudstack-agent/lib/jackson-databind-2.1.1.jar:/usr/share/cloudstack-agent/lib/jackson-jaxrs-json-provider-2.1.1.jar:/usr/share/cloudstack-agent/lib/jackson-mapper-asl-1.8.9.jar:/usr/share/cloudstack-agent/lib/jackson-module-jaxb-annotations-2.1.1.jar:/usr/share/cloudstack-agent/lib/jasypt-1.9.0.jar:/usr/share/cloudstack-agent/lib/java-ipv6-0.10.jar:/usr/share/cloudstack-agent/lib/javassist-3.12.1.GA.jar:/usr/share/cloudstack-agent/lib/javassist-3.18.1-GA.jar:/usr/share/cloudstack-agent/lib/javax.inject-1.jar:/usr/share/cloudstack-agent/lib/javax.persistence-2.0.0.jar:/usr/share/cloudstack-agent/lib/javax.ws.rs-api-2.0-m10.jar

重启服务


[root@node5 bin]# service cloudstack-agent status
cloudstack-agent (pid 6657) 正在运行...
[root@node5 bin]# service cloudstack-agent stop
Stopping Cloud Agent:

[root@node5 bin]# service cloudstack-agent status
cloudstack-agent (pid 6657) 正在运行..

ps ax |grep jsvc.exec 也验证了进程依然存在

眼前一亮的同时,也发现了之前使用restart带来的问题,stop不成功的问题被掩盖了~~~有没有懊恼? 不过来不及反思,接下来的问题还远不是这么简单......


[root@node5 bin]# kill -9 6655 6657
[root@node5 bin]# kill -9 6655 6657
-bash: kill: (6655) - 没有那个进程
-bash: kill: (6657) - 没有那个进程
[root@node5 bin]# service cloudstack-agent status
cloudstack-agent 已死,但 pid 文件仍存
[root@node5 bin]# rm /var/run/cloudstack-agent.pid
rm:是否删除普通文件 "/var/run/cloudstack-agent.pid"?y
[root@node5 bin]# service cloudstack-agent status
cloudstack-agent 已死,但是 subsys 被锁
[root@node5 bin]# service cloudstack-agent start
[root@node5 bin]# service cloudstack-agent status
cloudstack-agent (pid 109382) 正在运行...
[root@node5 bin]# netstat -antp |grep 8250
tcp    0   0 192.168.14.20:22220     192.168.14.10:8250     ESTABLISHED 109382/jsvc.exec

处理后状态恢复正常,但是libvirtd仍然无法杀掉, 很快netstat -antp |grep 8250 状态再次消失,cloudstack master平台监控主机记录由Up状态转为disconnect状态。不过毕竟不是down状态,较之前已经有了进步。

启动一个libvirtd -d看下,


[root@node5 bin]# libvirtd -d
[root@node5 bin]# ps ax |grep libvirtd
 6485 ?    R  863:37 libvirtd --daemon -l
130057 ?    Sl   0:38 libvirtd -d
28904 pts/0   S+   0:00 grep libvirtd

然后在cloudstack master平台上手工点击强制重新连接该主机,成功了。主机监控状态由disconnect转为Up,这时再次尝试杀掉6485仍然是不成功的,于是又在cloudstack master管理平台上尝试着点击操作了一下暂停vm命令,vm成功暂停。再返回服务器上观察原来hung死的libvirtd进程已经消失。


[root@node5 bin]# libvirtd -d
[root@node5 bin]# ps ax |grep libvirtd
130057 ?    Sl   0:38 libvirtd -d
28904 pts/0   S+   0:00 grep libvirtd

至此既恢复了平台对该主机的管控,也终止了libvirtd异常进程。问题初步归于cloudstack-agent在处理发送个libvirtd的信号上存在些小问题。以后再单独分析下jsvc进程,再现问题和根本解决。

问题反思

在处理服务异常的问题上,命令行参数不要用restart,用stop和kill来调试。说起来都是泪!

0
投稿

猜你喜欢

  • 本次测试包含以下10款ARP防火墙:360安全卫士ARP防火墙金山ARP防火墙风云防火墙antiarp-dnsARP防护大师Bitcomet
  • 虚拟基础架构时常遭遇简单解决方案效率低下的困扰。虚拟机执行匮乏的现状阻碍了虚拟化技术的普及。以下的十大虚拟机优化小窍门能帮助用户轻松解决每个
  • 我有一个感觉,不知道各位看观有没这种感觉。DedeCms在没组团队以前,我们BLT老大重点发展的应该是免费用户,也让V4.X辉煌一时。而在初
  • 国外的虚拟主机和国内的相比有不少的优点,例如不用去备案,价格相对低廉,稳定性也好,虽然从国内访问速度稍微慢一些,但的确是一个低成本创建一个网
  • 影响网站排名的因素据说有300多个,但对于网站优化来说,将这300多个因素都考虑到并且做得尽善尽美其实并不现实,也没必要。但下面列出的五个因
  • 面是Godaddy Windows主机支持的组件ADODB.CommandThe Command object is used to exe
  • C.对表的内容的一些说明            mysql> use po
  • 安装四个下载的软件包:#dpkg-i*.deb修改/etc/amavis/amavisd.conf:@inet_acl=qw(127/81.
  • 一、注册Google AdSense{$ggas_text$}单击申请按纽:开始注册:请在继续前确认所有信息都是正确的。在此之后不能更改付款
  • 现在主机商是遍地开花,怎么去选择主机?一个网站,选择一个质量好的虚拟主机就有关重要。因为很多主机商,在购买之前,和购买后所能实现的东西是根本
  • 首先自我介绍一下,我是华农人论坛站长,避免AD嫌疑,地址就不放上来了,是做高校论坛的,其间,做得不错,上来发表一点心得(两个月来,建站的一些
  • 通常地,大多数Web站点的设计目标都是:以最易接受的方式,为访问者提供即时的信息访问。在过去的几年中,越来越多的黑客、病毒和蠕虫带来的安全问
  • 最近几年来,基于Web的免费Email系统非常流行,当前,几个著名的免费Email网站基本上已经成为大多数人的选择,建立单纯提供免费Emai
  • 站点运营推广类的文章百度下可以搜索到36万条之多。站长对此类话题的关注程度可见一斑。许多中小站长面对一个新上线的站点并不知如何良好的运营和广
  • 据国家教育部统计,截至2009年7月1日,我国普通高校毕业生就业率为68%,约有200万毕业生未实现顺利就业。&ldquo;我们作为
  • 好几个站长朋友做资讯站就是网上下个CMS程序,然后复制粘贴。申请个域名就开始了他的网站之路,等待流量上来然后赚取广告费。如果这样就能成功。我
  • 服务器换了Win2003后,发现上传的附件如果超过204800字节后,系 统出错。这是因为win 2003默认上传字节为204800原因所致
  • 由于互联网普及,催生数量惊人的上网用户,如此庞大的网络用户群体,促生了新的经济增长方式,利用网络赚钱,让不少人走上了致富路。同时,还有一些网
  • 其实这三个模式大家肯定都各有了解, 天刃这里做的工作就是把大家都知道的三个东西放到一块,省去大家费力逐个翻阅的时间。08年闹得沸沸扬扬的百度
  • 很多网站设计的朋友,在未来的道路上,都会碰到这问题。下面我就将注册网络公司的相关步骤流程与相关信息整理出来,希望对要注册网络公司的朋友有所帮
手机版 网站运营 asp之家 www.aspxhome.com