Virtualization management platform
Technical Field
The invention relates to the technical field of server management, in particular to a virtualization management platform.
Background
The existing server cluster can be managed only by machine-by-machine login, a management system is not used for carrying out unified management on the server cluster, when one or more servers have problems and faults, resources cannot be immediately redistributed according to the conditions of other existing servers, and the personalized requirements of users cannot be met.
Disclosure of Invention
The invention aims to overcome the defects of the prior art and provide a virtualization management platform which can acquire the conditions of all hosts in real time and perform message notification and task allocation management.
In order to achieve the purpose, the following technical scheme is adopted: a virtualization management platform, the management platform comprising: the system comprises a host machine, a switch, a storage disk cabinet and a management cluster, wherein the management cluster comprises a management end, a DNS end, an ETCD end and a ZABBIX end; the host machine is respectively connected with the switch, the storage disk cabinet and the management cluster circuit;
the management platform builds a message framework through the following steps:
s1, the host machine sends self state information including memory, CPU, hard disk space and VM condition information to the ETCD terminal at regular time;
s2, the management end checks the state information of the host machine through the ETCD end in real time, and sends an arranging task to the task point of the host machine according to the load condition;
s3, combining the DNS end with a management end to provide domain name interpretation service for VM construction and production;
the management platform builds a monitoring framework through the following steps:
s4, sending monitoring data to the ZABBIX terminal through the host machine, the switch and the storage disk cabinet through SNMP;
s5, the VM point monitors push alerts on team, future load data may be further manipulated in conjunction with automatic volume scaling.
Further, the management platform realizes the information viewing function through the following steps:
the host reports the memory, CPU, hard disk space and VM condition information to the ETCD end through info _ script.sh;
and the management terminal acquires information from the ETCD terminal in real time.
Further, the management platform realizes a task pushing function through the following steps:
the management terminal pushes the input task information to the ETCD terminal;
the host machine receives task information from the ETCD end in real time, and returns a result to the ETCD end after analysis and processing;
and the management terminal receives the return result from the ETCD terminal and displays the return result.
Preferably, the task information includes: creating VM, increasing VM hard disk, controlling VM power supply, increasing and decreasing CPU and memory capacity.
Preferably, the management platform implements the VM template and the configuration file by:
after the host machine analyzes the task information, copying a VMDK file from the template and generating a VMX configuration file, putting the VMDK file and the VMX configuration file into a corresponding folder, registering the VMX file and generating a corresponding VM;
after a VM power supply is started, IP, a host name and a corresponding environment file are configured through a built-in start.sh file, and the environment data can be obtained from an ETCD end;
SALT pushes the relevant configuration to complete the file configuration.
Further, the management platform realizes the system injection function through the following steps:
the management end translates the IP condition of the production environment into an MAC address through an IP2MAC function;
when the VM is started, the VM is translated into an IP address through a mac2IP function in start.sh and is connected to the outside, and in addition, start.sh initializes the data disk and deploys the account number and the salt agent.
Further, the management platform realizes the high-energy function through the following steps:
the host machine elects through the ETCD end to select the MASTER node;
the host machine reports the self state at regular time, checks the self state at regular time and reports the survival VM of the host machine at regular time;
if the host machine cannot detect the self state, the storage state is checked in the preset time, and if the storage state is detected to have a problem, the host machine is automatically restarted in the preset time;
meanwhile, the MASTER host can find a host with a fault in a preset time;
if the MASTER host judges that the failed host is not the DAS storage problem, the VM originally surviving by the failed host is immediately distributed to restart according to the memory free condition of other hosts;
if the MASTER host judges that the failed host is the DAS storage problem, the DAS storage is checked to determine whether heartbeat information submitted by the failed host exists or not in a preset time, and if not, VMs which originally survive of the failed host are immediately distributed according to the memory free condition of each host to restart; if so, the solution is solved by manual detection.
Further, the virtualization management platform further comprises:
and the host machine is connected with the storage disk cabinet through the optical fiber switch.
Compared with the prior art, the invention has the beneficial effects that: the invention relates to a virtualization management platform, which comprises: the system comprises a host machine, a switch, a storage disk cabinet and a management cluster, wherein the management cluster comprises a management end, a DNS end, an ETCD end and a ZABBIX end; the host machine is respectively connected with the switch, the storage disk cabinet and the management cluster circuit; the management platform builds a message framework through the following steps: s1, the host sends self state information including memory, CPU, hard disk space and VM condition information to the ETCD terminal at regular time; s2, the management end checks the state information of the host machine through the ETCD end in real time and sends an arranging task to the task point of the host machine according to the load condition; s3, combining the DNS end with the management end to provide domain name interpretation service for VM construction and production; the management platform builds a monitoring framework through the following steps: s4, the host machine, the switch and the storage disk cabinet send monitoring data to the ZABBIX terminal through SNMP; s5, the VM point monitors alarms pushed by a team, future load data can be further operated by combining automatic capacity expansion, and therefore the intelligent management function that the situation of each host can be known in real time, and information notification and task allocation can be carried out is achieved.
Drawings
FIG. 1 is a flow chart of a method for building a message framework by a virtualization management platform according to the present invention;
FIG. 2 is a flowchart of a method for building a monitoring framework by the virtualization management platform of the present invention;
FIG. 3 is a flowchart of a method for implementing an information viewing function by the virtualization management platform of the present invention;
FIG. 4 is a flowchart of a method for implementing a task pushing function by the virtualization management platform according to the present invention;
FIG. 5 is a flowchart of a method for implementing a system injection function by the virtualization management platform of the present invention;
FIG. 6 is a flowchart of a method for implementing high-power function by the virtualization management platform of the present invention.
Detailed Description
The invention will be described in detail with reference to the drawings and specific embodiments, which are illustrative and not limiting.
DETAILED DESCRIPTION OF EMBODIMENT (S) OF INVENTION
A virtualization management platform, the management platform comprising: the system comprises a host machine, a switch, a storage disk cabinet and a management cluster, wherein the management cluster comprises a management end, a DNS end, an ETCD end and a ZABBIX end; the host machine is respectively connected with the switch, the storage disk cabinet and the management cluster circuit;
as shown in fig. 1, the management platform builds a message framework by the following steps:
s101: the host machine sends self state information including memory, CPU, hard disk space and VM condition information to the ETCD terminal at regular time;
s102: the management end checks the state information of the host machine through the ETCD end in real time and sends an arranging task to the task point of the host machine according to the load condition;
s103: the DNS terminal is combined with the management terminal to provide domain name interpretation service for VM construction and production;
as shown in fig. 2, the management platform builds a monitoring framework by the following steps:
s201: the host machine, the switch and the storage disk cabinet send monitoring data to the ZABBIX terminal through SNMP;
s202: the VM point monitors push alerts on a team basis, future load data can be further manipulated in conjunction with automatic capacity scaling.
As shown in fig. 3, the management platform implements an information viewing function by:
s301: the host reports the memory, CPU, hard disk space and VM condition information to the ETCD end through info _ script.sh;
s302: and the management terminal acquires information from the ETCD terminal in real time.
As shown in fig. 4, the management platform implements a task pushing function by:
s401: the management terminal pushes the input task information to the ETCD terminal;
s402: the host machine receives task information from the ETCD end in real time, and returns a result to the ETCD end after analysis and processing;
s403: and the management terminal receives the return result from the ETCD terminal and displays the return result.
The task information includes: creating VM, increasing VM hard disk, controlling VM power supply, increasing and decreasing CPU and memory capacity.
The management platform realizes the VM template and the configuration file through the following steps:
after the host machine analyzes the task information, copying a VMDK file from the template and generating a VMX configuration file, putting the VMDK file and the VMX configuration file into a corresponding folder, registering the VMX file and generating a corresponding VM;
after a VM power supply is started, IP, a host name and a corresponding environment file are configured through a built-in start.sh file, and the environment data can be obtained from an ETCD end;
SALT pushes the relevant configuration to complete the file configuration.
As shown in fig. 5, the management platform implements the system injection function by:
s501: the management end translates the IP condition of the production environment into an MAC address through an IP2MAC function;
s502: when the VM is started, the VM is translated into an IP address through a mac2IP function in start.sh and is connected to the outside, and in addition, start.sh initializes the data disk and deploys the account number and the salt agent.
As shown in fig. 6, the management platform implements the high-power function by the following steps:
s601: the host machine elects through the ETCD end to select the MASTER node;
s602: the host machine reports the self state at regular time, checks the self state at regular time and reports the survival VM of the host machine at regular time;
such as: the host machine reports the self state every 10 seconds, checks the self state every 60 seconds, and reports the self survival VM every 120 seconds;
s603: if the host machine cannot detect the self state, the storage state is checked in the preset time, and if the storage state is detected to have a problem, the host machine is automatically restarted in the preset time; such as: if the host machine cannot detect the self state, checking the storage state every 10 seconds, and if the storage state is detected to have problems, automatically restarting after 120 seconds;
s604: the MASTER host can find a host with fault in a preset time; such as: the MASTER host can find out a host with a fault after 30 seconds;
s605: if the MASTER host judges that the failed host is not the DAS storage problem, the VM originally surviving by the failed host is immediately distributed to restart according to the memory free condition of other hosts;
s606: if the MASTER host judges that the failed host is the DAS storage problem, the DAS storage is checked to determine whether heartbeat information submitted by the failed host exists or not in a preset time, and if not, VMs which originally survive of the failed host are immediately distributed according to the memory free condition of each host to restart; if so, the solution is solved by manual detection.
The virtualization management platform further comprises:
the host machine is connected with the storage disk cabinet through the optical fiber switch, the optical fiber switch is mutually standby, and the optical fiber switch can be recycled well and is standby when in failure.
The functions described in this embodiment, if implemented in the form of software functional units and sold or used as independent products, may be stored in a storage medium readable by a computing device. Based on such understanding, part of the contribution of the embodiments of the present invention to the prior art or part of the technical solution may be embodied in the form of a software product stored in a storage medium and including instructions for causing a computing device (which may be a personal computer, a server, a mobile computing device, a network device, or the like) to execute all or part of the steps of the method described in the embodiments of the present invention. And the aforementioned storage medium includes: a U-disk, a removable hard disk, a Read-Only Memory (ROM), a Random Access Memory (RAM), a magnetic disk or an optical disk, and other various media capable of storing program codes. The embodiments are described in a progressive manner, each embodiment focuses on differences from other embodiments, and the same or similar parts among the embodiments are referred to each other.
The previous description of the disclosed embodiments is provided to enable any person skilled in the art to make or use the present invention. Various modifications to these embodiments will be readily apparent to those skilled in the art, and the generic principles defined herein may be applied to other embodiments without departing from the spirit or scope of the invention. Thus, the present invention is not intended to be limited to the embodiments shown herein but is to be accorded the widest scope consistent with the principles and novel features disclosed herein.