Shulou(Shulou.com)06/02 Report--

SCCM Planning-High availability and redundancy

This article mainly discusses the high availability of ConfigMgr sites, site systems and clients, hoping to provide readers with a more comprehensive understanding of ConfigMgr high availability.

Unlike Exchange and SfB, applications such as ConfigMgr mainly exist in the form of interaction between the user and the client, which makes the application have very high requirements for server availability. After all, in the ConfigMgr scenario, the administrator wants to push a software or patch to the client, and these behaviors themselves may run in the background in a silent way, and ConfigMgr itself does not emphasize real-time in seconds or even minutes, so for the client After the failure of a service node, the client can choose to switch to the next node with the same function to request service, which is imperceptible to the user. Based on this objective situation, the so-called high availability in ConfigMgr is more like a kind of service redundancy from my personal point of view, because these forms of high availability themselves are not like the high availability in the traditional sense. After all, in many cases, when a node fails, the client can perceive that the service is unavailable, but it is as transparent to users as the traditional high availability.

Can provide highly available site components and role site components

Site databases: before ConfigMgr version 1602, site databases were highly available through SQL Server clusters, and ConfigMgr version 1602 began to support SQL Server Always On Availability Group as a high-availability solution

Secondary sites do not support the use of SQL Server clusters and SQL Server Always On Availability Group*

Introducing a central management site in the ConfigMgr hierarchy increases the recovery option in the event of an outage of the primary site below it, because data from all major sites in the hierarchy is replicated to the central management site, so when backups for the primary site are not available, the primary site can be restored through the central site.

SMS provider: redundancy can be improved by deploying multiple SMS providers. Site system role

Redundant roles are achieved by deploying multiple instances: management point, distribution point, state migration point, system health verification point, application directory Web service point, application directory network site, software update point, Report Service point

The multi-instance site system role also provides the ability to load balance. The multi-instance site system needs to be supported by ConfigMgr clients to reflect its redundancy. Publish service information to multiple locations for redundancy: it is typical to provide redundant information in multiple locations by publishing management point information to Active Directory through Active Directory integration and to DNS at the same time Unable to provide highly available site components and role site components site server: the site server currently does not have a high availability solution, and if the central management site or the main site goes down, it can be restored through ConfigMgr's own backup and recovery mechanism. If the secondary site is down, the only way is to reinstall the secondary site In SCCM 1806, the high availability scheme of independent primary site server was introduced. In SCCM 1810, the high availability scheme of central management site and sub-primary site server was introduced. The high availability of site server is accomplished by introducing a passive site server instance, when the passive site server instance uses the same site database and the same content library (Content Library) as the active site server in passive mode. The site server will not write any data site system role asset intelligent synchronization point Endpoint Protection point registration point registration agent point rollback state point out-of-band service point to the site database

These roles themselves do not assume critical services or related services have low requirements for high availability, so once the role involved goes down, you can remove the role in the hierarchy and reinstall the role to the new server.

ConfigMgr client

The ConfigMgr client also provides a highly available solution to maintain the normal operation of various functions as much as possible when there are problems with the service node or itself, mainly in the following aspects:

Autonomous operation of the client: there is no need for continuous and lasting information interaction between the client and the site system, and even if the client is temporarily unable to communicate with the site system, it will perform all kinds of tasks according to the received configuration information; the client can sense the unavailable service nodes and try to communicate with the remaining nodes from the list of service nodes to maintain the normal operation of all kinds of tasks.

Client self-repair: the client regularly evaluates its status and automatically repairs various components when needed

Client cache information: the client saves the configuration information and its own historical task running status in the local WMI namespace, and evaluates the current state if necessary according to the run plan to execute the client task

It is the original intention of this series of articles to refine the basic knowledge, precautions, operating mechanism and troubleshooting methods involved in SCCM for readers. There are already many articles in the network to refer to the information on the various components and functional deployment steps of SCCM, so this series of articles do not focus on providing deployment guides similar to Step-by-Step, please forgive me. At the same time, due to the limitations of personal ability and knowledge, there are inevitably some mistakes and mistakes in the article. I hope you can correct them. Thank you very much.

Welcome to subscribe "Shulou Technology Information " to get latest news, interesting things and hot topics in the IT industry, and controls the hottest and latest Internet news, technology news and IT industry trends.