Ibm afs secondary partition Virtualization

Ibm afs secondary partition Virtualization

IBM®AFS™The storage virtualization function of the storage Region network (SAN) allows you to migrate existing sub-partitions from one file system to another without restarting the file server or interrupting the client process. This article describes how to use the ibm afs File Server settings to migrate sub-partitions to different file servers and how to use the ibm afs san storage virtualization function to solve these problems.

Ibm afs Overview

AFS is a distributed file system that allows you to work together with hosts (clients and servers) and efficiently share file system resources across LAN and WAN.

Its main features include client/server architecture, management functions, scalability, location transparency, security, reliability, portability, and so on.

Main Components of IBM AFS

File Server: The file server is an AFS entity that provides a central disk memory card for a specific file set in the volume and allows authorized users running on the client to access these files.

Volume server: The volume server allows you to perform management tasks and detection on the AFS volume on the machine on which it is located. These operations include volume creation and deletion, renaming, dumping and recovery, modifying the copy Site List for read-only volumes, creating and adding new read-only volume images, creating and updating backup volumes, and listing partitions. all volumes, and check the volume status.

Basic OverSeer (BOS) Server: BOS Server is a management tool that runs on all file servers in cells. After the system is restarted, the server monitors the health status of the AFS agent in sequence, answers requests about their status, and restarts when they encounter a fault. It also accepts commands to restart, pause, and recover processes, and install the new server binary file.

Rescue server (salvager): Unlike other AFS servers, the rescue server only runs at the specified time. BOS Server will call salvager when the file Server, Volume Server, or both fail. Salvager will try to fix the faulty disk crash. As a system administrator, you can also call salvager as needed, even if the file server or volume server is not faulty.

Volume location server: The volume location server maintains and exports the volume location database. The Database tracks the server or server group where the volume instance is located. It supports operations including querying and returning volume location and status information, managing volume IDs, creating, deleting, and modifying VLDB entries.

Client component (Venus): the AFS Cache Management is implemented by the AFS client (afsd) and some daemon processes. It supports local disk cache and memory cache.

Secondary partition: the secondary partition is an ibm aix instance created using a physical device.®Partition. By convention, each partition is named/vicepx, where x is one or two English letters. The AFS partition can be named/vicepa or/vicepaa,/vicepb or/vicepbb until/vicepz or/vicepzz.

Volume: Disk Partitions used for AFS storage do not directly host various user files or directories. Instead, the subtree of the related system directory structure will be placed in a container called a volume.

Existing file server settings

As shown in settings 1 of the existing file server, multiple clients can access the volumes on fileserver1.

In the existing design, each file server has its own local storage, in which sub-partitions are created, and volumes and data are stored in local storage.

Figure 1. Existing file server settings

Existing file server settings

The main problem with the existing settings is that, no matter which file server, as long as the AFS administrator needs to upgrade the software, he must first move all the volumes on the server to other servers. Moving content includes copying all data in these volumes. After the software upgrade is completed, the volume is migrated back to the previous file server. The process is very long and may take several days to complete.

SAN storage virtualization solution

The new AFS Server is a larger machine and uses a SAN system with a higher degree of virtualization to store the data of the AFS sub-partition. We now have two or more AFS File servers to share the complete data storage in the SAN.

The new configuration shares SAN storage among multiple file servers. This helps reduce the time required for sub-partition migration because data does not need to be copied.

The existing AFS implementation can achieve this: conduct some configuration specific to AFS and deploy file servers. However, when moving the sub-partition, there will be downtime. Although the downtime is limited, you also need to restart the file server, which will affect the current and running client tasks.

Figure 2. New SAN-based file server settings

DirectionvosAdd two new commands to move all the sub-Partitions between different file servers.

The method is as follows:

vos detachpart -fromserver <machine name of server from where todetach> -partition <partition names to be moved> [-cell <cell name>][-noauth] [-localauth] [-verbose] [-timeout <timeout in seconds >] [-help]

The method is as follows:

vos attachpart -fromserver <machine name of server from where todetach> -toserver <machine name of server where to attach> -partition <partitionnames to be moved> [-cell <cell name>] [-noauth] [-localauth] [-verbose] [-timeout<timeout in seconds >] [-help]

vos detachpartCommand fromfromserverThe file servervos attachpartCommandfromserverPartition connections in the file servertoserverFile Server.

Because you do not need to restart the file server or interrupt client tasks, the migration process is very fast. They only get oneBusyMessage until the migration is complete.

We are supporting moving one or more sub-Partitions between different file servers.

For more details, please continue to read the highlights on the next page:

1 2 Next Page