Status of Document: This document describes service offered as of July 2013.
TSM Backup for Virtual Environments (TSM4VE) is a new HFS service to allow the backup and recovery of Virtual Machines running in a VMware vSphere infrastructure. The service is based around IBM Tivoli's TSM4VE software client which communicates with a dedicated TSM server hosted within the HFS service's virtual infrastructure. The TSM4VE client operates a 'forever incremental' backup of virtual machine images. It is packaged and pre-configured by the HFS service to perform client-side deduplication which allows the online copy of the backups to reside on disk. A second, full ('hydrated') copy of the data is made to tape and kept offline. Clients can perform image-level restores of virtual machines or file-level restores from these backups. The service is owned by the HFS section within Infrastructure Services and is available initially on an application-only basis to managers of small-to-medium sized vSphere infrastructures.
User support for the TSM4VE service is provided initially via the HFS Service. Problems that cannot be resolved will be escalated to IBM Tivoli support as per the current software maintenance service agreements that exist between us (the HFS) and IBM. Support can only be given for matters relating to backup and recovery. No support is offered by the TSM4VE service for the operation or otherwise of the vSphere software and hardware infrastructure, or any other tools used to manage it.
The initial point of contact for user support is the Help Centre - mentioning TSM4VE will expedite your call, or by emailing firstname.lastname@example.org again mentioning TSM4VE in the subject line to ensure the most effective routing of your call.
Hours of Service
2.1 The service is offered as follows:
- In common with other HFS services, TSM4VE will normally run 24/7.
- Section 2 of the HFS Service Level Description details hours of service, serviceability and resilience for all HFS services including TSM4VE. However, in nutshell these can be summarised as below:
- 9am - 5pm on weekdays: the service operates with full technical support.
- At all other times any support is provided informally on a best-efforts basis. There is no formal provision for out-of-hours support.
- In common with other HFS services, TSM4VE is hosted on resilient hardware with multi-pathed data channels. Disk systems are RAID protected. As a result simple, single hardware faults should not affect service availability, but may result in a degradation of service performance. The failure of multiple hardware components, and or more complicated hardware or software faults may leave the service unavailable.
- Exclusions: Service maintenance may entail an interruption to service. This will be announced to the itss-announce maillist and the Systems Status page.
Service Level Targets
2.2 It is intended, as far as is possible, to maintain service availability at all times apart from exclusions listed under 2.1 above however, there are no formal targets.
2.3 The HFS will, in the event of a disaster affecting its own service, endeavour to provide a usable service as soon as possible. This may very well be a degraded service as hardware resources may be constrained.
2.4 The HFS will, in the event of a widespread disaster affecting multiple services, endeavour to provide a usable service as soon as possible. It is likely that in such an event, demand for its (restore) services will be considerable and a possible triaging of client recovery requirements may be employed.
Administration and Support
2.7 Technical support (operations and 2nd/3rd line user support) for the service is provided by IT Services; User support (1st line) provision is via IT Services Help Desk and HFS User Support, as described above.
2.8 Notification of scheduled maintenance, outages, and other information of general interest in relation to the service will be circulated on the itss-announce mailing list.
2.9 Service requests and fault reports relating to the service should be sent to the IT Services Help Desk.
3.1 Users are responsible for ensuring that the service is suitable for their needs; in particular that the service offers adequate recoverability for their systems.
3.2 Users should be competent in the setup, operation and management of a vSphere infrastructure and invoke best practices wherever possible.
3.3 Users should ensure that there is adequate network bandwidth available to ensure that data is transferred both to and from the service server within service limits (see 3.5 below).
3.4 Users should ensure that the configuration dictating that deduplication occurs client-side is not changed. They should ensure that the local machine performing this operation is of adequate specification to complete the task - see the prerequisites section .
3.5 Client sessions should not under normal circumstances run longer than 8 hours. Most client sessions should complete well within that cutoff time.
3.6 Users should try to ensure that use of the service remains reasonable such that the service can be accessed and shared by multiple users.
3.7 Users agree to make configuration or other changes as requested by the HFS service in order to continue to use ther service efficiently and effectively.
3.8 Users should report any defect, malfunction, or performance degradation of the service promptly to enable remedial action to be taken.