BEFORE YOU BEGIN
Check before you begin
This public guide covers planning and acceptance. Multiple repositories, inter-site replication, centralized monitoring and automatic failover are different scopes; a single console or automatic switchover is not assumed for every platform. Contact support for internal topology and technical access guidance.
Visual note: no suitable real management screenshot without internal topology or customer information is available; an unrelated platform screen was not substituted. This work changed no site, repository, access or replication settings and ran no recovery operation.
- Product release, platforms and site owners
- Data scope, capacity and connectivity objectives
- Access boundaries and secret protection
- Isolated testing and recovery acceptance plan
011. Define site roles
Define source, backup-storage and recovery-site roles. Record data owners and operational responsibility. Two repositories at one site do not establish geographically separate protection.
Expected resultScope and acceptance results are documented with the responsible owner.
022. Verify supported scope
Validate backup-copy, replication and recovery methods with support for the release and platforms. Do not generalize a platform-specific inter-site replication guide to every VM, database or container workload.
Expected resultScope and acceptance results are documented with the responsible owner.
033. Plan data and repository mappings
Define protected data, capacity reserve and repository ownership at each site. Do not reuse an existing repository identity for a different physical data area. Catalog visibility does not prove data exists at the remote site.
Expected resultScope and acceptance results are documented with the responsible owner.
044. Assess connectivity and transfer windows
Plan initial transfer, incremental change volume, shared-link use and outage scenarios. Separate scheduling from guaranteed transfer rates. This public guide contains no internal addresses, port matrix or connection files; prepare the technical plan with support.
Expected resultScope and acceptance results are documented with the responsible owner.
055. Define access and operational boundaries
Define who can monitor and initiate operations at each site through authorized organizational procedures. Never place credentials in site inventories or public documents. Validate duplicate-operation prevention during connectivity loss against the supported method.
Expected resultScope and acceptance results are documented with the responsible owner.
066. Plan isolated site-recovery testing
Test target readability of the required recovery point, application dependencies and network isolation. Compare data acceptance and measured recovery time with objectives. Successful replication is not automatic failover or application acceptance.
Expected resultScope and acceptance results are documented with the responsible owner.
077. Document outage and return plans
Define decision ownership, monitoring, escalation and return-to-normal conditions during outages. Verify failover/failback support by platform and release. Contact destek@menabyte.com.tr for a detailed technical scenario; do not send sensitive topology in the initial message.
Expected resultScope and acceptance results are documented with the responsible owner.
COMMON QUESTIONS
Do multiple repositories establish multi-site disaster recovery?
Not alone. Physical separation, copy freshness, method support and recovery acceptance at the target site require separate validation.