vSAN's repair activity, in which components of objects are rebuilt or relocated so they meet policy again. Causes include device, host or network failure (after the repair delay), maintenance mode, policy edits and rebalancing; watch it under Monitor > vSAN > Resyncing Objects.
Also called resynchronisation.
Read more: Broadcom TechDocs
In the Ultra Transcenders books
Each book explains resync in context, with comparison tables and the common traps.
Terms in this definition
- vSAN
VMware's software-defined storage, which combines local devices across a cluster's ESX hosts into one shared datastore. Available as OSA or ESA, VCF can use it as principal storage.
- Exclusions
Scopes taken out of a policy assignment; they can only narrow its coverage, never widen it.
- maintenance mode
A state a host is placed in, only on request, so that it can be serviced. Entry completes only once every running VM has been moved off or shut down, and vSAN hosts additionally require a choice about how their data is handled.
- RAMP
The approach used by the Cloud Adoption Framework's Manage methodology to structure cloud operations: teams, duties, processes and tools. It divides what the central platform handles from what each workload handles.
Related terms
- Automatic rebalance
A vSAN cluster option that shifts components across capacity devices when the gap in usage between any two exceeds a threshold, 30% unless changed. VCF advises turning it on to keep data evenly spread, despite more resync traffic when hosts join; it is distinct from reactive rebalancing, which begins once a device is 80% full.
- Data locality
Rule in a vSAN stretched cluster policy that pins an object to one site's fault domain (Preferred or Secondary) or leaves it unpinned (None). If the sites swap roles, pinned objects move, and more resynchronisation may follow.
- DRS Awareness of vSAN Stretched Cluster
On a vSAN stretched cluster, DRS automatically favours the site that can serve reads locally when it places VMs. After an HA restart, VMs do not move back until resynchronisation is complete. No configuration is offered.
- Service mesh
In HCX, the object that links a source compute profile to a destination one (together with their network profiles) and deploys paired service appliances, such as Interconnect and Network Extension, at both sites. Moving where appliances sit requires a resync, and adding further meshes scales out migration capacity.