DRS Rules

Configure DRS, affinity rules, and resource pools for vSphere clusters.

Reference Files

Read these before answering:

• references/drs-tuning.md -- migration thresholds, share calculations, affinity patterns
• references/cluster-design-patterns.md -- if cluster-level design context is relevant
• references/ha-patterns.md -- if HA interaction with DRS rules comes up
• references/vm-reservations.md -- if resource pool reservations or shares are discussed

Response Structure

1. Understand the placement constraint needed -- keep together, keep apart, pin to hosts, avoid hosts
2. Recommend the specific rule type: VM-VM affinity/anti-affinity, VM-Host affinity/anti-affinity
3. Discuss Required vs Preferred -- Required rules can block HA failover and DRS evacuation
4. Warn about HA interaction: Required anti-affinity with too few hosts can prevent restart
5. Provide specific configuration guidance for the scenario

Key Concepts

• DRS automation levels: Manual (recommendations only), Partially Automated (initial placement only), Fully Automated
• Migration threshold 1-5: Level 1 = only mandatory moves, Level 5 = all improvements including minor ones
• Required rules are hard constraints -- DRS will not violate them even during maintenance mode
• Preferred rules are soft constraints -- DRS tries to honor them but will violate if necessary
• Resource pools: reservations guarantee minimums, limits cap maximums, shares set relative priority during contention
• Anti-pattern: mixing VMs directly in a cluster with resource pools (orphaned VMs get unfair share allocation)

$ARGUMENTS