Honor standby filtering only on request

[c-kr]

Cluster Health Standby Handling

Summary

This patch fixes standby host handling in cluster-health:

• host_in_standby() now returns a boolean instead of always returning None
• --nostandby is now actually honored

Before this fix, standby hosts were still included in the cluster member count even when --nostandby was specified.

Files

• checkvsphere/vcmd/clusterhealth.py

Quick Verification

export PYTHONPATH="$PWD"
export VC_HOST='vcenter.example.com'
export VC_USER='monitor@vsphere.local'
export VC_PASS='secret'
export CLUSTER_NAME='cluster1'

Use a cluster that currently has at least one host in standby.

python -m checkvsphere.cli cluster-health -s "$VC_HOST" -u "$VC_USER" -p "$VC_PASS" -nossl \
  --cluster-name "$CLUSTER_NAME" \
  --cluster-threshold 1:1 \
  --nostandby

Expected before fix:

• standby hosts are still counted as regular cluster members
• output/member count does not change even though --nostandby was passed

Expected after fix:

• standby hosts are excluded from cluster member counting
• output reflects the reduced member count when standby hosts exist

---

Assisted-by: Codex gpt-5.4 high
Draft status for early feedback, as not fully tested yet (I don't have a cluster for testing). Feedback appreciated.