using Microsoft.Extensions.Options;
namespace ScadaLink.ClusterInfrastructure;
///
/// CI-004: Validates at startup. The values it
/// guards carry cluster-wide consequences — the design doc
/// (Component-ClusterInfrastructure.md) is emphatic that misconfiguring
/// them produces a total cluster shutdown or an indefinitely blocked singleton.
/// Registered with ValidateOnStart() so a bad appsettings.json
/// fails fast at boot rather than failing far from the cause.
///
public sealed class ClusterOptionsValidator : IValidateOptions
{
/// Split-brain resolver strategies safe for ScadaLink's two-node clusters.
private static readonly HashSet AllowedStrategies = new(StringComparer.OrdinalIgnoreCase)
{
"keep-oldest"
};
public ValidateOptionsResult Validate(string? name, ClusterOptions options)
{
var failures = new List();
if (options.SeedNodes is null || options.SeedNodes.Count == 0)
{
failures.Add("ClusterOptions.SeedNodes must contain at least one seed node.");
}
if (string.IsNullOrWhiteSpace(options.SplitBrainResolverStrategy)
|| !AllowedStrategies.Contains(options.SplitBrainResolverStrategy))
{
failures.Add(
$"ClusterOptions.SplitBrainResolverStrategy must be 'keep-oldest' for a two-node cluster; " +
$"'{options.SplitBrainResolverStrategy}' would risk a total cluster shutdown on a partition.");
}
if (options.MinNrOfMembers != 1)
{
failures.Add(
$"ClusterOptions.MinNrOfMembers must be 1 (was {options.MinNrOfMembers}); " +
"any other value blocks the cluster singleton after failover and halts all data collection.");
}
if (options.StableAfter <= TimeSpan.Zero)
{
failures.Add("ClusterOptions.StableAfter must be a positive duration.");
}
if (options.HeartbeatInterval <= TimeSpan.Zero)
{
failures.Add("ClusterOptions.HeartbeatInterval must be a positive duration.");
}
if (options.FailureDetectionThreshold <= TimeSpan.Zero)
{
failures.Add("ClusterOptions.FailureDetectionThreshold must be a positive duration.");
}
if (options.HeartbeatInterval >= options.FailureDetectionThreshold)
{
failures.Add(
$"ClusterOptions.HeartbeatInterval ({options.HeartbeatInterval}) must be well below " +
$"FailureDetectionThreshold ({options.FailureDetectionThreshold}); otherwise nodes are " +
"declared unreachable before a heartbeat can arrive.");
}
if (!options.DownIfAlone)
{
failures.Add(
"ClusterOptions.DownIfAlone must be true for the keep-oldest resolver "
+ "(Component-ClusterInfrastructure.md → Split-Brain Resolution); with it false the "
+ "oldest node can run as an isolated single-node cluster during a partition while the "
+ "younger node forms its own, producing two live clusters.");
}
return failures.Count > 0
? ValidateOptionsResult.Fail(failures)
: ValidateOptionsResult.Success;
}
}