feat: support disk smart config and status

Adds DiskSMARTConfig and SMARTStatus

Signed-off-by: Maja Bojarska <maja.bojarska@siderolabs.com>

Author: majabojarska

.golangci.yml

@@ -219,6 +219,7 @@ linters:
         - golang.zx2c4.com/wireguard/wgctrl
         - github.com/mdlayher/ethtool
         - github.com/secure-io/siv-go
+        - github.com/anatol/smart.go
       replace-local: true
       exclude-forbidden: false
       retract-allow-no-explanation: false
@@ -247,7 +248,18 @@ linters:
           disabled: true
     rowserrcheck: {}
     staticcheck:
-      checks: ["all", "-ST1000", "-ST1003", "-ST1016", "-ST1020", "-ST1021", "-ST1022", "-QF1001", "-QF1008"]
+      checks:
+        [
+          "all",
+          "-ST1000",
+          "-ST1003",
+          "-ST1016",
+          "-ST1020",
+          "-ST1021",
+          "-ST1022",
+          "-QF1001",
+          "-QF1008",
+        ]
     testpackage: {}
     unparam:
       check-exported: false

api/resource/definitions/block/block.proto

@@ -242,6 +242,45 @@ message ProvisioningSpec {
   FilesystemSpec filesystem_spec = 4;
 }
 
+// SMARTAttribute is a single ATA SMART attribute.
+message SMARTAttribute {
+  uint32 id = 1;
+  string name = 2;
+  uint32 current = 3;
+  uint32 worst = 4;
+  uint32 threshold = 5;
+  uint64 raw_value = 6;
+  bool failing = 7;
+}
+
+// SMARTStatusSpec is the spec for SMARTStatus resource.
+message SMARTStatusSpec {
+  // DevPath is the path to the block device, e.g. /dev/sda.
+  string dev_path = 1;
+  // DeviceType is the SMART device type: "nvme", "sata" or "scsi".
+  string device_type = 2;
+  // Healthy is the overall SMART health verdict computed by Talos.
+  bool healthy = 3;
+  // Message carries additional details, e.g. why the disk is unhealthy or that
+  // the probe was skipped because the disk was in standby.
+  string message = 4;
+  // PowerState reports the disk power mode observed during the last probe, e.g.
+  // "active", "idle" or "standby". SMART data is not refreshed while the disk is
+  // in standby (to avoid spinning it up).
+  string power_state = 5;
+  // Temperature is the disk temperature in degrees Celsius.
+  uint32 temperature = 6;
+  uint64 power_on_hours = 7;
+  uint64 power_cycles = 8;
+  // NVMe-specific fields.
+  uint32 percent_used = 9;
+  uint32 available_spare = 10;
+  uint32 critical_warning = 11;
+  uint64 media_errors = 12;
+  // Attributes is the raw ATA SMART attribute table (empty for NVMe/SCSI).
+  repeated SMARTAttribute attributes = 13;
+}
+
 // SwapStatusSpec is the spec for SwapStatuss resource.
 message SwapStatusSpec {
   string device = 1;

go.mod

@@ -3,6 +3,9 @@ module github.com/siderolabs/talos
 go 1.26.5
 
 replace (
+	// forked to set O_CLOEXEC on device open and add SataDevice.CheckPowerMode (skip standby disks)
+	github.com/anatol/smart.go => github.com/majabojarska/smart.go v0.0.0-20260707122632-d93e71f8295c
+
 	// forked coredns so we don't carry caddy and other stuff into the Talos
 	github.com/coredns/coredns => github.com/siderolabs/coredns v1.14.53
 
@@ -46,6 +49,7 @@ require (
 	github.com/Azure/azure-sdk-for-go/sdk/security/keyvault/azcertificates v1.5.0
 	github.com/Azure/azure-sdk-for-go/sdk/security/keyvault/azkeys v1.5.0
 	github.com/alexflint/go-filemutex v1.3.0
+	github.com/anatol/smart.go v0.0.0-20260707122632-d93e71f8295c
 	github.com/aws/aws-sdk-go-v2 v1.43.0
 	github.com/aws/aws-sdk-go-v2/config v1.32.31
 	github.com/aws/aws-sdk-go-v2/feature/ec2/imds v1.18.31

go.sum

@@ -72,6 +72,8 @@ github.com/aead/cmac v0.0.0-20160719120800-7af84192f0b1 h1:+JkXLHME8vLJafGhOH4ao
 github.com/aead/cmac v0.0.0-20160719120800-7af84192f0b1/go.mod h1:nuudZmJhzWtx2212z+pkuy7B6nkBqa+xwNXZHL1j8cg=
 github.com/alexflint/go-filemutex v1.3.0 h1:LgE+nTUWnQCyRKbpoceKZsPQbs84LivvgwUymZXdOcM=
 github.com/alexflint/go-filemutex v1.3.0/go.mod h1:U0+VA/i30mGBlLCrFPGtTe9y6wGQfNAWPBTekHQ+c8A=
+github.com/anatol/vmtest v0.0.0-20260313235012-c2b0479898b4 h1:E9KSKETdT7YYxgOAWMRshWLoMF40nV53ZFTPskPXUTI=
+github.com/anatol/vmtest v0.0.0-20260313235012-c2b0479898b4/go.mod h1:LQKSHEViwHPr4yRxK/6U+WpYY8ytg1uvVrjt/+PYgfI=
 github.com/antlr4-go/antlr/v4 v4.13.1 h1:SqQKkuVZ+zWkMMNkjy5FZe5mr5WURWnlpmOuzYWrPrQ=
 github.com/antlr4-go/antlr/v4 v4.13.1/go.mod h1:GKmUxMtwp6ZgGwZSva4eWPC5mS6vUAmOABFgjdkM7Nw=
 github.com/apparentlymart/go-cidr v1.1.1 h1:oEEk8CE0HP0YpHxsegk/TaOtR2FLHdWv4p3eM4ceUwg=
@@ -589,6 +591,8 @@ github.com/lmittmann/tint v1.2.0 h1:AogHRHy8HUJUnNJBHJlYa+fR4YY8mko2cnCp67xn9JY=
 github.com/lmittmann/tint v1.2.0/go.mod h1:HIS3gSy7qNwGCj+5oRjAutErFBl4BzdQP6cJZ0NfMwE=
 github.com/lucasb-eyer/go-colorful v1.4.0 h1:UtrWVfLdarDgc44HcS7pYloGHJUjHV/4FwW4TvVgFr4=
 github.com/lucasb-eyer/go-colorful v1.4.0/go.mod h1:R4dSotOR9KMtayYi1e77YzuveK+i7ruzyGqttikkLy0=
+github.com/majabojarska/smart.go v0.0.0-20260707122632-d93e71f8295c h1:9eLHH+oFEf6zuz49vUtL0hI0FRD5qGgXn1G4wBEr4cc=
+github.com/majabojarska/smart.go v0.0.0-20260707122632-d93e71f8295c/go.mod h1:PrHdljtLe/VXVRH+zAJOcPjHkOHjF8/T2Arblruu/ac=
 github.com/martinlindhe/base36 v1.1.1 h1:1F1MZ5MGghBXDZ2KJ3QfxmiydlWOGB8HCEtkap5NkVg=
 github.com/martinlindhe/base36 v1.1.1/go.mod h1:vMS8PaZ5e/jV9LwFKlm0YLnXl/hpOihiBxKkIoc3g08=
 github.com/mattn/go-colorable v0.1.15 h1:+u9SLTRGnXv73cEsnsmoZBom+dMU88B2M0aDcWy0/jY=
@@ -950,6 +954,8 @@ github.com/titanous/rocacheck v0.0.0-20171023193734-afe73141d399 h1:e/5i7d4oYZ+C
 github.com/titanous/rocacheck v0.0.0-20171023193734-afe73141d399/go.mod h1:LdwHTNJT99C5fTAzDz0ud328OgXz+gierycbcIx2fRs=
 github.com/tmc/grpc-websocket-proxy v0.0.0-20220101234140-673ab2c3ae75 h1:6fotK7otjonDflCTK0BCfls4SPy3NcCVb5dqqmbRknE=
 github.com/tmc/grpc-websocket-proxy v0.0.0-20220101234140-673ab2c3ae75/go.mod h1:KO6IkyS8Y3j8OdNO85qEYBsRPuteD+YciPomcXdrMnk=
+github.com/tmc/scp v0.0.0-20170824174625-f7b48647feef h1:7D6Nm4D6f0ci9yttWaKjM1TMAXrH5Su72dojqYGntFY=
+github.com/tmc/scp v0.0.0-20170824174625-f7b48647feef/go.mod h1:WLFStEdnJXpjK8kd4qKLwQKX/1vrDzp5BcDyiZJBHJM=
 github.com/transparency-dev/formats v0.1.1 h1:4bVHJc+KdBgpA1OJD1yjI+g0i5Z1graCppTMH8lWKJI=
 github.com/transparency-dev/formats v0.1.1/go.mod h1:qtZ8goRuJ8FTBG9c9+Bj0rn2rUG7eG/AUTkr+Aw3jFw=
 github.com/transparency-dev/merkle v0.0.2 h1:Q9nBoQcZcgPamMkGn7ghV8XiTZ/kRxn1yCG81+twTK4=

hack/release.toml

@@ -466,6 +466,32 @@ trim:
   enabled: true
   interval: 24h0m0s
 ```
+"""
+
+    [notes.disk_smart]
+        title = "Disk SMART Status"
+        description = """\
+Talos now collects SMART (Self-Monitoring, Analysis and Reporting Technology) health information from disks,
+exposed via the new `SMARTStatus` resource (one per disk, keyed by the disk device name):
+
+```bash
+talosctl get smart
+```
+
+The resource reports an overall health verdict, temperature, power-on hours and power cycles, NVMe wear
+indicators (percent used, available spare, critical warning, media errors) and, for ATA/SATA disks, the raw
+SMART attribute table with per-attribute thresholds.
+
+SMART collection is enabled by default. Rotational disks are never spun up just to be probed: a disk in
+standby is detected via ATA `CHECK POWER MODE` and its SMART data is left untouched until it is active again.
+Collection can be tuned or disabled via the new `DiskSMARTConfig` document:
+
+```yaml
+apiVersion: v1alpha1
+kind: DiskSMARTConfig
+enabled: true
+interval: 30m0s
+```
 """
 
     [notes.unattended_install]