feat: customize vm_id with libvirt metadata

By default the name chosen to define the Virtual Machine ID is the one
found in /proc/<pid>/cgroup/.

Many IaaS platform like Openstack use metadata in libvirt to customise
among other things the instance's ID.

Allow the user to use the metadata in libvirt to set the name of the
vm metrics and gives a better user experience.

Set "LIBVIRT_METADATA_URI" with the uri of the XML namespace identifier.
Set "LIBVIRT_METADATA_TOKEN" to choose the string used for VM ID. By
default, the token "name" is used.

Signed-off-by: Anthony Harivel <[email protected]>

go.mod

@@ -8,7 +8,9 @@ require (
 	github.com/HabanaAI/gohlml v1.16.0
 	github.com/NVIDIA/go-dcgm v0.0.0-20240118201113-3385e277e49f
 	github.com/NVIDIA/go-nvml v0.12.0-1
+	github.com/beevik/etree v1.4.0
 	github.com/cilium/ebpf v0.15.0
+	github.com/digitalocean/go-libvirt v0.0.0-20240709142323-d8406205c752
 	github.com/jaypipes/ghw v0.12.0
 	github.com/joho/godotenv v1.5.1
 	github.com/jszwec/csvutil v1.10.0
@@ -69,6 +71,7 @@ require (
 	github.com/prometheus/procfs v0.12.0 // indirect
 	github.com/spf13/pflag v1.0.5 // indirect
 	github.com/stretchr/testify v1.9.0 // indirect
+	golang.org/x/crypto v0.25.0 // indirect
 	golang.org/x/net v0.27.0 // indirect
 	golang.org/x/oauth2 v0.21.0 // indirect
 	golang.org/x/term v0.22.0 // indirect

go.sum

@@ -10,6 +10,8 @@ github.com/StackExchange/wmi v1.2.1 h1:VIkavFPXSjcnS+O8yTq7NI32k0R5Aj+v39y29VYDO
 github.com/StackExchange/wmi v1.2.1/go.mod h1:rcmrprowKIVzvc+NUiLncP2uuArMWLCbu9SBzvHz7e8=
 github.com/armon/go-socks5 v0.0.0-20160902184237-e75332964ef5 h1:0CwZNZbxp69SHPdPJAN/hZIm0C4OItdklCFmMRWYpio=
 github.com/armon/go-socks5 v0.0.0-20160902184237-e75332964ef5/go.mod h1:wHh0iHkYZB8zMSxRWpUBQtwG5a7fFgvEO+odwuTv2gs=
+github.com/beevik/etree v1.4.0 h1:oz1UedHRepuY3p4N5OjE0nK1WLCqtzHf25bxplKOHLs=
+github.com/beevik/etree v1.4.0/go.mod h1:cyWiXwGoasx60gHvtnEh5x8+uIjUVnjWqBvEnhnqKDA=
 github.com/beorn7/perks v1.0.1 h1:VlbKKnNfV8bJzeqoa4cOKqO6bYr3WgKZxO8Z16+hsOM=
 github.com/beorn7/perks v1.0.1/go.mod h1:G2ZrVWU2WbWT9wwq4/hrbKbnv/1ERSJQ0ibhJ6rlkpw=
 github.com/bits-and-blooms/bitset v1.13.0 h1:bAQ9OPNFYbGHV6Nez0tmNI0RiEu7/hxlYJRUA0wFAVE=
@@ -22,6 +24,8 @@ github.com/davecgh/go-spew v1.1.0/go.mod h1:J7Y8YcW2NihsgmVo/mv3lAwl/skON4iLHjSs
 github.com/davecgh/go-spew v1.1.1/go.mod h1:J7Y8YcW2NihsgmVo/mv3lAwl/skON4iLHjSsI+c5H38=
 github.com/davecgh/go-spew v1.1.2-0.20180830191138-d8f796af33cc h1:U9qPSI2PIWSS1VwoXQT9A3Wy9MM3WgvqSxFWenqJduM=
 github.com/davecgh/go-spew v1.1.2-0.20180830191138-d8f796af33cc/go.mod h1:J7Y8YcW2NihsgmVo/mv3lAwl/skON4iLHjSsI+c5H38=
+github.com/digitalocean/go-libvirt v0.0.0-20240709142323-d8406205c752 h1:NI7XEcHzWVvBfVjSVK6Qk4wmrUfoyQxCNpBjrHelZFk=
+github.com/digitalocean/go-libvirt v0.0.0-20240709142323-d8406205c752/go.mod h1:/Ok8PA2qi/ve0Py38+oL+VxoYmlowigYRyLEODRYdgc=
 github.com/emicklei/go-restful/v3 v3.11.0 h1:rAQeMHw1c7zTmncogyy8VvRZwtkmkZ4FxERmMY4rD+g=
 github.com/emicklei/go-restful/v3 v3.11.0/go.mod h1:6n3XBCmQQb25CM2LCACGz8ukIrRry+4bhvbpWn3mrbc=
 github.com/ghodss/yaml v1.0.0 h1:wQHKEahhL6wmXdzwWG11gIVCkOv05bNOh+Rxn0yngAk=
@@ -149,6 +153,8 @@ go.uber.org/goleak v1.3.0/go.mod h1:CoHD4mav9JJNrW/WLlf7HGZPjdw8EucARQHekz1X6bE=
 golang.org/x/crypto v0.0.0-20190308221718-c2843e01d9a2/go.mod h1:djNgcEr1/C05ACkg1iLfiJU5Ep61QUkGW8qpdssI0+w=
 golang.org/x/crypto v0.0.0-20191011191535-87dc89f01550/go.mod h1:yigFU9vqHzYiE8UmvKecakEJjdnWj3jj499lnFckfCI=
 golang.org/x/crypto v0.0.0-20200622213623-75b288015ac9/go.mod h1:LzIPMQfyMNhhGPhUkYOs5KpL4U8rLKemX1yGLhDgUto=
+golang.org/x/crypto v0.25.0 h1:ypSNr+bnYL2YhwoMt2zPxHFmbAN1KZs/njMG3hxUp30=
+golang.org/x/crypto v0.25.0/go.mod h1:T+wALwcMOSE0kXgUAnPAHqTLW+XHgcELELW8VaDgm/M=
 golang.org/x/exp v0.0.0-20240719175910-8a7402abbf56 h1:2dVuKD2vS7b0QIHQbpyTISPd0LeHDbnYEryqj5Q1ug8=
 golang.org/x/exp v0.0.0-20240719175910-8a7402abbf56/go.mod h1:M4RDyNAINzryxdtnbRXRL/OHtkFuWGRjvuhBJpk2IlY=
 golang.org/x/mod v0.2.0/go.mod h1:s0Qsj1ACt9ePp/hMypM3fl4fZqREWJwdYDEqhRiZZUA=

pkg/config/config.go

@@ -133,6 +133,12 @@ var (
 	// KubeConfig is used to start k8s client with the pod running outside the cluster
 	KubeConfig      = ""
 	EnableAPIServer = false
+
+	////////////////////////////////////
+
+	// Libvirt Metadata config
+	LibvirtMetadataURI   = getConfig("LIBVIRT_METADATA_URI", "")
+	LibvirtMetadataToken = getConfig("LIBVIRT_METADATA_TOKEN", "name")
 )
 
 // return local path to power model weight
@@ -423,3 +429,11 @@ func GetModelConfigMap() map[string]string {
 func SetGpuUsageMetric(metric string) {
 	GpuUsageMetric = metric
 }
+
+func GetLibvirtMetadataURI() string {
+	return LibvirtMetadataURI
+}
+
+func GetLibvirtMetadataToken() string {
+	return LibvirtMetadataToken
+}

pkg/libvirt/resolve_vm.go

@@ -18,10 +18,15 @@ package libvirt
 
 import (
 	"fmt"
+	"net/url"
 	"os"
 	"regexp"
 	"strings"
 
+	"github.com/beevik/etree"
+	"github.com/digitalocean/go-libvirt"
+
+	"github.com/sustainable-computing-io/kepler/pkg/config"
 	"github.com/sustainable-computing-io/kepler/pkg/utils"
 )
 
@@ -66,7 +71,14 @@ func getVMID(pid uint64, fileName string) (string, error) {
 	vmID = strings.ReplaceAll(vmID, "\\x2d", "-")
 	vmID = strings.ReplaceAll(vmID, ".scope", "")
 
+	if config.GetLibvirtMetadataURI() != "" {
+		// NOTE: use VM Metadata instead
+		if metaID, err := getMetadataVMID(vmID); metaID != "" && err == nil {
+			vmID = metaID
+		}
+	}
 	addVMIDToCache(pid, vmID)
+
 	return vmID, nil
 }
 
@@ -99,3 +111,50 @@ func addToNotExistCache(pid uint64) {
 	}
 	cacheNotExist[pid] = false
 }
+
+func getMetadataVMID(vmid string) (metaVmid string, err error) {
+	metadataURI := config.GetLibvirtMetadataURI()
+	domainName := ""
+	metaVmid = ""
+
+	parts := strings.Split(vmid, "-")
+	if len(parts) < 4 {
+		return metaVmid, fmt.Errorf("the vmid provided does not match machine-qemu-number-name : %v", vmid)
+	}
+
+	domainName = strings.Join(parts[3:], "-")
+
+	uri, _ := url.Parse(string(libvirt.QEMUSystem))
+	l, err := libvirt.ConnectToURI(uri)
+	if err != nil {
+		return metaVmid, fmt.Errorf("libvirt failed to connect: %w", err)
+	}
+
+	defer func() {
+		e := l.Disconnect()
+		if err == nil {
+			err = e
+		}
+	}()
+
+	d, err := l.DomainLookupByName(domainName)
+	if err != nil {
+		return metaVmid, fmt.Errorf("libvirt Domainlookup fail %w", err)
+	}
+
+	metaURI := libvirt.OptString{metadataURI}
+	// Domain Metadata Element = 0x2
+	metaXML, _ := l.DomainGetMetadata(d, 0x2, metaURI, 0)
+
+	doc := etree.NewDocument()
+	if err := doc.ReadFromString(metaXML); err != nil {
+		return metaVmid, fmt.Errorf("failed to read metadata XML: %w", err)
+	}
+
+	token := config.GetLibvirtMetadataToken()
+	nameElement := doc.FindElement(fmt.Sprintf("//%s", token))
+	if nameElement == nil {
+		return metaVmid, fmt.Errorf("no <%s> element found", token)
+	}
+	return nameElement.Text(), nil
+}