package spec

Get desktop application:
View/edit binary Protocol Buffers messages

Used in: Provider

• string secretKey = 1

• string accessKey = 2

Used in: DNS

• string hostname = 1

• string endpoint = 2

Used in: Update.ApiEndpoint

• NoChange = 0

  NoChange represents the 1st case - no change is needed as the LB cluster is currently attached and the desired spec contains no changes.

• AttachingLoadBalancer = 1

  AttachingLoadBalancer represents 2nd case - the K8s cluster previously didn't have an LB cluster attached and the ports needed to communicate with the API server were exposed. After attaching an LB cluster to the existing K8s cluster the ports were closed and are no longer accessible, and thus we need to change the API endpoint.

• DetachingLoadBalancer = 2

  DetachingLoadBalancer represents 3rd. case - the K8s cluster had an existing LB cluster attached but the new state removed the LB cluster and thus the API endpoint needs to be changed back to one of the control nodes of the cluster.

• EndpointRenamed = 3

  EndpointRenamed represents the 4th. case - the K8s cluster has an existing LB cluster attached and also keeps it but the endpoint has changed in the desired state.

• MoveEndpoint = 4

  MoveEndpoint represents the 5th. case - the K8s cluster has an existing LB cluster attached, but it will be switched to a different LB cluster in the desired state.

Autoscaler configuration on per node pool basis.

Used in: DynamicNodePool, Update.MovedNodePoolFromAutoscaled, Update.TerraformerMoveNodePoolToAutoscaled

• int32 min = 1

  Minimum number of nodes in node pool.

• int32 max = 2

  Maximum number of nodes in node pool.

• int32 targetSize = 3

  Target size within the [Min, Max] range to which the number of nodes in the cluster should be matched againts to meet the compute demands.

Used in: Provider

• string subscriptionID = 1

• string tenantID = 2

• string clientID = 3

• string clientSecret = 4

Used in: Provider

• string token = 1

• optional string teamId = 2

Used in: Provider

• string token = 1

• string accountID = 2

ClusterInfo holds general information about the clusters.

Used in: K8scluster, LBcluster

• string name = 1

  Name of the cluster.

• string hash = 2

  Random hash of the cluster.

• repeated NodePool nodePools = 5

  Array of node pools this cluster is made of.

Used in: Config

• optional Clusters current = 1

• optional Workflow state = 4

• optional TaskEvent inFlight = 5

• optional Counters counters = 6

Used in: ClusterState

• optional K8scluster k8s = 1

• optional LoadBalancers loadBalancers = 2

Config holds data for a single manifest.

Used in: claudie.GetConfigResponse, claudie.ListConfigsResponse

• uint64 version = 1

  version of the config.

• string name = 2

  Config name - same as input manifest name.

• optional KubernetesContext k8sCtx = 3

  Information related within a k8s context.

• optional Manifest manifest = 4

  Client defined manifest.

• map<string, ClusterState> clusters = 5

  Clusters parsed from the supplied manifest.

Used in: ClusterState

• map<string, int64> k8sNodePoolScaleUpFailed = 1

Creates a new kubernetes cluster with the attached [LBcluster], if any.

Used in: Task

• optional K8scluster k8s = 1

  The desired state of the kuberenetes cluster to create. This field is required to be present.

• repeated LBcluster loadBalancers = 2

  The desired state of the LoadBalancers attached to the kubernetes cluster to create. This field is optional, if the kuberentes cluster does not have any loadbalacners.

DNS holds general information about the DNS records.

Used in: LBcluster, Update.TerraformerReplaceDns

• string dnsZone = 1

  DNS zone for the DNS records.

• string hostname = 2

  User specified hostname. (optional)

• optional Provider provider = 3

  Provider for the DNS records.

• string endpoint = 4

  The whole hostname of the DNS record.

• repeated AlternativeName alternativeNames = 5

  alternative names for which A records will be created in addition to the hostname.

Deletes an existing kubernetes cluster along with its attached [LBcluster], if any.

Used in: Task

• optional K8scluster k8s = 1

  The current state of the kuberentes cluster to delete. This field is required

• repeated LBcluster loadBalancers = 2

  The current state of the loadbalancers to delete, that are attached to the kubernetes cluster. This field depends on the above [k8s] field and is required if the kuberentes cluster has any loadbalancers attached.

DynamicNodePool represents dynamic node pool used in cluster.

Used in: NodePool

• string serverType = 1

  Machine type of the nodes.

• string image = 2

  OS image of the nodes.

• int32 storageDiskSize = 3

  Storage disk size of the nodes.

• string region = 4

  Geograhical region of the nodes.

• string zone = 5

  Datacentre zone for the nodes.

• int32 count = 6

  Number of the nodes in the node pool.

• optional Provider provider = 7

  Provider configuarion of the node pool.

• optional AutoscalerConf autoscalerConfig = 9

  Autoscaler configuration. (optional)

• optional MachineSpec machineSpec = 10

  MachineSpec configuration. (optional)

• string publicKey = 11

• string privateKey = 12

• string cidr = 14

  Cidr initially empty, will be set once the exising state is transferred.

• string externalNetworkName = 15

  Network Name with public IPs (required for Openstack)

• bool spot = 16

  Spot requests a discounted, pre-emptible (Spot) instance. Worker pools only. Currently supported on GCP, Verda, AWS, Azure, OCI.

Different error levels supported by each scheduled 'Work' in claudie.

Used in: StageDescription

• ERROR_FATAL = 0

  Error FATAL describes that whatever the operation/scheduled 'Work' failed in, no further work must be done on the state and whatever changes has been made, if any, must be reported back to be synced.

• ERROR_WARN = 1

  Error WARN describes that whatever the operation/scheduled 'Work' failed in, it may continue work on the state, even if partial changes were made into the state of the accepted 'Work' that is currently being processed. **Use with caution**, make sure that all of the function in the chain with this error level can handle partial state.

Used in: TaskEvent

• UNKNOWN = 0

• CREATE = 1

• UPDATE = 2

• DELETE = 3

Used in: Provider

• string apiKey = 1

• string apiSecret = 2

Used in: Workflow

• Workflow.Status status = 1

• string taskDescription = 2

• string stage = 3

• optional google.protobuf.Timestamp timestamp = 4

Used in: Provider

• string key = 1

• string project = 2

Used in: Provider

• string token = 1

InstallationProxy holds general information about a proxy used to build a K8s cluster.

Used in: K8scluster, Update.AnsiblerReplaceProxySettings

• string mode = 1

  Proxy installation mode.

• string endpoint = 2

  Proxy endpoint used to access the proxy.

• string noProxy = 3

  NoProxy is a comma-separated list of values that will be added to the default list of NoProxies used by Claudie. The default no proxy list is: 127.0.0.1/8,localhost,cluster.local,10.244.0.0/16,10.96.0.0/12" Any values specified will be appended to the end of the default NoProxy list. This field only has an effect if the Proxy is turned on.

K8scluster represents a single kubernetes cluster specified in the manifest.

Used in: Clusters, Create, Delete, TaskResult.UpdateState, Update.State

• optional ClusterInfo clusterInfo = 1

  General info about the cluster.

• string network = 2

  Network range for the VPN.

• string kubeconfig = 3

  Kubeconfig of the cluster.

• string kubernetes = 4

  Kubernetes version.

• optional InstallationProxy installationProxy = 5

  General information about a proxy used to build a K8s cluster.

Used in: claudie.UpsertManifestRequest, Config

• string name = 1

  name represents the input manifest resource name in Kubernetes

• string namespace = 2

  namespace represents the input manifest resource namespace in Kubernetes

LBcluster represents a single load balancer cluster specified in the manifest.

Used in: Create, Delete, LoadBalancers, Update.State, Update.TerraformerAddLoadBalancer

• optional ClusterInfo clusterInfo = 1

  General info about the cluster.

• repeated Role roles = 2

  Array of Load balancer roles.

• optional DNS dns = 3

  DNS information.

• string targetedK8s = 4

  Kubernetes cluster name of the cluster this load balancer is attached to.

• bool usedApiEndpoint = 5

  usedApiEndpoint signals which LB is actually used as the api endpoint. Claudie manifest validation does not allow for multiple API endpoints to be present, however. Internally claudie can deal with more loadbalancers that have the API role, this is due to the creation of intermediate representation which always add new infrastructure before making any further changes. To recognize which of them is actually used this field was added.

Used in: Clusters, TaskResult.UpdateState

• repeated LBcluster clusters = 1

MachineSpec further specifies the requested server type.

Used in: DynamicNodePool

• int32 cpuCount = 1

  Number of CPU cores.

• int32 memory = 2

  Memory in MB.

• int32 nvidiaGpuCount = 3

  Number of NVIDIA GPUs.

• string nvidiaGpuType = 4

  NVIDIA GPU accelerator type (required for GCP when using GPUs).

Used in: claudie.UpsertManifestRequest, Config

• string raw = 1

• bytes checksum = 2

• bytes lastAppliedChecksum = 3

• Manifest.State state = 4

• optional google.protobuf.Timestamp stateTimestamp = 5

Used in: Manifest

• Pending = 0

• Scheduled = 1

• Done = 2

• Error = 3

Node represents a single node from the node pool.

Used in: NodePool, Update.DeletedK8sNodes.Partial, Update.DeletedLoadBalancerNodes.Partial, Update.TerraformerAddK8sNodes.Existing, Update.TerraformerAddLoadBalancerNodes.Existing

• string name = 1

  Name of the node.

• string private = 2

  Private IP of the node in the VPN.

• string public = 3

  Public IP of the node.

• NodeType nodeType = 4

  Type of the node.

• string username = 5

  Username of a user with root privileges. Also used in SSH connection

• NodeStatus status = 6

  Status of the node.

• int32 sshPort = 7

  Per-node SSH port. Overrides NodePool.sshPort when > 0. Used by shared-IP / NAT nodes (e.g. CloudRift) where each VM gets its own mapped SSH port. 0 means inherit the node pool's sshPort (which itself defaults to 22).

• int32 wireguardPort = 8

  Per-node WireGuard endpoint port advertised to peers. Used when the host-mapped UDP port differs from the in-VM ListenPort (shared-IP / NAT nodes). 0 means use the default WireGuard listen port (51820).

NodePool represents a single node pool used in cluster.

Used in: ClusterInfo, Update.DeletedK8sNodes.WholeNodePool, Update.DeletedLoadBalancerNodes.WholeNodePool, Update.TerraformerAddK8sNodes.New, Update.TerraformerAddLoadBalancerNodes.New

• oneof Type

  Dynamic or Static node pool info.

  • DynamicNodePool dynamicNodePool = 1

  • StaticNodePool staticNodePool = 2

• string name = 3

  Name of static node pool.

• repeated Node nodes = 4

  Array of nodes in this node pool.

• bool isControl = 5

  Flag to differentiate between control and compute nodes.

• map<string, string> labels = 6

  User defined labels.

• repeated Taint taints = 7

  User defined taints.

• map<string, string> annotations = 8

  User definded annotations.

• int32 sshPort = 9

  SSH port for the nodes in this node pool. Default 0 means port 22.

Used in: Node

• Preparing = 0

  Preparing states that the node is not joined into the cluster yet but is in the creation process whether spawning the infrastructure or configuring it or simply pending to be scheduled for creation.

• Joined = 1

  Joined states that the node's infrastructure is spawned and configured and was successfully joined into the cluster.

• MarkedForDeletion = 2

  MarkedForDeletion states that the nodes will at some point in the future be deleted from the current state of the cluster, Depending on the target size of the nodepool the node may be replaced by claudie or may simply be deleted.

NodeType specifies the type of the node.

Used in: Node

• worker = 0

  Worker/Compute node.

• master = 1

  Master/Control node.

• apiEndpoint = 2

  Master node, which is also an API endpoint.

Used in: Provider

• string userOCID = 1

• string tenancyOCID = 2

• string keyFingerprint = 3

• string compartmentOCID = 4

• string privateKey = 5

Used in: Provider

• string clientId = 1

• string clientSecret = 2

• string serviceName = 3

• optional string endpoint = 4

Used in: Provider

• string authURL = 1

• string domainID = 2

• string projectID = 3

• string applicationCredentialID = 4

• string applicationCredentialSecret = 5

Used in: DNS, DynamicNodePool

• string specName = 1

• string cloudProviderName = 2

• oneof ProviderType

  = 5; deprecated hetznerdns.

  • GCPProvider gcp = 3

  • HetznerProvider hetzner = 4

  • OCIProvider oci = 6

  • AWSProvider aws = 7

  • AzureProvider azure = 8

  • CloudflareProvider cloudflare = 9

  • OpenstackProvider openstack = 11

  • ExoscaleProvider exoscale = 12

  • CloudRiftProvider cloudrift = 14

  • VerdaProvider verda = 15

  • OVHProvider ovh = 16

• optional TemplateRepository templates = 13

Role represents a single loadbalancer role from the manifest.

Used in: LBcluster, Update.TerraformerAddLoadBalancerRoles

• string name = 1

  Name of the role.

• string protocol = 2

  Protocol that load balancer uses to forward traffic. ["tcp", "udp"]

• int32 port = 3

  Port that load balancer will forward from.

• int32 targetPort = 4

  Port that load balancer will forward to.

• RoleType roleType = 6

  Type of the role.

• repeated string targetPools = 7

  Targeted nodes in Kubernetes clusters.

• optional Role.Settings settings = 8

  Additional settings for the role.

Used in: Role, Update.AnsiblerReplaceRoleInternalSettings

• bool proxyProtocol = 1

• bool stickySessions = 2

• int32 envoy_admin_port = 3

  required port for the envoy admin interface, on change will issue restart of the envoy proxy.

RoleType specifies the type of the role.

Used in: Role, Update.TerraformerReplaceRoleExternalSettings

• ApiServer = 0

  API server load balancer.

• Ingress = 1

  Ingress load balancer.

Used in: TaskEvent

• oneof StageKind

  • StageTerraformer terraformer = 1

  • StageAnsibler ansibler = 2

  • StageKubeEleven kubeEleven = 3

  • StageKuber kuber = 4

Used in: Stage

• optional StageDescription description = 1

• repeated StageAnsibler.SubPass subPasses = 2

Used in: StageAnsibler

• SubPassKind kind = 1

• optional StageDescription description = 2

Used in: SubPass

• INSTALL_NODE_REQUIREMENTS = 0

• INSTALL_VPN = 1

• DETERMINE_API_ENDPOINT_CHANGE = 2

• RECONCILE_LOADBALANCERS = 3

• REMOVE_CLAUDIE_UTILITIES = 4

• UPDATE_API_ENDPOINT = 5

• COMMIT_PROXY_ENVS = 6

• UPDATE_PROXY_ENVS_ON_NODES = 7

• CLEAR_PROXY_ENVS_ON_NODES = 8

Used in: StageAnsibler, StageAnsibler.SubPass, StageKubeEleven, StageKubeEleven.SubPass, StageKuber, StageKuber.SubPass, StageTerraformer, StageTerraformer.SubPass

• string about = 1

• ErrorLevel errorLevel = 2

Used in: Stage

• optional StageDescription description = 1

• repeated StageKubeEleven.SubPass subPasses = 2

Used in: StageKubeEleven

• SubPassKind kind = 1

• optional StageDescription description = 2

Used in: SubPass

• RECONCILE_CLUSTER = 0

• DESTROY_CLUSTER = 1

Used in: Stage

• optional StageDescription description = 1

• repeated StageKuber.SubPass subPasses = 2

Used in: StageKuber

• SubPassKind kind = 1

• optional StageDescription description = 2

Used in: SubPass

• CILIUM_RESTART = 0

• DELETE_NODES = 1

• PATCH_CLUSTER_INFO_CM = 2

• PATCH_KUBEADM = 4

• REMOVE_LB_SCRAPE_CONFIG = 5

• STORE_LB_SCRAPE_CONFIG = 6

• PATCH_NODES = 7

• DEPLOY_LONGHORN = 8

• ENABLE_LONGHORN_CA = 9

• DISABLE_LONGHORN_CA = 10

• RECONCILE_LONGHORN_STORAGE_CLASSES = 11

• DEPLOY_KUBELET_CSR_APPROVER = 12

Used in: Stage

• optional StageDescription description = 1

• repeated StageTerraformer.SubPass subPasses = 2

Used in: StageTerraformer

• SubPassKind kind = 1

• optional StageDescription description = 2

Used in: SubPass

• BUILD_INFRASTRUCTURE = 0

• UPDATE_INFRASTRUCTURE = 1

• DESTROY_INFRASTRUCTURE = 2

• API_PORT_ON_KUBERNETES = 3

StaticNodePool represents static node pool used in cluster.

Used in: NodePool

• map<string, string> nodeKeys = 1

  Map of keys for each static node in [<Node Endpoint>]<Key> form.

Enum used to replace data not present in static nodepools.

• STATIC_PROVIDER = 0

• STATIC_REGION = 1

• STATIC_ZONE = 2

Taint defines a custom defined taint for the node pools.

Used in: NodePool, Update.KuberPatchNodes.ListOfTaints

• string key = 1

  Taint key.

• string value = 2

  Taint value.

• string effect = 3

  Taint effect.

Used in: TaskEvent, Work

• oneof Do

  • Create create = 7

  • Update update = 8

  • Delete delete = 9

Used in: ClusterState

• string id = 1

• optional google.protobuf.Timestamp timestamp = 2

• Event event = 3

• optional Task task = 4

• string description = 5

• repeated Stage pipeline = 8

  Pipeline stages of the task through which it has to pass to be considered as done.

• uint32 currentStage = 9

  Index into the currently to be worked on pipeline stage.

• optional TaskEvent lowerPriority = 21

  LowerPriority Task is an optional field that is set whenever a higher priority task should be scheduled but the lower priority task should immediately follow after the higher priority task finishes. Only Higher priority tasks will have this field set. It is possible to create a chain of tasks with this that would follow based on priority until `nil` is hit.

• optional TaskResult.Error error = 1

  If the Error.Kind is FATAL no useful information is stored in the result. Otherwise, if error is not set or PARTIAL, the result will contain valid data,.

• oneof Result

  • TaskResult.None none = 2

  • TaskResult.UpdateState update = 3

  • TaskResult.ClearState clear = 4

ClearState specifies cluster IDs which should be cleared as changes processed by a scheduled task destroyed the infrastructure. Any non-nil value signals an update that should be done.

Used in: TaskResult

• optional bool k8s = 1

  If the kuberentes cluster is specified than all of its loadbalancers should be deleted even if they're not specified within this message.

• repeated string loadBalancersIDs = 2

  ID of the loadbalancers that should be cleared.

Used in: TaskResult

• Error.Kind kind = 1

• string description = 2

Used in: Error

• FATAL = 0

• PARTIAL = 1

None specifies that nothing should be done as a result of processing a task other than acknowledging that the task was processed, and thus moving to next task, if any.

Used in: TaskResult

(message has no fields)

UpdateState specifies the current state should be updated as changes may have been done to it. All of the values are optionals and the receiving end should make sense of them. Any non-nil value signals an update that should be handled.

Used in: TaskResult

• optional K8scluster k8s = 1

  If the [k8s] field is set then its current state should be merged/replaced with the state in this message.

• optional LoadBalancers loadBalancers = 2

  Updates the loadbalancers that are attached to the [k8s] cluster. The loadbalancers in this message should only update existing matching loadbalancers and or add newly created ones. If any loadbalancers are missing while they're present in the current state they should not be deleted, as for that there is an explicit task result [ClearState].

Used in: Provider

• string repository = 1

• optional string tag = 2

• string path = 3

• string commitHash = 4

Unreachable are the nodes of the kubernetes cluster and all of its loadbalancer clusters which are unreachable. If this is included in some message, one should omit connecting to those nodes during the processing of the pipeline, if relevant.

Used in: Update.DeleteLoadBalancer, Update.DeletedK8sNodes, Update.DeletedLoadBalancerNodes, Update.KuberDeleteK8sNodes, Update.TerraformerDeleteLoadBalancerNodes

• optional Unreachable.UnreachableNodePools kubernetes = 1

• map<string, Unreachable.UnreachableNodePools> loadbalancers = 2

Used in: UnreachableNodePools

• repeated string endpoints = 1

Used in: Unreachable

• map<string, ListOfNodeEndpoints> nodepools = 1

Used in: Task

• optional Update.State state = 1

  State that gets continiously updated as the task gets processed.

• oneof Delta

  Additional information while the [Update] message is passed through the build pipeline to give context what has been scheduled and what to process.

  • Update.None none = 2

  • Update.TerraformerAddLoadBalancer tfAddLoadBalancer = 3

    Terraformer messages to consume

  • Update.TerraformerAddLoadBalancerNodes tfAddLoadBalancerNodes = 4

  • Update.TerraformerReplaceDns tfReplaceDns = 5

  • Update.TerraformerAddK8sNodes tfAddK8sNodes = 6

  • Update.TerraformerAddLoadBalancerRoles tfAddLoadBalancerRoles = 7

  • Update.TerraformerDeleteLoadBalancerNodes tfDeleteLoadBalancerNodes = 8

  • Update.TerraformerMoveNodePoolToAutoscaled tfMoveNodePoolToAutoscaled = 9

  • Update.TerraformerMoveNodePoolFromAutoscaled tfMoveNodePoolFromAutoscaled = 10

  • Update.TerraformerReplaceRoleExternalSettings tfReplaceRoleExternalSettings = 12

  • Update.AnsiblerReplaceProxySettings ansReplaceProxy = 14

    Ansibler message to consume

  • Update.AnsiblerReplaceTargetPools ansReplaceTargetPools = 15

  • Update.AnsiblerReplaceRoleInternalSettings ansReplaceRoleInternalSettings = 16

  • Update.KuberPatchNodes kpatchNodes = 18

    Kuber messages to consume

  • Update.KuberDeleteK8sNodes kDeleteNodes = 19

  • Update.AddedLoadBalancer addedLoadBalancer = 21

    Consumed messages.

  • Update.AddedLoadBalancerNodes addedLoadBalancerNodes = 22

  • Update.ReplacedDns replacedDns = 23

  • Update.AddedK8sNodes addedK8sNodes = 24

  • Update.ReplacedProxySettings replacedProxy = 25

  • Update.PatchedNodes patchedNodes = 26

  • Update.AddedLoadBalancerRoles addedLoadBalancerRoles = 27

  • Update.ReplacedTargetPools replacedTargetPools = 28

  • Update.MovedNodePoolToAutoscaled movedNodePoolToAutoscaled = 29

  • Update.MovedNodePoolFromAutoscaled movedNodePoolFromAutoscaled = 30

  • Update.ReplacedRoleInternalSettings replacedRoleInternalSettings = 31

  • Update.ReplacedRoleExternalSettings replacedRoleExternalSettings = 32

  • Update.DeleteLoadBalancer deleteLoadBalancer = 34

    Deletions.

  • Update.DeletedK8sNodes deletedK8sNodes = 35

  • Update.DeletedLoadBalancerNodes deletedLoadBalancerNodes = 36

  • Update.DeleteLoadBalancerRoles deleteLoadBalancerRoles = 37

  • Update.ApiEndpoint apiEndpoint = 38

    Utilities.

  • Update.ApiPortOnCluster clusterApiPort = 39

  • Update.K8sOnlyApiEndpoint k8sApiEndpoint = 40

  • Update.UpgradeVersion upgradeVersion = 41

Used in: Update

• bool newNodePool = 1

• string nodepool = 2

• repeated string nodes = 3

Used in: Update

• string handle = 1

Used in: Update

• string handle = 1

• bool newNodePool = 2

• string nodePool = 3

• repeated string nodes = 4

Used in: Update

• string handle = 1

• repeated string roles = 2

AnsiblerReplaceProxySettings is a message that once processed by the Ansibler service a result is send back to the Manager, it will consume the message and further stages will only have a [ReplacedProxySettings] message to process. New, "desired", state needs only to be present in the ansibler stage and all other services need to only see what has been done which will already be present in the [State] of the [Update] message.

Used in: Update

• optional InstallationProxy proxy = 1

Used in: Update

• string handle = 1

• string role = 2

• int32 targetPort = 3

  updated settings.

• optional Role.Settings settings = 4

AnsiblerReplaceTargetPools is a message that once processed by the Ansibler service a result is send back to the Manager, it will consume the message and further stages will only have a [ReplacedTargetPools] message to process. New, "desired", state needs only to be present in the ansibler stage and all other services need to only see what has been done which will already be present in the [State] of the [Update] message.

Used in: Update

• string handle = 1

• map<string, AnsiblerReplaceTargetPools.TargetPools> roles = 2

  Role to target pools

Used in: AnsiblerReplaceTargetPools

• repeated string pools = 1

ApiEndpoint works with the [State] that is part of the [Update] message thus no "message consuming" needs to happen.

Used in: Update

• ApiEndpointChangeState state = 1

• string currentEndpointId = 2

• string desiredEndpointId = 3

ApiPortOnCluster works with the [State] that is part of the [Update] message thus no "message consuming" needs to happen.

Used in: Update

• bool open = 1

DeleteLoadBalancer works with the [State] that is part of the [Update] message, thus no "message consuming" needs to happen.

Used in: Update

• string handle = 1

• optional Unreachable unreachable = 2

  Unreachable infrastructure that is optionally set for the deletion process, in which presense, connecting to the unreachable infrastructure should be omitted.

DeleteLoadBalnacerRoles works with the [State] that is part of the [Update] message, thus no "message consuming" needs to happen.

Used in: Update

• string handle = 1

• repeated string roles = 2

Used in: Update

• optional Unreachable unreachable = 1

  Unreachable infrastructure that is optionally set for the deletion process, in which presense, connecting to the unreachable infrastructure should be omitted.

• oneof Kind

  • DeletedK8sNodes.WholeNodePool whole = 2

  • DeletedK8sNodes.Partial partial = 3

Used in: DeletedK8sNodes

• string nodepool = 1

  The nodepool which was modified with the deletion of the nodes, which is still present in the [State] passed along the [Update] msg.

• repeated Node nodes = 2

  Nodes that were deleted.

• map<string, string> staticNodeKeys = 3

  Map of keys for each node. Will only be set if the partially deleted nodes were part of a static nodepool. [<Node Endpoint>]<Key>.

Used in: DeletedK8sNodes

• optional NodePool nodepool = 1

  The nodepool that was deleted from the current state along with all of its nodes.

Used in: Update

• optional Unreachable unreachable = 1

  Unreachable infrastructure that is optionally set for the deletion process, in which presense, connecting to the unreachable infrastructure should be omitted.

• string handle = 2

  id of the loadbalancer.

• oneof Kind

  • DeletedLoadBalancerNodes.WholeNodePool whole = 3

  • DeletedLoadBalancerNodes.Partial partial = 4

Used in: DeletedLoadBalancerNodes

• string nodepool = 1

  The nodepool which was modified with the deletion of the nodes, which is still present in the [State] passed along the [Update] msg.

• repeated Node nodes = 2

  The nodes that were deleted.

• map<string, string> staticNodeKeys = 3

  Map of keys for each node. Will only be set if the partially deleted nodes were part of a static nodepool. [<Node Endpoint>]<Key>.

Used in: DeletedLoadBalancerNodes

• optional NodePool nodepool = 1

  The nodepool that was deleted from the current state along with all of its nodes.

K8sOnlyApiEndpoint works with the [State] that is part of the [Update] message, thus no "message consuming" needs to happen.

Used in: Update

• string nodepool = 1

• string node = 2

KuberDeleteK8sNodes is a message that is processed by the Kuber service and a result is send back to the Manager, it will consume the message and further stages will only have a [DeletedK8sNodes] message to process. Old, "current", state needs to only be present in the Kuber stage and all other services need to only see the changes after the deletion for reconciliation which will already be present in the [State] of the [Update] message.

Used in: Update

• bool withNodePool = 1

  Whether the NodePool itself was deleted from the cluster. Meaning that all nodes of the nodepool were deleted. Note that with autoscaling its possible to have a zero sized nodepool so this is only set if the nodepool is removed from the cluster.

• string nodepool = 2

  Affected Nodepool.

• repeated string nodes = 3

  Nodes that were deleted.

• optional Unreachable unreachable = 4

  Unreachable infrastructure that is optionally set for the deletion process, in which presense, connecting to the unreachable infrastructure should be omitted.

KuberPatchNodes is a message that once processed by the Kuber service and a result is send back to the Manager, it will consume the message and further stages will only have a [PatchedNodes] message to process. New, "desired", state needs only to be present in the kuber stage and all other services need to only see what has been done which will already be present in the [State] of the [Update] message.

Used in: Update

• optional KuberPatchNodes.AddBatch add = 1

• optional KuberPatchNodes.RemoveBatch remove = 2

Used in: KuberPatchNodes

• map<string, ListOfTaints> taints = 1

• map<string, MapOfLabels> labels = 2

• map<string, MapOfAnnotations> annotations = 3

Used in: RemoveBatch

• repeated string annotations = 1

Used in: RemoveBatch

• repeated string labels = 1

Used in: AddBatch, RemoveBatch

• repeated Taint taints = 1

Used in: AddBatch

• map<string, string> annotations = 1

Used in: AddBatch

• map<string, string> labels = 1

Used in: KuberPatchNodes

• map<string, ListOfTaints> taints = 1

• map<string, ListOfAnnotationKeys> annotations = 2

• map<string, ListOfLabelKeys> labels = 3

Used in: Update

• string nodepool = 1

• optional AutoscalerConf config = 2

Used in: Update

• string nodepool = 1

Used in: Update

(message has no fields)

Used in: Update

(message has no fields)

Used in: Update

• string handle = 1

• optional string oldApiEndpoint = 2

Used in: Update

(message has no fields)

Used in: Update

• string handle = 1

• string role = 2

Used in: Update

• string handle = 1

• string role = 2

Used in: Update

• string handle = 1

• map<string, ReplacedTargetPools.TargetPools> roles = 2

  Role to target pools.

Used in: ReplacedTargetPools

• repeated string pools = 1

State is the current state as known by the Manager service that is continiously updated and merged into the [Update] message as it traverses the build pipeline.

Used in: Update

• optional K8scluster k8s = 1

  The state of the kubernetes cluster to update.

• repeated LBcluster loadBalancers = 2

  The state of the loadbalancers to update.

TerraformerAddK8sNodes is a message that is processed by the Terraformer service and a result is send back to the Manager, it will consume the message and further stages will only have a [AddedK8sNodes] message to process. New, "desired", state needs only to be present in the Terraformer stage and all other services need to only see what has been done which will already be present in the [State] of the [Update] message.

Used in: Update

• oneof Kind

  • TerraformerAddK8sNodes.Existing existing = 1

  • TerraformerAddK8sNodes.New new = 2

Used in: TerraformerAddK8sNodes

• string nodepool = 1

  The nodepool which is to be modified with the addition of new nodes, which is already present in the [State] passed along the [Update] msg.

• repeated Node nodes = 2

  Nodes to be added.

Used in: TerraformerAddK8sNodes

• optional NodePool nodepool = 1

  The new nodepool which is to be added to the [State] passed along the [Update] msg.

TerraformerAddLoadBalancer is a message that once processed by the Terraformer service and a result is send back to the Manager, the manager will consume the message further stages will only have a [AddedLoadBalancer] message to process. New, "desired", state needs only to be present in the terraformer stage and all other services need to only see what has been done and ocasionally index the loadbalancer, which will already be present in the [State] of the [Update] message.

Used in: Update

• optional LBcluster handle = 1

TerraformerAddLoadBalancerNodes is a message that is processed by the Terraformer service and a result is send back to the Manager, it will consume the message and further stages will only have a [AddedLoadBalancerNodes] message to process. New, "desired", state needs only to be present in the Terraformer stage and all other services need to only see what has been done which will already be present in the [State] of the [Update] message.

Used in: Update

• string handle = 1

  Id of the loadbalancer.

• oneof Kind

  • TerraformerAddLoadBalancerNodes.Existing existing = 2

  • TerraformerAddLoadBalancerNodes.New new = 3

Used in: TerraformerAddLoadBalancerNodes

• string nodepool = 1

  The nodepool which is to be modified with the addition of new nodes, which is already present in the [State] passed along the [Update] msg.

• repeated Node nodes = 2

  Nodes to be added.

Used in: TerraformerAddLoadBalancerNodes

• optional NodePool nodepool = 1

  The new nodepool which is to be added to the [State] passed along the [Update] msg.

TerraformerAddLoadBalancerRoles is a message that is processed by the Terraformer service and a result is send back to the Manager, it will consume the message and further stages will only have a [AddedLoadBalancerRoles] message to process. New, "desired", state needs only to be present in the Terraformer stage and all other services need to only see what has been done which will already be present in the [State] of the [Update] message.

Used in: Update

• string handle = 1

• repeated Role roles = 2

TerraformerDeleteLoadBalancerNodes is a message that is processed by the Terraformer service and a result is send back to the Manager, it will consume the message and further stages will only have a [DeletedLoadBalancerNodes] message to process. Old, "current", state needs to only be present in the Terraformer stage and all other services need to only see the changes after the deletion for reconciliation which will already be present in the [State] of the [Update] message.

Used in: Update

• string handle = 1

  Id of the loadbalancer.

• bool withNodePool = 2

  Whether the NodePool itself was deleted from the cluster. Meaning that all nodes of the nodepool were deleted.

• string nodepool = 3

  Affected Nodepool.

• repeated string nodes = 4

  Nodes that were deleted.

• optional Unreachable unreachable = 5

  Unreachable infrastructure that is optionally set for the deletion process, in which presense, connecting to the unreachable infrastructure should be omitted.

Does exactly the same thing as [TerraformerMoveNodePoolToAutoscaled] but instead moves it to a fixed size nodepool.

Used in: Update

• string nodepool = 1

TerraformerMoveNodePoolToAutoscaled is a message that once processed by the Terraformer service a result is send back to the Manager, the manager will consume the message and further stages will only have a [MovedNodePoolToAutoscaled] message to process. New, "desired", state needs to only be present in the terraformer stage and all other services need to only see what has been done and ocasionally index the performed change on the nodepool which will already be present in the [State] of the [Update] message.

Used in: Update

• string nodepool = 1

• optional AutoscalerConf config = 2

TerraformerReplaceDns is a message that once processed by the Terraformer service and a result is send back to the Manager, it will consume the message and further stages will only have a [ReplacedDns] message to process. New, "desired", state needs only to be present in the terraformer stage and all other services need to only see what has been done and occasionally index the LoadBalancer, which will already be present in the [State] of the [Update] message.

Used in: Update

• string handle = 1

  Loadbalancer ID to which the new [Dns] is going to created for.

• optional DNS dns = 2

  The new dns to replace the old within [LoadBalancerId].

• optional string oldApiEndpoint = 3

  If the [Dns] is part of a Api loadbalancer this field will be set to give extra information for when changing the Api endpoint from the old to the new.

Used in: Update

• string handle = 1

• string role = 2

• RoleType roleType = 3

  updated settings.

• string protocol = 4

• int32 port = 5

UpgradeVersion works with the [State] that is part of the [Update] message, thus no "message consuming" needs to happen.

Used in: Update

• string version = 1

Used in: Provider

• string clientId = 1

• string clientSecret = 2

• optional string baseUrl = 3

The message that is used for the Message Queue.

• optional Task task = 1

  The task to be worked on.

• repeated google.protobuf.Any passes = 2

  What exactly should be done with the task.

Used in: ClusterState

• Workflow.Status status = 2

• string description = 3

  additional information describing the state.

• repeated FinishedWorkflow previous = 9

  Information about previously finished workflows.

• int32 ticksUntilRefresh = 10

  Ticks until refresh are the number of ticks until a refresh of the infrastructure is to be scheduled. Once this value reaches 0 the refresh of the infrastructure should be scheduled and this value should be reset back to its original starting value.

Used in: FinishedWorkflow, Workflow

• DONE = 0

  DONE indicates that the workflow for a scheduled [Task] has finished.

• ERROR = 1

  ERROR indicates that an error occurred while processing the scheduled [Task]

• IN_PROGRESS = 2

  IN_PROGRESS indicates that the [Task] is currently being worked on.

• WAIT_FOR_PICKUP = 3

  WAIT_FOR_PICKUP indicates that new a new [Task] was created or moved to the next stage to be worked on.