def generate_pod_spec_for_task():
# Primary containers do not require us to specify an image, the default image built for flyte tasks will get used.
primary_container = V1Container(name="primary")
# Note: for non-primary containers we must specify an image.
secondary_container = V1Container(name="secondary", image="alpine",)
secondary_container.command.extend(["/bin/sh"])
secondary_container.args.extend(
["-c", "echo hi pod world > {}".format(_SHARED_DATA_PATH)]
)
resources = V1ResourceRequirements(
requests={"cpu": "1", "memory": "100Mi"}, limits={"cpu": "1", "memory": "100Mi"}
)
primary_container.resources = resources
secondary_container = resources
shared_volume_mount = V1VolumeMount(name="shared-data", mount_path="/data",)
secondary_container.volumeMounts = [shared_volume_mount]
primary_container.volumeMounts = [shared_volume_mount]
pod_spec = V1PodSpec(
containers=[primary_container, secondary_container],
volumes=[
V1Volume(
name="shared-data", empty_dir=V1EmptyDirVolumeSource(medium="Memory")
)
],
)
return pod_spec
def test_fast_pod_task_serialization():
pod = Pod(
pod_spec=V1PodSpec(restart_policy="OnFailure",
containers=[V1Container(name="primary")]),
primary_container_name="primary",
)
@task(task_config=pod, environment={"FOO": "bar"})
def simple_pod_task(i: int):
pass
default_img = Image(name="default", fqn="test", tag="tag")
serialization_settings = SerializationSettings(
project="project",
domain="domain",
version="version",
env={"FOO": "baz"},
image_config=ImageConfig(default_image=default_img,
images=[default_img]),
fast_serialization_settings=FastSerializationSettings(enabled=True),
)
serialized = get_serializable(OrderedDict(), serialization_settings,
simple_pod_task)
assert serialized.template.k8s_pod.pod_spec["containers"][0]["args"] == [
"pyflyte-fast-execute",
"--additional-distribution",
"{{ .remote_package_path }}",
"--dest-dir",
"{{ .dest_dir }}",
"--",
"pyflyte-execute",
"--inputs",
"{{.input}}",
"--output-prefix",
"{{.outputPrefix}}",
"--raw-output-data-prefix",
"{{.rawOutputDataPrefix}}",
"--checkpoint-path",
"{{.checkpointOutputPrefix}}",
"--prev-checkpoint",
"{{.prevCheckpointPrefix}}",
"--resolver",
"flytekit.core.python_auto_container.default_task_resolver",
"--",
"task-module",
"tests.test_pod",
"task-name",
"simple_pod_task",
]
def test_map_pod_task_serialization():
pod = Pod(
pod_spec=V1PodSpec(restart_policy="OnFailure",
containers=[V1Container(name="primary")]),
primary_container_name="primary",
)
@task(task_config=pod, environment={"FOO": "bar"})
def simple_pod_task(i: int):
pass
mapped_task = map_task(simple_pod_task, metadata=TaskMetadata(retries=1))
default_img = Image(name="default", fqn="test", tag="tag")
serialization_settings = SerializationSettings(
project="project",
domain="domain",
version="version",
env={"FOO": "baz"},
image_config=ImageConfig(default_image=default_img,
images=[default_img]),
)
# Test that target is correctly serialized with an updated command
pod_spec = mapped_task.get_k8s_pod(serialization_settings).pod_spec
assert len(pod_spec["containers"]) == 1
assert pod_spec["containers"][0]["args"] == [
"pyflyte-map-execute",
"--inputs",
"{{.input}}",
"--output-prefix",
"{{.outputPrefix}}",
"--raw-output-data-prefix",
"{{.rawOutputDataPrefix}}",
"--checkpoint-path",
"{{.checkpointOutputPrefix}}",
"--prev-checkpoint",
"{{.prevCheckpointPrefix}}",
"--resolver",
"flytekit.core.python_auto_container.default_task_resolver",
"--",
"task-module",
"tests.test_pod",
"task-name",
"simple_pod_task",
]
assert {
"primary_container_name": "primary"
} == mapped_task.get_config(serialization_settings)
def get_pod_spec():
a_container = V1Container(name="a container", )
a_container.command = ["fee", "fi", "fo", "fum"]
a_container.volume_mounts = [
V1VolumeMount(
name="volume mount",
mount_path="some/where",
)
]
pod_spec = V1PodSpec(
restart_policy="OnFailure",
containers=[a_container,
V1Container(name="another container")])
return pod_spec
def make_pod(
name,
cmd,
port,
image,
image_pull_policy,
image_pull_secrets=None,
node_selector=None,
run_as_uid=None,
run_as_gid=None,
fs_gid=None,
supplemental_gids=None,
run_privileged=False,
allow_privilege_escalation=True,
env=None,
working_dir=None,
volumes=None,
volume_mounts=None,
labels=None,
annotations=None,
cpu_limit=None,
cpu_guarantee=None,
mem_limit=None,
mem_guarantee=None,
extra_resource_limits=None,
extra_resource_guarantees=None,
lifecycle_hooks=None,
init_containers=None,
service_account=None,
extra_container_config=None,
extra_pod_config=None,
extra_containers=None,
scheduler_name=None,
tolerations=None,
node_affinity_preferred=None,
node_affinity_required=None,
pod_affinity_preferred=None,
pod_affinity_required=None,
pod_anti_affinity_preferred=None,
pod_anti_affinity_required=None,
priority_class_name=None,
logger=None,
):
"""
Make a k8s pod specification for running a user notebook.
Parameters
----------
name:
Name of pod. Must be unique within the namespace the object is
going to be created in. Must be a valid DNS label.
image:
Image specification - usually a image name and tag in the form
of image_name:tag. Same thing you would use with docker commandline
arguments
image_pull_policy:
Image pull policy - one of 'Always', 'IfNotPresent' or 'Never'. Decides
when kubernetes will check for a newer version of image and pull it when
running a pod.
image_pull_secrets:
Image pull secrets - a list of references to Kubernetes Secret resources
with credentials to pull images from image registries. This list can
either have strings in it or objects with the string value nested under
a name field.
port:
Port the notebook server is going to be listening on
cmd:
The command used to execute the singleuser server.
node_selector:
Dictionary Selector to match nodes where to launch the Pods
run_as_uid:
The UID used to run single-user pods. The default is to run as the user
specified in the Dockerfile, if this is set to None.
run_as_gid:
The GID used to run single-user pods. The default is to run as the primary
group of the user specified in the Dockerfile, if this is set to None.
Setting this parameter requires that *feature-gate* **RunAsGroup** be enabled,
otherwise the effective GID of the pod will be 0 (root). In addition, not
setting `run_as_gid` once feature-gate RunAsGroup is enabled will also
result in an effective GID of 0 (root).
fs_gid
The gid that will own any fresh volumes mounted into this pod, if using
volume types that support this (such as GCE). This should be a group that
the uid the process is running as should be a member of, so that it can
read / write to the volumes mounted.
supplemental_gids:
A list of GIDs that should be set as additional supplemental groups to
the user that the container runs as. You may have to set this if you are
deploying to an environment with RBAC/SCC enforced and pods run with a
'restricted' SCC which results in the image being run as an assigned
user ID. The supplemental group IDs would need to include the
corresponding group ID of the user ID the image normally would run as.
The image must setup all directories/files any application needs access
to, as group writable.
run_privileged:
Whether the container should be run in privileged mode.
allow_privilege_escalation:
Controls whether a process can gain more privileges than its parent process.
env:
Dictionary of environment variables.
volumes:
List of dictionaries containing the volumes of various types this pod
will be using. See k8s documentation about volumes on how to specify
these
volume_mounts:
List of dictionaries mapping paths in the container and the volume(
specified in volumes) that should be mounted on them. See the k8s
documentaiton for more details
working_dir:
String specifying the working directory for the notebook container
labels:
Labels to add to the spawned pod.
annotations:
Annotations to add to the spawned pod.
cpu_limit:
Float specifying the max number of CPU cores the user's pod is
allowed to use.
cpu_guarentee:
Float specifying the max number of CPU cores the user's pod is
guaranteed to have access to, by the scheduler.
mem_limit:
String specifying the max amount of RAM the user's pod is allowed
to use. String instead of float/int since common suffixes are allowed
mem_guarantee:
String specifying the max amount of RAM the user's pod is guaranteed
to have access to. String ins loat/int since common suffixes
are allowed
lifecycle_hooks:
Dictionary of lifecycle hooks
init_containers:
List of initialization containers belonging to the pod.
service_account:
Service account to mount on the pod. None disables mounting
extra_container_config:
Extra configuration (e.g. envFrom) for notebook container which is not covered by parameters above.
extra_pod_config:
Extra configuration (e.g. tolerations) for pod which is not covered by parameters above.
extra_containers:
Extra containers besides notebook container. Used for some housekeeping jobs (e.g. crontab).
scheduler_name:
The pod's scheduler explicitly named.
tolerations:
Tolerations can allow a pod to schedule or execute on a tainted node. To
learn more about pod tolerations, see
https://kubernetes.io/docs/concepts/configuration/taint-and-toleration/.
Pass this field an array of "Toleration" objects.*
* https://kubernetes.io/docs/reference/generated/kubernetes-api/v1.10/#nodeselectorterm-v1-core
node_affinity_preferred:
Affinities describe where pods prefer or require to be scheduled, they
may prefer or require a node to have a certain label or be in proximity
/ remoteness to another pod. To learn more visit
https://kubernetes.io/docs/concepts/configuration/assign-pod-node/
Pass this field an array of "PreferredSchedulingTerm" objects.*
* https://kubernetes.io/docs/reference/generated/kubernetes-api/v1.10/#preferredschedulingterm-v1-core
node_affinity_required:
Affinities describe where pods prefer or require to be scheduled, they
may prefer or require a node to have a certain label or be in proximity
/ remoteness to another pod. To learn more visit
https://kubernetes.io/docs/concepts/configuration/assign-pod-node/
Pass this field an array of "NodeSelectorTerm" objects.*
* https://kubernetes.io/docs/reference/generated/kubernetes-api/v1.10/#nodeselectorterm-v1-core
pod_affinity_preferred:
Affinities describe where pods prefer or require to be scheduled, they
may prefer or require a node to have a certain label or be in proximity
/ remoteness to another pod. To learn more visit
https://kubernetes.io/docs/concepts/configuration/assign-pod-node/
Pass this field an array of "WeightedPodAffinityTerm" objects.*
* https://kubernetes.io/docs/reference/generated/kubernetes-api/v1.10/#weightedpodaffinityterm-v1-core
pod_affinity_required:
Affinities describe where pods prefer or require to be scheduled, they
may prefer or require a node to have a certain label or be in proximity
/ remoteness to another pod. To learn more visit
https://kubernetes.io/docs/concepts/configuration/assign-pod-node/
Pass this field an array of "PodAffinityTerm" objects.*
* https://kubernetes.io/docs/reference/generated/kubernetes-api/v1.10/#podaffinityterm-v1-core
pod_anti_affinity_preferred:
Affinities describe where pods prefer or require to be scheduled, they
may prefer or require a node to have a certain label or be in proximity
/ remoteness to another pod. To learn more visit
https://kubernetes.io/docs/concepts/configuration/assign-pod-node/
Pass this field an array of "WeightedPodAffinityTerm" objects.*
* https://kubernetes.io/docs/reference/generated/kubernetes-api/v1.10/#weightedpodaffinityterm-v1-core
pod_anti_affinity_required:
Affinities describe where pods prefer or require to be scheduled, they
may prefer or require a node to have a certain label or be in proximity
/ remoteness to another pod. To learn more visit
https://kubernetes.io/docs/concepts/configuration/assign-pod-node/
Pass this field an array of "PodAffinityTerm" objects.*
* https://kubernetes.io/docs/reference/generated/kubernetes-api/v1.10/#podaffinityterm-v1-core
priority_class_name:
The name of the PriorityClass to be assigned the pod. This feature is Beta available in K8s 1.11.
"""
pod = V1Pod()
pod.kind = "Pod"
pod.api_version = "v1"
pod.metadata = V1ObjectMeta(name=name,
labels=(labels or {}).copy(),
annotations=(annotations or {}).copy())
pod.spec = V1PodSpec(containers=[])
pod.spec.restart_policy = 'OnFailure'
if image_pull_secrets is not None:
# image_pull_secrets as received by the make_pod function should always
# be a list, but it is allowed to have "a-string" elements or {"name":
# "a-string"} elements.
pod.spec.image_pull_secrets = [
V1LocalObjectReference(name=secret_ref) if type(secret_ref) == str
else get_k8s_model(V1LocalObjectReference, secret_ref)
for secret_ref in image_pull_secrets
]
if node_selector:
pod.spec.node_selector = node_selector
if lifecycle_hooks:
lifecycle_hooks = get_k8s_model(V1Lifecycle, lifecycle_hooks)
# There are security contexts both on the Pod level or the Container level.
# The security settings that you specify for a Pod apply to all Containers
# in the Pod, but settings on the container level can override them.
#
# We configure the pod to be spawned on the container level unless the
# option is only available on the pod level, such as for those relating to
# the volumes as compared to the running user of the container. Volumes
# belong to the pod and are only mounted by containers after all.
#
# ref: https://kubernetes.io/docs/tasks/configure-pod-container/security-context/
# ref: https://kubernetes.io/docs/reference/generated/kubernetes-api/v1.16/#securitycontext-v1-core (container)
# ref: https://kubernetes.io/docs/reference/generated/kubernetes-api/v1.16/#podsecuritycontext-v1-core (pod)
pod_security_context = V1PodSecurityContext()
if fs_gid is not None:
pod_security_context.fs_group = int(fs_gid)
if supplemental_gids is not None and supplemental_gids:
pod_security_context.supplemental_groups = [
int(gid) for gid in supplemental_gids
]
# Only clutter pod spec with actual content
if not all([e is None for e in pod_security_context.to_dict().values()]):
pod.spec.security_context = pod_security_context
container_security_context = V1SecurityContext()
if run_as_uid is not None:
container_security_context.run_as_user = int(run_as_uid)
if run_as_gid is not None:
container_security_context.run_as_group = int(run_as_gid)
if run_privileged:
container_security_context.privileged = True
if not allow_privilege_escalation:
container_security_context.allow_privilege_escalation = False
# Only clutter container spec with actual content
if all([e is None for e in container_security_context.to_dict().values()]):
container_security_context = None
# Transform a dict into valid Kubernetes EnvVar Python representations. This
# representation shall always have a "name" field as well as either a
# "value" field or "value_from" field. For examples see the
# test_make_pod_with_env function.
prepared_env = []
for k, v in (env or {}).items():
if type(v) == dict:
if not "name" in v:
v["name"] = k
prepared_env.append(get_k8s_model(V1EnvVar, v))
else:
prepared_env.append(V1EnvVar(name=k, value=v))
notebook_container = V1Container(
name='notebook',
image=image,
working_dir=working_dir,
ports=[V1ContainerPort(name='notebook-port', container_port=port)],
env=prepared_env,
args=cmd,
image_pull_policy=image_pull_policy,
lifecycle=lifecycle_hooks,
resources=V1ResourceRequirements(),
volume_mounts=[
get_k8s_model(V1VolumeMount, obj) for obj in (volume_mounts or [])
],
security_context=container_security_context,
)
if service_account is None:
# This makes sure that we don't accidentally give access to the whole
# kubernetes API to the users in the spawned pods.
pod.spec.automount_service_account_token = False
else:
pod.spec.service_account_name = service_account
notebook_container.resources.requests = {}
if cpu_guarantee:
notebook_container.resources.requests['cpu'] = cpu_guarantee
if mem_guarantee:
notebook_container.resources.requests['memory'] = mem_guarantee
if extra_resource_guarantees:
notebook_container.resources.requests.update(extra_resource_guarantees)
notebook_container.resources.limits = {}
if cpu_limit:
notebook_container.resources.limits['cpu'] = cpu_limit
if mem_limit:
notebook_container.resources.limits['memory'] = mem_limit
if extra_resource_limits:
notebook_container.resources.limits.update(extra_resource_limits)
if extra_container_config:
notebook_container = update_k8s_model(
target=notebook_container,
changes=extra_container_config,
logger=logger,
target_name="notebook_container",
changes_name="extra_container_config",
)
pod.spec.containers.append(notebook_container)
if extra_containers:
pod.spec.containers.extend(
[get_k8s_model(V1Container, obj) for obj in extra_containers])
if tolerations:
pod.spec.tolerations = [
get_k8s_model(V1Toleration, obj) for obj in tolerations
]
if init_containers:
pod.spec.init_containers = [
get_k8s_model(V1Container, obj) for obj in init_containers
]
if volumes:
pod.spec.volumes = [get_k8s_model(V1Volume, obj) for obj in volumes]
else:
# Keep behaving exactly like before by not cleaning up generated pod
# spec by setting the volumes field even though it is an empty list.
pod.spec.volumes = []
if scheduler_name:
pod.spec.scheduler_name = scheduler_name
node_affinity = None
if node_affinity_preferred or node_affinity_required:
node_selector = None
if node_affinity_required:
node_selector = V1NodeSelector(node_selector_terms=[
get_k8s_model(V1NodeSelectorTerm, obj)
for obj in node_affinity_required
], )
preferred_scheduling_terms = None
if node_affinity_preferred:
preferred_scheduling_terms = [
get_k8s_model(V1PreferredSchedulingTerm, obj)
for obj in node_affinity_preferred
]
node_affinity = V1NodeAffinity(
preferred_during_scheduling_ignored_during_execution=
preferred_scheduling_terms,
required_during_scheduling_ignored_during_execution=node_selector,
)
pod_affinity = None
if pod_affinity_preferred or pod_affinity_required:
weighted_pod_affinity_terms = None
if pod_affinity_preferred:
weighted_pod_affinity_terms = [
get_k8s_model(V1WeightedPodAffinityTerm, obj)
for obj in pod_affinity_preferred
]
pod_affinity_terms = None
if pod_affinity_required:
pod_affinity_terms = [
get_k8s_model(V1PodAffinityTerm, obj)
for obj in pod_affinity_required
]
pod_affinity = V1PodAffinity(
preferred_during_scheduling_ignored_during_execution=
weighted_pod_affinity_terms,
required_during_scheduling_ignored_during_execution=
pod_affinity_terms,
)
pod_anti_affinity = None
if pod_anti_affinity_preferred or pod_anti_affinity_required:
weighted_pod_affinity_terms = None
if pod_anti_affinity_preferred:
weighted_pod_affinity_terms = [
get_k8s_model(V1WeightedPodAffinityTerm, obj)
for obj in pod_anti_affinity_preferred
]
pod_affinity_terms = None
if pod_anti_affinity_required:
pod_affinity_terms = [
get_k8s_model(V1PodAffinityTerm, obj)
for obj in pod_anti_affinity_required
]
pod_anti_affinity = V1PodAffinity(
preferred_during_scheduling_ignored_during_execution=
weighted_pod_affinity_terms,
required_during_scheduling_ignored_during_execution=
pod_affinity_terms,
)
affinity = None
if (node_affinity or pod_affinity or pod_anti_affinity):
affinity = V1Affinity(
node_affinity=node_affinity,
pod_affinity=pod_affinity,
pod_anti_affinity=pod_anti_affinity,
)
if affinity:
pod.spec.affinity = affinity
if priority_class_name:
pod.spec.priority_class_name = priority_class_name
if extra_pod_config:
pod.spec = update_k8s_model(
target=pod.spec,
changes=extra_pod_config,
logger=logger,
target_name="pod.spec",
changes_name="extra_pod_config",
)
return pod
def make_pod(
name,
cmd,
port,
image_spec,
image_pull_policy,
image_pull_secret=None,
node_selector=None,
run_as_uid=None,
run_as_gid=None,
fs_gid=None,
supplemental_gids=None,
run_privileged=False,
env=None,
working_dir=None,
volumes=None,
volume_mounts=None,
labels=None,
annotations=None,
cpu_limit=None,
cpu_guarantee=None,
mem_limit=None,
mem_guarantee=None,
extra_resource_limits=None,
extra_resource_guarantees=None,
lifecycle_hooks=None,
init_containers=None,
service_account=None,
extra_container_config=None,
extra_pod_config=None,
extra_containers=None,
scheduler_name=None,
tolerations=None,
node_affinity_preferred=None,
node_affinity_required=None,
pod_affinity_preferred=None,
pod_affinity_required=None,
pod_anti_affinity_preferred=None,
pod_anti_affinity_required=None,
priority_class_name=None,
logger=None,
userdir={},
):
"""
Make a k8s pod specification for running a user notebook.
Parameters
----------
name:
Name of pod. Must be unique within the namespace the object is
going to be created in. Must be a valid DNS label.
image_spec:
Image specification - usually a image name and tag in the form
of image_name:tag. Same thing you would use with docker commandline
arguments
image_pull_policy:
Image pull policy - one of 'Always', 'IfNotPresent' or 'Never'. Decides
when kubernetes will check for a newer version of image and pull it when
running a pod.
image_pull_secret:
Image pull secret - Default is None -- set to your secret name to pull
from private docker registry.
port:
Port the notebook server is going to be listening on
cmd:
The command used to execute the singleuser server.
node_selector:
Dictionary Selector to match nodes where to launch the Pods
run_as_uid:
The UID used to run single-user pods. The default is to run as the user
specified in the Dockerfile, if this is set to None.
run_as_gid:
The GID used to run single-user pods. The default is to run as the primary
group of the user specified in the Dockerfile, if this is set to None.
fs_gid
The gid that will own any fresh volumes mounted into this pod, if using
volume types that support this (such as GCE). This should be a group that
the uid the process is running as should be a member of, so that it can
read / write to the volumes mounted.
supplemental_gids:
A list of GIDs that should be set as additional supplemental groups to
the user that the container runs as. You may have to set this if you are
deploying to an environment with RBAC/SCC enforced and pods run with a
'restricted' SCC which results in the image being run as an assigned
user ID. The supplemental group IDs would need to include the
corresponding group ID of the user ID the image normally would run as.
The image must setup all directories/files any application needs access
to, as group writable.
run_privileged:
Whether the container should be run in privileged mode.
env:
Dictionary of environment variables.
volumes:
List of dictionaries containing the volumes of various types this pod
will be using. See k8s documentation about volumes on how to specify
these
volume_mounts:
List of dictionaries mapping paths in the container and the volume(
specified in volumes) that should be mounted on them. See the k8s
documentaiton for more details
working_dir:
String specifying the working directory for the notebook container
labels:
Labels to add to the spawned pod.
annotations:
Annotations to add to the spawned pod.
cpu_limit:
Float specifying the max number of CPU cores the user's pod is
allowed to use.
cpu_guarentee:
Float specifying the max number of CPU cores the user's pod is
guaranteed to have access to, by the scheduler.
mem_limit:
String specifying the max amount of RAM the user's pod is allowed
to use. String instead of float/int since common suffixes are allowed
mem_guarantee:
String specifying the max amount of RAM the user's pod is guaranteed
to have access to. String ins loat/int since common suffixes
are allowed
lifecycle_hooks:
Dictionary of lifecycle hooks
init_containers:
List of initialization containers belonging to the pod.
service_account:
Service account to mount on the pod. None disables mounting
extra_container_config:
Extra configuration (e.g. envFrom) for notebook container which is not covered by parameters above.
extra_pod_config:
Extra configuration (e.g. tolerations) for pod which is not covered by parameters above.
extra_containers:
Extra containers besides notebook container. Used for some housekeeping jobs (e.g. crontab).
scheduler_name:
The pod's scheduler explicitly named.
tolerations:
Tolerations can allow a pod to schedule or execute on a tainted node. To
learn more about pod tolerations, see
https://kubernetes.io/docs/concepts/configuration/taint-and-toleration/.
Pass this field an array of "Toleration" objects.*
* https://kubernetes.io/docs/reference/generated/kubernetes-api/v1.10/#nodeselectorterm-v1-core
node_affinity_preferred:
Affinities describe where pods prefer or require to be scheduled, they
may prefer or require a node to have a certain label or be in proximity
/ remoteness to another pod. To learn more visit
https://kubernetes.io/docs/concepts/configuration/assign-pod-node/
Pass this field an array of "PreferredSchedulingTerm" objects.*
* https://kubernetes.io/docs/reference/generated/kubernetes-api/v1.10/#preferredschedulingterm-v1-core
node_affinity_required:
Affinities describe where pods prefer or require to be scheduled, they
may prefer or require a node to have a certain label or be in proximity
/ remoteness to another pod. To learn more visit
https://kubernetes.io/docs/concepts/configuration/assign-pod-node/
Pass this field an array of "NodeSelectorTerm" objects.*
* https://kubernetes.io/docs/reference/generated/kubernetes-api/v1.10/#nodeselectorterm-v1-core
pod_affinity_preferred:
Affinities describe where pods prefer or require to be scheduled, they
may prefer or require a node to have a certain label or be in proximity
/ remoteness to another pod. To learn more visit
https://kubernetes.io/docs/concepts/configuration/assign-pod-node/
Pass this field an array of "WeightedPodAffinityTerm" objects.*
* https://kubernetes.io/docs/reference/generated/kubernetes-api/v1.10/#weightedpodaffinityterm-v1-core
pod_affinity_required:
Affinities describe where pods prefer or require to be scheduled, they
may prefer or require a node to have a certain label or be in proximity
/ remoteness to another pod. To learn more visit
https://kubernetes.io/docs/concepts/configuration/assign-pod-node/
Pass this field an array of "PodAffinityTerm" objects.*
* https://kubernetes.io/docs/reference/generated/kubernetes-api/v1.10/#podaffinityterm-v1-core
pod_anti_affinity_preferred:
Affinities describe where pods prefer or require to be scheduled, they
may prefer or require a node to have a certain label or be in proximity
/ remoteness to another pod. To learn more visit
https://kubernetes.io/docs/concepts/configuration/assign-pod-node/
Pass this field an array of "WeightedPodAffinityTerm" objects.*
* https://kubernetes.io/docs/reference/generated/kubernetes-api/v1.10/#weightedpodaffinityterm-v1-core
pod_anti_affinity_required:
Affinities describe where pods prefer or require to be scheduled, they
may prefer or require a node to have a certain label or be in proximity
/ remoteness to another pod. To learn more visit
https://kubernetes.io/docs/concepts/configuration/assign-pod-node/
Pass this field an array of "PodAffinityTerm" objects.*
* https://kubernetes.io/docs/reference/generated/kubernetes-api/v1.10/#podaffinityterm-v1-core
priority_class_name:
The name of the PriorityClass to be assigned the pod. This feature is Beta available in K8s 1.11.
"""
pod = V1Pod()
pod.kind = "Pod"
pod.api_version = "v1"
pod.metadata = V1ObjectMeta(name=name,
labels=(labels or {}).copy(),
annotations=(annotations or {}).copy())
pod.spec = V1PodSpec(containers=[])
pod.spec.restartPolicy = 'Never'
security_context = V1PodSecurityContext()
if fs_gid is not None:
security_context.fs_group = int(fs_gid)
if supplemental_gids is not None and supplemental_gids:
security_context.supplemental_groups = [
int(gid) for gid in supplemental_gids
]
if run_as_uid is not None:
security_context.run_as_user = int(run_as_uid)
if run_as_gid is not None:
security_context.run_as_group = int(run_as_gid)
pod.spec.security_context = security_context
if image_pull_secret is not None:
pod.spec.image_pull_secrets = []
image_secret = V1LocalObjectReference()
image_secret.name = image_pull_secret
pod.spec.image_pull_secrets.append(image_secret)
if node_selector:
pod.spec.node_selector = node_selector
if lifecycle_hooks:
lifecycle_hooks = get_k8s_model(V1Lifecycle, lifecycle_hooks)
notebook_container = V1Container(
name='notebook',
image=image_spec,
working_dir=working_dir,
ports=[V1ContainerPort(name='notebook-port', container_port=port)],
env=[V1EnvVar(k, v) for k, v in (env or {}).items()],
args=cmd,
image_pull_policy=image_pull_policy,
lifecycle=lifecycle_hooks,
resources=V1ResourceRequirements(),
volume_mounts=[
get_k8s_model(V1VolumeMount, obj) for obj in (volume_mounts or [])
],
)
if service_account is None:
# This makes sure that we don't accidentally give access to the whole
# kubernetes API to the users in the spawned pods.
pod.spec.automount_service_account_token = False
else:
pod.spec.service_account_name = service_account
if run_privileged:
notebook_container.security_context = V1SecurityContext(
privileged=True)
notebook_container.resources.requests = {}
if userdir.get('jp_cpu_request'):
notebook_container.resources.requests['cpu'] = userdir.get(
'jp_cpu_request')
elif cpu_guarantee:
notebook_container.resources.requests['cpu'] = cpu_guarantee
if userdir.get('jp_mem_request'):
notebook_container.resources.requests['memory'] = userdir.get(
'jp_mem_request')
elif mem_guarantee:
notebook_container.resources.requests['memory'] = mem_guarantee
if extra_resource_guarantees:
notebook_container.resources.requests.update(extra_resource_guarantees)
notebook_container.resources.limits = {}
if userdir.get('jp_cpu_limit'):
notebook_container.resources.limits['cpu'] = userdir.get(
'jp_cpu_limit')
elif cpu_limit:
notebook_container.resources.limits['cpu'] = cpu_limit
if userdir.get('jp_mem_limit'):
notebook_container.resources.limits['memory'] = userdir.get(
'jp_mem_limit')
elif mem_limit:
notebook_container.resources.limits['memory'] = mem_limit
if extra_resource_limits:
notebook_container.resources.limits.update(extra_resource_limits)
if userdir.get("jp_gpu_enable", False):
notebook_container.resources.limits[r'nvidia.com/gpu'] = userdir.get(
'jp_gpu_number', 0)
if extra_container_config:
notebook_container = update_k8s_model(
target=notebook_container,
changes=extra_container_config,
logger=logger,
target_name="notebook_container",
changes_name="extra_container_config",
)
pod.spec.containers.append(notebook_container)
if extra_containers:
pod.spec.containers.extend(
[get_k8s_model(V1Container, obj) for obj in extra_containers])
if tolerations:
pod.spec.tolerations = [
get_k8s_model(V1Toleration, obj) for obj in tolerations
]
if init_containers:
pod.spec.init_containers = [
get_k8s_model(V1Container, obj) for obj in init_containers
]
if volumes:
pod.spec.volumes = [get_k8s_model(V1Volume, obj) for obj in volumes]
else:
# Keep behaving exactly like before by not cleaning up generated pod
# spec by setting the volumes field even though it is an empty list.
pod.spec.volumes = []
if scheduler_name:
pod.spec.scheduler_name = scheduler_name
node_affinity = None
if node_affinity_preferred or node_affinity_required:
node_selector = None
if node_affinity_required:
node_selector = V1NodeSelector(node_selector_terms=[
get_k8s_model(V1NodeSelectorTerm, obj)
for obj in node_affinity_required
], )
preferred_scheduling_terms = None
if node_affinity_preferred:
preferred_scheduling_terms = [
get_k8s_model(V1PreferredSchedulingTerm, obj)
for obj in node_affinity_preferred
]
node_affinity = V1NodeAffinity(
preferred_during_scheduling_ignored_during_execution=
preferred_scheduling_terms,
required_during_scheduling_ignored_during_execution=node_selector,
)
pod_affinity = None
if pod_affinity_preferred or pod_affinity_required:
weighted_pod_affinity_terms = None
if pod_affinity_preferred:
weighted_pod_affinity_terms = [
get_k8s_model(V1WeightedPodAffinityTerm, obj)
for obj in pod_affinity_preferred
]
pod_affinity_terms = None
if pod_affinity_required:
pod_affinity_terms = [
get_k8s_model(V1PodAffinityTerm, obj)
for obj in pod_affinity_required
]
pod_affinity = V1PodAffinity(
preferred_during_scheduling_ignored_during_execution=
weighted_pod_affinity_terms,
required_during_scheduling_ignored_during_execution=
pod_affinity_terms,
)
pod_anti_affinity = None
if pod_anti_affinity_preferred or pod_anti_affinity_required:
weighted_pod_affinity_terms = None
if pod_anti_affinity_preferred:
weighted_pod_affinity_terms = [
get_k8s_model(V1WeightedPodAffinityTerm, obj)
for obj in pod_anti_affinity_preferred
]
pod_affinity_terms = None
if pod_anti_affinity_required:
pod_affinity_terms = [
get_k8s_model(V1PodAffinityTerm, obj)
for obj in pod_anti_affinity_required
]
pod_anti_affinity = V1PodAffinity(
preferred_during_scheduling_ignored_during_execution=
weighted_pod_affinity_terms,
required_during_scheduling_ignored_during_execution=
pod_affinity_terms,
)
affinity = None
if (node_affinity or pod_affinity or pod_anti_affinity):
affinity = V1Affinity(
node_affinity=node_affinity,
pod_affinity=pod_affinity,
pod_anti_affinity=pod_anti_affinity,
)
if affinity:
pod.spec.affinity = affinity
if priority_class_name:
pod.spec.priority_class_name = priority_class_name
if extra_pod_config:
pod.spec = update_k8s_model(
target=pod.spec,
changes=extra_pod_config,
logger=logger,
target_name="pod.spec",
changes_name="extra_pod_config",
)
return pod
def make_pod(
name,
cmd,
port,
image,
image_pull_policy,
image_pull_secrets=None,
node_selector=None,
uid=None,
gid=None,
fs_gid=None,
supplemental_gids=None,
privileged=False,
allow_privilege_escalation=True,
container_security_context=None,
pod_security_context=None,
env=None,
working_dir=None,
volumes=None,
volume_mounts=None,
labels=None,
annotations=None,
cpu_limit=None,
cpu_guarantee=None,
mem_limit=None,
mem_guarantee=None,
extra_resource_limits=None,
extra_resource_guarantees=None,
lifecycle_hooks=None,
init_containers=None,
service_account=None,
automount_service_account_token=None,
extra_container_config=None,
extra_pod_config=None,
extra_containers=None,
scheduler_name=None,
tolerations=None,
node_affinity_preferred=None,
node_affinity_required=None,
pod_affinity_preferred=None,
pod_affinity_required=None,
pod_anti_affinity_preferred=None,
pod_anti_affinity_required=None,
priority_class_name=None,
ssl_secret_name=None,
ssl_secret_mount_path=None,
logger=None,
):
"""
Make a k8s pod specification for running a user notebook.
Parameters
----------
name:
Name of pod. Must be unique within the namespace the object is
going to be created in. Must be a valid DNS label.
image:
Image specification - usually a image name and tag in the form
of image_name:tag. Same thing you would use with docker commandline
arguments
image_pull_policy:
Image pull policy - one of 'Always', 'IfNotPresent' or 'Never'. Decides
when kubernetes will check for a newer version of image and pull it when
running a pod.
image_pull_secrets:
Image pull secrets - a list of references to Kubernetes Secret resources
with credentials to pull images from image registries. This list can
either have strings in it or objects with the string value nested under
a name field.
port:
Port the notebook server is going to be listening on
cmd:
The command used to execute the singleuser server.
node_selector:
Dictionary Selector to match nodes where to launch the Pods
uid:
The UID used to run single-user pods. The default is to run as the user
specified in the Dockerfile, if this is set to None.
gid:
The GID used to run single-user pods. The default is to run as the primary
group of the user specified in the Dockerfile, if this is set to None.
Setting this parameter requires that *feature-gate* **RunAsGroup** be enabled,
otherwise the effective GID of the pod will be 0 (root). In addition, not
setting `gid` once feature-gate RunAsGroup is enabled will also
result in an effective GID of 0 (root).
fs_gid
The gid that will own any fresh volumes mounted into this pod, if using
volume types that support this (such as GCE). This should be a group that
the uid the process is running as should be a member of, so that it can
read / write to the volumes mounted.
supplemental_gids:
A list of GIDs that should be set as additional supplemental groups to
the user that the container runs as. You may have to set this if you are
deploying to an environment with RBAC/SCC enforced and pods run with a
'restricted' SCC which results in the image being run as an assigned
user ID. The supplemental group IDs would need to include the
corresponding group ID of the user ID the image normally would run as.
The image must setup all directories/files any application needs access
to, as group writable.
privileged:
Whether the container should be run in privileged mode.
allow_privilege_escalation:
Controls whether a process can gain more privileges than its parent process.
container_security_context:
A kubernetes securityContext to apply to the container.
pod_security_context:
A kubernetes securityContext to apply to the pod.
env:
Dictionary of environment variables.
volumes:
List of dictionaries containing the volumes of various types this pod
will be using. See k8s documentation about volumes on how to specify
these
volume_mounts:
List of dictionaries mapping paths in the container and the volume(
specified in volumes) that should be mounted on them. See the k8s
documentaiton for more details
working_dir:
String specifying the working directory for the notebook container
labels:
Labels to add to the spawned pod.
annotations:
Annotations to add to the spawned pod.
cpu_limit:
Float specifying the max number of CPU cores the user's pod is
allowed to use.
cpu_guarantee:
Float specifying the max number of CPU cores the user's pod is
guaranteed to have access to, by the scheduler.
mem_limit:
String specifying the max amount of RAM the user's pod is allowed
to use. String instead of float/int since common suffixes are allowed
mem_guarantee:
String specifying the max amount of RAM the user's pod is guaranteed
to have access to. String ins loat/int since common suffixes
are allowed
lifecycle_hooks:
Dictionary of lifecycle hooks
init_containers:
List of initialization containers belonging to the pod.
service_account:
Service account to mount on the pod. None disables mounting
extra_container_config:
Extra configuration (e.g. envFrom) for notebook container which is not covered by parameters above.
extra_pod_config:
Extra configuration (e.g. tolerations) for pod which is not covered by parameters above.
extra_containers:
Extra containers besides notebook container. Used for some housekeeping jobs (e.g. crontab).
scheduler_name:
The pod's scheduler explicitly named.
tolerations:
Tolerations can allow a pod to schedule or execute on a tainted node. To
learn more about pod tolerations, see
https://kubernetes.io/docs/concepts/configuration/taint-and-toleration/.
Pass this field an array of "Toleration" objects.
* https://kubernetes.io/docs/reference/generated/kubernetes-api/v1.20/#toleration-v1-core
node_affinity_preferred:
Affinities describe where pods prefer or require to be scheduled, they
may prefer or require a node to have a certain label or be in proximity
/ remoteness to another pod. To learn more visit
https://kubernetes.io/docs/concepts/configuration/assign-pod-node/
Pass this field an array of "PreferredSchedulingTerm" objects.
* https://kubernetes.io/docs/reference/generated/kubernetes-api/v1.20#preferredschedulingterm-v1-core
node_affinity_required:
Affinities describe where pods prefer or require to be scheduled, they
may prefer or require a node to have a certain label or be in proximity
/ remoteness to another pod. To learn more visit
https://kubernetes.io/docs/concepts/configuration/assign-pod-node/
Pass this field an array of "NodeSelectorTerm" objects.
* https://kubernetes.io/docs/reference/generated/kubernetes-api/v1.20/#nodeselectorterm-v1-core
pod_affinity_preferred:
Affinities describe where pods prefer or require to be scheduled, they
may prefer or require a node to have a certain label or be in proximity
/ remoteness to another pod. To learn more visit
https://kubernetes.io/docs/concepts/configuration/assign-pod-node/
Pass this field an array of "WeightedPodAffinityTerm" objects.
* https://kubernetes.io/docs/reference/generated/kubernetes-api/v1.20/#weightedpodaffinityterm-v1-core
pod_affinity_required:
Affinities describe where pods prefer or require to be scheduled, they
may prefer or require a node to have a certain label or be in proximity
/ remoteness to another pod. To learn more visit
https://kubernetes.io/docs/concepts/configuration/assign-pod-node/
Pass this field an array of "PodAffinityTerm" objects.
* https://kubernetes.io/docs/reference/generated/kubernetes-api/v1.20/#podaffinityterm-v1-core
pod_anti_affinity_preferred:
Affinities describe where pods prefer or require to be scheduled, they
may prefer or require a node to have a certain label or be in proximity
/ remoteness to another pod. To learn more visit
https://kubernetes.io/docs/concepts/configuration/assign-pod-node/
Pass this field an array of "WeightedPodAffinityTerm" objects.
* https://kubernetes.io/docs/reference/generated/kubernetes-api/v1.20/#weightedpodaffinityterm-v1-core
pod_anti_affinity_required:
Affinities describe where pods prefer or require to be scheduled, they
may prefer or require a node to have a certain label or be in proximity
/ remoteness to another pod. To learn more visit
https://kubernetes.io/docs/concepts/configuration/assign-pod-node/
Pass this field an array of "PodAffinityTerm" objects.
* https://kubernetes.io/docs/reference/generated/kubernetes-api/v1.20/#podaffinityterm-v1-core
priority_class_name:
The name of the PriorityClass to be assigned the pod. This feature is Beta available in K8s 1.11 and GA in 1.14.
ssl_secret_name:
Specifies the name of the ssl secret
ssl_secret_mount_path:
Specifies the name of the ssl secret mount path for the pod
"""
pod = V1Pod()
pod.kind = "Pod"
pod.api_version = "v1"
pod.metadata = V1ObjectMeta(
name=name,
labels=(labels or {}).copy(),
annotations=(annotations or {}).copy(),
)
pod.spec = V1PodSpec(containers=[])
pod.spec.restart_policy = 'OnFailure'
if image_pull_secrets is not None:
# image_pull_secrets as received by the make_pod function should always
# be a list, but it is allowed to have "a-string" elements or {"name":
# "a-string"} elements.
pod.spec.image_pull_secrets = [
V1LocalObjectReference(name=secret_ref)
if type(secret_ref) == str
else get_k8s_model(V1LocalObjectReference, secret_ref)
for secret_ref in image_pull_secrets
]
if ssl_secret_name and ssl_secret_mount_path:
if not volumes:
volumes = []
volumes.append(
{
'name': 'jupyterhub-internal-certs',
'secret': {'secretName': ssl_secret_name, 'defaultMode': 511},
}
)
env['JUPYTERHUB_SSL_KEYFILE'] = ssl_secret_mount_path + "ssl.key"
env['JUPYTERHUB_SSL_CERTFILE'] = ssl_secret_mount_path + "ssl.crt"
env['JUPYTERHUB_SSL_CLIENT_CA'] = (
ssl_secret_mount_path + "notebooks-ca_trust.crt"
)
if not volume_mounts:
volume_mounts = []
volume_mounts.append(
{
'name': 'jupyterhub-internal-certs',
'mountPath': ssl_secret_mount_path,
}
)
if node_selector:
pod.spec.node_selector = node_selector
if lifecycle_hooks:
lifecycle_hooks = get_k8s_model(V1Lifecycle, lifecycle_hooks)
# Security contexts can be configured on Pod and Container level. The
# Dedicated KubeSpawner API will bootstraps the container_security_context
# except for if can only be configured on the Pod level, then it bootstraps
# pod_security_context.
#
# The pod|container_security_context configuration is given a higher
# priority than the dedicated KubeSpawner API options.
#
# Note that validation against the Python kubernetes-client isn't made as
# the security contexts has evolved significantly and kubernetes-client is
# too outdated.
#
# | Dedicated KubeSpawner API | Kubernetes API | Security contexts |
# | -------------------------- | ------------------------ | ----------------- |
# | supplemental_gids | supplementalGroups | Pod only |
# | fs_gid | fsGroup | Pod only |
# | - | fsGroupChangePolicy | Pod only |
# | - | sysctls | Pod only |
# | privileged | privileged | Container only |
# | allow_privilege_escalation | allowPrivilegeEscalation | Container only |
# | - | capabilities | Container only |
# | - | procMount | Container only |
# | - | readOnlyRootFilesystem | Container only |
# | uid | runAsUser | Pod and Container |
# | gid | runAsGroup | Pod and Container |
# | - | runAsNonRoot | Pod and Container |
# | - | seLinuxOptions | Pod and Container |
# | - | seccompProfile | Pod and Container |
# | - | windowsOptions | Pod and Container |
#
# ref: https://kubernetes.io/docs/tasks/configure-pod-container/security-context/
# ref: https://kubernetes.io/docs/reference/generated/kubernetes-api/v1.20/#securitycontext-v1-core (container)
# ref: https://kubernetes.io/docs/reference/generated/kubernetes-api/v1.20/#podsecuritycontext-v1-core (pod)
#
psc = {}
# populate with fs_gid / supplemental_gids
if fs_gid is not None:
psc["fsGroup"] = int(fs_gid)
if supplemental_gids:
psc["supplementalGroups"] = [int(gid) for gid in supplemental_gids]
if pod_security_context:
for key in pod_security_context.keys():
if "_" in key:
raise ValueError(
f"pod_security_context's keys should have k8s camelCase names, got '{key}'"
)
psc.update(pod_security_context)
if not psc:
psc = None
pod.spec.security_context = psc
csc = {}
# populate with uid / gid / privileged / allow_privilege_escalation
if uid is not None:
csc["runAsUser"] = int(uid)
if gid is not None:
csc["runAsGroup"] = int(gid)
if privileged: # false as default
csc["privileged"] = True
if not allow_privilege_escalation: # true as default
csc["allowPrivilegeEscalation"] = False
if container_security_context:
for key in container_security_context.keys():
if "_" in key:
raise ValueError(
f"container_security_context's keys should have k8s camelCase names, got '{key}'"
)
csc.update(container_security_context)
if not csc:
csc = None
def _get_env_var_deps(env):
# only consider env var objects with an explicit string value
if not env.value:
return set()
# $(MY_ENV) pattern: $( followed by non-)-characters to be captured, followed by )
re_k8s_env_reference_pattern = r"\$\(([^\)]+)\)"
deps = set(re.findall(re_k8s_env_reference_pattern, env.value))
return deps - {env.name}
unsorted_env = {}
for key, env in (env or {}).items():
# Normalize KubeSpawners env input to valid Kubernetes EnvVar Python
# representations. They should have a "name" field as well as either a
# "value" field or "value_from" field. For examples see the
# test_make_pod_with_env function.
if type(env) == dict:
if not "name" in env:
env["name"] = key
env = get_k8s_model(V1EnvVar, env)
else:
env = V1EnvVar(name=key, value=env)
# Extract information about references to other envs as we want to use
# those to make an intelligent sorting before we render this into a list
# with an order that matters.
unsorted_env[env.name] = {
"deps": _get_env_var_deps(env),
"key": key,
"env": env,
}
# We sort environment variables in a way that allows dependencies to other
# env to resolve as much as possible. There could be circular dependencies
# so we will just do our best and settle with that.
#
# Algorithm description:
#
# - loop step:
# - pop all unsorted_env entries with dependencies in sorted_env
# - sort popped env based on key and extend the sorted_env list
# - loop exit:
# - exit if loop step didn't pop anything from unsorted_env
# - before exit, sort what remains and extending the sorted_env list
#
sorted_env = []
while True:
already_resolved_env_names = [e.name for e in sorted_env]
extracted_env = {}
for k, v in unsorted_env.copy().items():
if v["deps"].issubset(already_resolved_env_names):
extracted_env[k] = unsorted_env.pop(k)
if extracted_env:
extracted_env = [
d["env"]
for d in sorted(extracted_env.values(), key=operator.itemgetter("key"))
]
sorted_env.extend(extracted_env)
else:
remaining_env = [
d["env"]
for d in sorted(unsorted_env.values(), key=operator.itemgetter("key"))
]
sorted_env.extend(remaining_env)
break
notebook_container = V1Container(
name='notebook',
image=image,
working_dir=working_dir,
ports=[V1ContainerPort(name='notebook-port', container_port=port)],
env=sorted_env,
args=cmd,
image_pull_policy=image_pull_policy,
lifecycle=lifecycle_hooks,
resources=V1ResourceRequirements(),
volume_mounts=[
get_k8s_model(V1VolumeMount, obj) for obj in (volume_mounts or [])
],
security_context=csc,
)
if service_account is not None:
pod.spec.service_account_name = service_account
if automount_service_account_token is None:
if service_account is None:
# This makes sure that we don't accidentally give access to the whole
# kubernetes API to the users in the spawned pods.
pod.spec.automount_service_account_token = False
else:
pod.spec.automount_service_account_token = automount_service_account_token
notebook_container.resources.requests = {}
if cpu_guarantee:
notebook_container.resources.requests['cpu'] = cpu_guarantee
if mem_guarantee:
notebook_container.resources.requests['memory'] = mem_guarantee
if extra_resource_guarantees:
notebook_container.resources.requests.update(extra_resource_guarantees)
notebook_container.resources.limits = {}
if cpu_limit:
notebook_container.resources.limits['cpu'] = cpu_limit
if mem_limit:
notebook_container.resources.limits['memory'] = mem_limit
if extra_resource_limits:
notebook_container.resources.limits.update(extra_resource_limits)
if extra_container_config:
notebook_container = update_k8s_model(
target=notebook_container,
changes=extra_container_config,
logger=logger,
target_name="notebook_container",
changes_name="extra_container_config",
)
pod.spec.containers.append(notebook_container)
if extra_containers:
pod.spec.containers.extend(
[get_k8s_model(V1Container, obj) for obj in extra_containers]
)
if tolerations:
pod.spec.tolerations = [get_k8s_model(V1Toleration, obj) for obj in tolerations]
if init_containers:
pod.spec.init_containers = [
get_k8s_model(V1Container, obj) for obj in init_containers
]
if volumes:
pod.spec.volumes = [get_k8s_model(V1Volume, obj) for obj in volumes]
else:
# Keep behaving exactly like before by not cleaning up generated pod
# spec by setting the volumes field even though it is an empty list.
pod.spec.volumes = []
if scheduler_name:
pod.spec.scheduler_name = scheduler_name
node_affinity = None
if node_affinity_preferred or node_affinity_required:
node_selector = None
if node_affinity_required:
node_selector = V1NodeSelector(
node_selector_terms=[
get_k8s_model(V1NodeSelectorTerm, obj)
for obj in node_affinity_required
],
)
preferred_scheduling_terms = None
if node_affinity_preferred:
preferred_scheduling_terms = [
get_k8s_model(V1PreferredSchedulingTerm, obj)
for obj in node_affinity_preferred
]
node_affinity = V1NodeAffinity(
preferred_during_scheduling_ignored_during_execution=preferred_scheduling_terms,
required_during_scheduling_ignored_during_execution=node_selector,
)
pod_affinity = None
if pod_affinity_preferred or pod_affinity_required:
weighted_pod_affinity_terms = None
if pod_affinity_preferred:
weighted_pod_affinity_terms = [
get_k8s_model(V1WeightedPodAffinityTerm, obj)
for obj in pod_affinity_preferred
]
pod_affinity_terms = None
if pod_affinity_required:
pod_affinity_terms = [
get_k8s_model(V1PodAffinityTerm, obj) for obj in pod_affinity_required
]
pod_affinity = V1PodAffinity(
preferred_during_scheduling_ignored_during_execution=weighted_pod_affinity_terms,
required_during_scheduling_ignored_during_execution=pod_affinity_terms,
)
pod_anti_affinity = None
if pod_anti_affinity_preferred or pod_anti_affinity_required:
weighted_pod_affinity_terms = None
if pod_anti_affinity_preferred:
weighted_pod_affinity_terms = [
get_k8s_model(V1WeightedPodAffinityTerm, obj)
for obj in pod_anti_affinity_preferred
]
pod_affinity_terms = None
if pod_anti_affinity_required:
pod_affinity_terms = [
get_k8s_model(V1PodAffinityTerm, obj)
for obj in pod_anti_affinity_required
]
pod_anti_affinity = V1PodAffinity(
preferred_during_scheduling_ignored_during_execution=weighted_pod_affinity_terms,
required_during_scheduling_ignored_during_execution=pod_affinity_terms,
)
affinity = None
if node_affinity or pod_affinity or pod_anti_affinity:
affinity = V1Affinity(
node_affinity=node_affinity,
pod_affinity=pod_affinity,
pod_anti_affinity=pod_anti_affinity,
)
if affinity:
pod.spec.affinity = affinity
if priority_class_name:
pod.spec.priority_class_name = priority_class_name
if extra_pod_config:
pod.spec = update_k8s_model(
target=pod.spec,
changes=extra_pod_config,
logger=logger,
target_name="pod.spec",
changes_name="extra_pod_config",
)
return pod
def make_pod_spec(self, tls_secret, worker_name=None):
annotations = self.common_annotations
env = self.get_env()
if worker_name is not None:
# Worker
name = "dask-gateway-worker-%s" % worker_name
container_name = "dask-gateway-worker"
labels = self.get_labels_for("dask-gateway-worker",
worker_name=worker_name)
mem_req = self.worker_memory
mem_lim = self.worker_memory_limit
cpu_req = self.worker_cores
cpu_lim = self.worker_cores_limit
env["DASK_GATEWAY_WORKER_NAME"] = worker_name
cmd = self.worker_command
extra_pod_config = self.worker_extra_pod_config
extra_container_config = self.worker_extra_container_config
else:
# Scheduler
name = "dask-gateway-scheduler-%s" % self.cluster_name
container_name = "dask-gateway-scheduler"
labels = self.get_labels_for("dask-gateway-scheduler")
mem_req = self.scheduler_memory
mem_lim = self.scheduler_memory_limit
cpu_req = self.scheduler_cores
cpu_lim = self.scheduler_cores_limit
cmd = self.scheduler_command
extra_pod_config = self.scheduler_extra_pod_config
extra_container_config = self.scheduler_extra_container_config
volume = V1Volume(name="dask-credentials",
secret=V1SecretVolumeSource(secret_name=tls_secret))
container = V1Container(
name=container_name,
image=self.image,
args=cmd,
env=[V1EnvVar(k, v) for k, v in env.items()],
image_pull_policy=self.image_pull_policy,
resources=V1ResourceRequirements(
requests={
"cpu": cpu_req,
"memory": mem_req
},
limits={
"cpu": cpu_lim,
"memory": mem_lim
},
),
volume_mounts=[
V1VolumeMount(
name=volume.name,
mount_path="/etc/dask-credentials/",
read_only=True,
)
],
)
if extra_container_config:
container = merge_kube_objects(container, extra_container_config)
pod = V1Pod(
kind="Pod",
api_version="v1",
metadata=V1ObjectMeta(name=name,
labels=labels,
annotations=annotations),
spec=V1PodSpec(containers=[container],
volumes=[volume],
restart_policy="Never"),
)
# Ensure we don't accidentally give access to the kubernetes API
pod.spec.automount_service_account_token = False
if extra_pod_config:
pod.spec = merge_kube_objects(pod.spec, extra_pod_config)
return pod
def make_pod(name,
cmd,
port,
image_spec,
image_pull_policy,
image_pull_secret=None,
node_selector=None,
run_as_uid=None,
run_as_gid=None,
fs_gid=None,
supplemental_gids=None,
run_privileged=False,
env={},
working_dir=None,
volumes=[],
volume_mounts=[],
labels={},
annotations={},
cpu_limit=None,
cpu_guarantee=None,
mem_limit=None,
mem_guarantee=None,
extra_resource_limits=None,
extra_resource_guarantees=None,
lifecycle_hooks=None,
init_containers=None,
service_account=None,
extra_container_config=None,
extra_pod_config=None,
extra_containers=None,
scheduler_name=None):
"""
Make a k8s pod specification for running a user notebook.
Parameters
----------
name:
Name of pod. Must be unique within the namespace the object is
going to be created in. Must be a valid DNS label.
image_spec:
Image specification - usually a image name and tag in the form
of image_name:tag. Same thing you would use with docker commandline
arguments
image_pull_policy:
Image pull policy - one of 'Always', 'IfNotPresent' or 'Never'. Decides
when kubernetes will check for a newer version of image and pull it when
running a pod.
image_pull_secret:
Image pull secret - Default is None -- set to your secret name to pull
from private docker registry.
port:
Port the notebook server is going to be listening on
cmd:
The command used to execute the singleuser server.
node_selector:
Dictionary Selector to match nodes where to launch the Pods
run_as_uid:
The UID used to run single-user pods. The default is to run as the user
specified in the Dockerfile, if this is set to None.
run_as_gid:
The GID used to run single-user pods. The default is to run as the primary
group of the user specified in the Dockerfile, if this is set to None.
fs_gid
The gid that will own any fresh volumes mounted into this pod, if using
volume types that support this (such as GCE). This should be a group that
the uid the process is running as should be a member of, so that it can
read / write to the volumes mounted.
supplemental_gids:
A list of GIDs that should be set as additional supplemental groups to
the user that the container runs as. You may have to set this if you are
deploying to an environment with RBAC/SCC enforced and pods run with a
'restricted' SCC which results in the image being run as an assigned
user ID. The supplemental group IDs would need to include the
corresponding group ID of the user ID the image normally would run as.
The image must setup all directories/files any application needs access
to, as group writable.
run_privileged:
Whether the container should be run in privileged mode.
env:
Dictionary of environment variables.
volumes:
List of dictionaries containing the volumes of various types this pod
will be using. See k8s documentation about volumes on how to specify
these
volume_mounts:
List of dictionaries mapping paths in the container and the volume(
specified in volumes) that should be mounted on them. See the k8s
documentaiton for more details
working_dir:
String specifying the working directory for the notebook container
labels:
Labels to add to the spawned pod.
annotations:
Annotations to add to the spawned pod.
cpu_limit:
Float specifying the max number of CPU cores the user's pod is
allowed to use.
cpu_guarentee:
Float specifying the max number of CPU cores the user's pod is
guaranteed to have access to, by the scheduler.
mem_limit:
String specifying the max amount of RAM the user's pod is allowed
to use. String instead of float/int since common suffixes are allowed
mem_guarantee:
String specifying the max amount of RAM the user's pod is guaranteed
to have access to. String ins loat/int since common suffixes
are allowed
lifecycle_hooks:
Dictionary of lifecycle hooks
init_containers:
List of initialization containers belonging to the pod.
service_account:
Service account to mount on the pod. None disables mounting
extra_container_config:
Extra configuration (e.g. envFrom) for notebook container which is not covered by parameters above.
extra_pod_config:
Extra configuration (e.g. tolerations) for pod which is not covered by parameters above.
extra_containers:
Extra containers besides notebook container. Used for some housekeeping jobs (e.g. crontab).
scheduler_name:
A custom scheduler's name.
"""
pod = V1Pod()
pod.kind = "Pod"
pod.api_version = "v1"
pod.metadata = V1ObjectMeta(name=name,
labels=labels.copy(),
annotations=annotations.copy())
pod.spec = V1PodSpec(containers=[])
pod.spec.restartPolicy = 'Never'
security_context = V1PodSecurityContext()
if fs_gid is not None:
security_context.fs_group = int(fs_gid)
if supplemental_gids is not None and supplemental_gids:
security_context.supplemental_groups = [
int(gid) for gid in supplemental_gids
]
if run_as_uid is not None:
security_context.run_as_user = int(run_as_uid)
if run_as_gid is not None:
security_context.run_as_group = int(run_as_gid)
pod.spec.security_context = security_context
if image_pull_secret is not None:
pod.spec.image_pull_secrets = []
image_secret = V1LocalObjectReference()
image_secret.name = image_pull_secret
pod.spec.image_pull_secrets.append(image_secret)
if node_selector:
pod.spec.node_selector = node_selector
notebook_container = V1Container(
name='notebook',
image=image_spec,
working_dir=working_dir,
ports=[V1ContainerPort(name='notebook-port', container_port=port)],
env=[V1EnvVar(k, v) for k, v in env.items()],
args=cmd,
image_pull_policy=image_pull_policy,
lifecycle=lifecycle_hooks,
resources=V1ResourceRequirements())
if service_account is None:
# Add a hack to ensure that no service accounts are mounted in spawned pods
# This makes sure that we don"t accidentally give access to the whole
# kubernetes API to the users in the spawned pods.
# Note: We don't simply use `automountServiceAccountToken` here since we wanna be compatible
# with older kubernetes versions too for now.
hack_volume = V1Volume(name='no-api-access-please', empty_dir={})
hack_volumes = [hack_volume]
hack_volume_mount = V1VolumeMount(
name='no-api-access-please',
mount_path="/var/run/secrets/kubernetes.io/serviceaccount",
read_only=True)
hack_volume_mounts = [hack_volume_mount]
# Non-hacky way of not mounting service accounts
pod.spec.automount_service_account_token = False
else:
hack_volumes = []
hack_volume_mounts = []
pod.spec.service_account_name = service_account
if run_privileged:
notebook_container.security_context = V1SecurityContext(
privileged=True)
notebook_container.resources.requests = {}
if cpu_guarantee:
notebook_container.resources.requests['cpu'] = cpu_guarantee
if mem_guarantee:
notebook_container.resources.requests['memory'] = mem_guarantee
if extra_resource_guarantees:
for k in extra_resource_guarantees:
notebook_container.resources.requests[
k] = extra_resource_guarantees[k]
notebook_container.resources.limits = {}
if cpu_limit:
notebook_container.resources.limits['cpu'] = cpu_limit
if mem_limit:
notebook_container.resources.limits['memory'] = mem_limit
if extra_resource_limits:
for k in extra_resource_limits:
notebook_container.resources.limits[k] = extra_resource_limits[k]
notebook_container.volume_mounts = volume_mounts + hack_volume_mounts
pod.spec.containers.append(notebook_container)
if extra_container_config:
for key, value in extra_container_config.items():
setattr(notebook_container,
_map_attribute(notebook_container.attribute_map, key),
value)
if extra_pod_config:
for key, value in extra_pod_config.items():
setattr(pod.spec, _map_attribute(pod.spec.attribute_map, key),
value)
if extra_containers:
pod.spec.containers.extend(extra_containers)
pod.spec.init_containers = init_containers
pod.spec.volumes = volumes + hack_volumes
if scheduler_name:
pod.spec.scheduler_name = scheduler_name
return pod
@task(task_config=Pod(
pod_spec=V1PodSpec(
containers=[
V1Container(
name="primary",
resources=V1ResourceRequirements(
requests={
"cpu": ".5",
"memory": "500Mi"
},
limits={
"cpu": ".5",
"memory": "500Mi"
},
),
)
],
init_containers=[
V1Container(
image="alpine",
name="init",
command=["/bin/sh"],
args=["-c", 'echo "I\'m a customizable init container"'],
resources=V1ResourceRequirements(limits={
"cpu": ".5",
"memory": "500Mi"
}, ),
)
],
),
primary_container_name="primary",
def test_pod_resources(self):
cases = [
{
"pod":
V1Pod(spec=V1PodSpec(
init_containers=[
V1Container(name="1",
resources=V1ResourceRequirements(
limits={}, requests={}))
],
containers=[
V1Container(name="1",
resources=V1ResourceRequirements(
limits={}, requests={})),
V1Container(name="2",
resources=V1ResourceRequirements(
limits={}, requests={}))
])),
# req_cpu, req_memory, lim_cpu, lim_memory, gpu_spec
"expected": (0, 0, 0, 0, '')
},
{
"pod":
V1Pod(
spec=V1PodSpec(
init_containers=[
V1Container(name="1",
resources=V1ResourceRequirements(
limits={
"cpu": "1",
"memory": "512Mi"
},
requests={
"cpu": "0.5",
"memory": "256Mi"
}))
],
containers=[
V1Container(name="1",
resources=V1ResourceRequirements(
limits={
"cpu": "6",
"memory": "6Gi"
},
requests={
"cpu": "3",
"memory": "0.5Gi"
})),
V1Container(name="2",
resources=V1ResourceRequirements(
limits={
"cpu": "2",
"memory": "2Gi"
},
requests={
"cpu": "1.5",
"memory": "1Gi"
}))
])),
# req_cpu, req_memory, lim_cpu, lim_memory, gpu_spec
"expected": (4.5, 1.5, 8.0, 8.0, '')
}
]
for case in cases:
got = k8s_pod_resource_requirements(case['pod'])
self.assertEqual(got, case['expected'])
}
test_dep = V1Deployment(
'apps/v1',
'Deployment',
V1ObjectMeta(name='web-main', labels={'app': 'web'}, annotations={'opsani-current-mode': 'canary'}),
V1DeploymentSpec(
replicas=3,
selector=V1LabelSelector(match_labels={'app': 'web', 'role': 'main'}),
template=V1PodTemplateSpec(
metadata=V1ObjectMeta(labels={'app': 'web', 'role': 'main'}),
spec=V1PodSpec(containers=[V1Container(
name='main',
image='opsani/co-http:latest',
args=['busy=400'],
resources=V1ResourceRequirements(
limits={'cpu': '125m', 'memory': '512Mi'},
requests={'cpu': '125m', 'memory': '512Mi'}
),
ports=[V1ContainerPort(container_port=8080)]
)])
)
)
)
def setup_module(module):
if os.getenv('KUBERNETES_SERVICE_HOST'): # If running in a kubernetes cluster
kubernetes.config.load_incluster_config()
else:
kubernetes.config.load_kube_config()
# setup desired ns
def get_minimal_pod_task_config() -> Pod:
primary_container = V1Container(name="flytetask")
pod_spec = V1PodSpec(containers=[primary_container])
return Pod(pod_spec=pod_spec, primary_container_name="flytetask")