Basic Deployment Interaction

Import Functions

import os

from kubecustom.kubecustom import (
    create_secret_deployment,
    delete_secret_deployment,
    get_deployment_name
)
from kubecustom.deployment import (
    utilization_per_deployment,
    scale_deployment,
)
from kubecustom.pod import (
    get_pods_resource_info,
    print_pods_summary,
    get_active_tasks,
    delete_pods_by_status,
)
from kubecustom import MyData
MyDataInstance = MyData()
MyDataInstance.set_configuration()

Available configurations:
   - qca-psi4

Check Active Tasks in Pods

MyDataInstance.set_configuration(configuration="qca-psi4")
path = "/some/path/to/target/directory/394_pyddx-600"
tag, cpus, mem = "pyddx-600", 16, 32
deployment_name = get_deployment_name(tag)

#create_secret_deployment(path, tag, cpus, mem, replicas=1)
#delete_secret_deployment(deployment_name)
#scale_deployment(deployment_name, 10)
#get_pods_resource_info(verbose=True, keep_key=tag);
print_pods_summary(deployment_name=deployment_name)
#get_active_tasks(get_pod_list(deployment_name=deployment_name));
#delete_pods_by_status("ContainerCreating")

Pod Name,                           Restart Count,  Memory, CPUs,   |       Current State,  Current Status, |       Previous State, Previous Status
openff-jac-qca-psi4-pyddx-600-6cf5cf797c-462kx,     23,     0.1,    0.0     |       running,        Running,        |       terminated,     OOMKilled
openff-jac-qca-psi4-pyddx-600-6cf5cf797c-9bgsm,     0,      12.8,   12.2    |       running,        Running,        |       None,   None
openff-jac-qca-psi4-pyddx-600-6cf5cf797c-ddcmn,     4,      0.1,    0.0     |       running,        Running,        |       terminated,     OOMKilled
openff-jac-qca-psi4-pyddx-600-6cf5cf797c-htssf,     0,      10.5,   11.6    |       running,        Running,        |       None,   None
openff-jac-qca-psi4-pyddx-600-6cf5cf797c-kwlvx,     35,     0.1,    0.0     |       running,        Running,        |       terminated,     OOMKilled
openff-jac-qca-psi4-pyddx-600-6cf5cf797c-nbk7d,     0,      14.4,   10.1    |       running,        Running,        |       None,   None
openff-jac-qca-psi4-pyddx-600-6cf5cf797c-pcv67,     5,      0.1,    0.0     |       running,        Running,        |       terminated,     OOMKilled
openff-jac-qca-psi4-pyddx-600-6cf5cf797c-q287c,     1,      10.8,   11.2    |       running,        Running,        |       terminated,     OOMKilled
openff-jac-qca-psi4-pyddx-600-6cf5cf797c-wshwz,     7,      15.2,   11.9    |       running,        Running,        |       terminated,     OOMKilled
openff-jac-qca-psi4-pyddx-600-6cf5cf797c-wvj5q,     19,     14.0,   10.5    |       running,        Running,        |       terminated,     OOMKilled

UserWarning: No CPU unit detected, assuming whole CPU.
  warnings.warn("No CPU unit detected, assuming whole CPU.")

Use a Different Configuration

Multiple configurations can be used. The configurations are stored and accessed in the MyData class. One can switch between them by calling MyDataInstance.set_configuration(configuration=”qca-psi4”) to switch to another configuration. Internally, kubecustom will use the configuration to determine the docker image and other settings. See Use a Different Configuration for the template options.

tag = "my-tag"

MyDataInstance.set_configuration(configuration="qca-psi4")
print(get_deployment_name(tag))

MyDataInstance.set_configuration(configuration="qca-xtb")
print(get_deployment_name(tag))

openff-jac-qca-psi4-my-tag
openff-jac-qca-xtb-my-tag

Get Pod CPU and Memory Usage

output = utilization_per_deployment(namespace="openforcefield", verbose=True,) # keep_key="pyddx")
# When containers are "waiting" their resources aren't counted... this means this plot shows higher utilization than grafana

Dep Name, Nreplicas, |  Avg Mem Ut%, Min Mem Ut%, Max Mem Ut%, Mem Req, |  Avg CPU Ut%, Min CPU Ut%, Max CPU Ut%, CPU Req
_____________________________________________________________________________
openff-jac-qca-psi4-pyddx-600       10      |       24.4    0.2     47.4    32.0    |       42.2    0.0     76.1    16.0