gen3-dataops-toolkit 2.0.0__py3-none-any.whl

g3dt/services/indexd/register_indexd.py

@@ -0,0 +1,240 @@
+"""Register S3 data files with Gen3 indexd.
+
+Scans one or more S3 prefixes, collects file metadata (name, md5, size),
+registers each file with the indexd API, and writes results to Glue
+Iceberg tables for downstream use by dbt silver models.
+
+Usage
+-----
+python register_indexd.py \
+    --s3-paths s3://bucket/path1 s3://bucket/path2 \
+    --study edcad \
+    --env staging
+
+Studies come from the g3dt.yaml marker or s3://<metadata-bucket>/config/studies.yaml.
+"""
+
+import sys
+import logging
+import argparse
+import json
+
+import yaml
+from gen3.auth import Gen3Auth
+from gen3.index import Gen3Index
+
+from g3dt.upload.metadata_submitter import (
+    create_boto3_session,
+    get_gen3_api_key_aws_secret,
+    infer_api_endpoint_from_jwt,
+)
+from g3dt.indexd.indexd_registrar import (
+    scan_s3_files,
+    register_files_with_indexd,
+    write_to_glue,
+)
+
+
+def setup_logger():
+    logger = logging.getLogger()
+    logger.setLevel(logging.INFO)
+    if not logger.handlers:
+        handler = logging.StreamHandler(sys.stdout)
+        formatter = logging.Formatter(
+            "%(asctime)s - %(name)s - %(levelname)s - %(message)s"
+        )
+        handler.setFormatter(formatter)
+        logger.addHandler(handler)
+    return logger
+
+
+# Shared config resolution (SSM-backed) — see src/g3dt/config.py
+from g3dt import config as g3dt_config, resolver  # noqa: E402
+
+
+def main():
+    logger = setup_logger()
+
+    parser = argparse.ArgumentParser(
+        description="Register S3 files with Gen3 indexd"
+    )
+    parser.add_argument(
+        "--s3-paths",
+        nargs="+",
+        required=True,
+        help="One or more S3 prefixes to scan for files",
+    )
+    parser.add_argument(
+        "--study",
+        required=True,
+        help="Study key (e.g. edcad, ausdiab, caughtcad)",
+    )
+    parser.add_argument(
+        "--env",
+        required=True,
+        help="Environment (e.g. test, staging, prod, "
+        "staging_ec2, prod_ec2)",
+    )
+    parser.add_argument(
+        "--dry-run",
+        action="store_true",
+        help="Scan and write file_metadata only; skip indexd "
+        "registration",
+    )
+
+    args = parser.parse_args()
+
+    # Env facts + resource names from SSM; the study registry from the marker
+    # or s3://<metadata-bucket>/config/studies.yaml.
+    try:
+        env_cfg = g3dt_config.resolve_env(args.env)
+        study_cfg = g3dt_config.resolve_study(args.study, args.env)
+    except g3dt_config.ConfigError as exc:
+        logger.error(str(exc))
+        sys.exit(1)
+
+    project_id = study_cfg.project_id
+    program_id = study_cfg.program_id
+    authz = [f"/programs/{program_id}/projects/{project_id}"]
+
+    aws_secret_name = env_cfg.aws_secret_name
+    aws_profile = env_cfg.aws_profile
+    aws_region = env_cfg.region
+
+    rc = resolver.resolve(
+        g3dt_config.require_project(),
+        g3dt_config.env_base(args.env),
+        profile=aws_profile,
+    )
+    athena_s3_output = rc.athena_output_location
+    workgroup = rc.athena_workgroup
+
+    # Indexd tables: conventional names in the env's metadata DB/bucket
+    # (exactly like the CDK's `releases` table).
+    fm_database = rc.metadata_db
+    fm_table = g3dt_config.FILE_METADATA_TABLE
+    reg_database = rc.metadata_db
+    reg_table = g3dt_config.INDEXD_REGISTRY_TABLE
+    indexd_s3_path = f"s3://{rc.metadata_bucket}/{g3dt_config.INDEXD_PREFIX}"
+
+    # --- boto3 session ---
+    session = create_boto3_session(
+        aws_profile=aws_profile, aws_region=aws_region
+    )
+
+    # --- Fetch API key early so endpoint can be stamped on tables ---
+    if aws_secret_name.startswith("/"):
+        logger.info(
+            "Loading Gen3 API key from local file: %s",
+            aws_secret_name,
+        )
+        with open(aws_secret_name, "r") as f:
+            api_key = json.load(f)
+    else:
+        logger.info(
+            "Fetching Gen3 API key from Secrets Manager: %s",
+            aws_secret_name,
+        )
+        api_key = get_gen3_api_key_aws_secret(
+            secret_name=aws_secret_name,
+            region_name=aws_region,
+            session=session,
+        )
+
+    base_url = infer_api_endpoint_from_jwt(api_key["api_key"])
+    indexd_endpoint = base_url.replace("api/v0", "index/index")
+    logger.info("Indexd endpoint: %s", indexd_endpoint)
+
+    # --- Phase 1: scan S3 ---
+    logger.info(
+        "Scanning %d S3 path(s): %s",
+        len(args.s3_paths),
+        args.s3_paths,
+    )
+    file_df = scan_s3_files(args.s3_paths, boto3_session=session)
+    file_df["study_id"] = args.study
+    file_df["indexd_endpoint"] = indexd_endpoint
+    logger.info("Found %d files.", len(file_df))
+
+    if file_df.empty:
+        logger.warning("No files found. Exiting.")
+        sys.exit(0)
+
+    # --- Write file_metadata table ---
+    fm_location = (
+        f"{indexd_s3_path.rstrip('/')}/file_metadata/"
+        if indexd_s3_path
+        else None
+    )
+    logger.info(
+        "Writing file metadata to %s.%s", fm_database, fm_table
+    )
+    write_to_glue(
+        df=file_df,
+        database=fm_database,
+        table=fm_table,
+        athena_s3_output=athena_s3_output,
+        table_location=fm_location,
+        workgroup=workgroup,
+        partition_cols=["study_id", "indexd_endpoint"],
+        merge_cols=["file_name", "study_id", "indexd_endpoint"],
+        schema_evolution=True,
+        boto3_session=session,
+    )
+
+    if args.dry_run:
+        logger.info("Dry run — skipping indexd registration.")
+        sys.exit(0)
+
+    auth = Gen3Auth(refresh_token=api_key)
+    index = Gen3Index(auth)
+
+    logger.info(
+        "Registering %d files with indexd (authz=%s)",
+        len(file_df),
+        authz,
+    )
+    registry_df = register_files_with_indexd(
+        index, file_df, authz
+    )
+    logger.info(
+        "Successfully registered %d / %d files.",
+        len(registry_df),
+        len(file_df),
+    )
+
+    if registry_df.empty:
+        logger.warning(
+            "No new registrations. All files may already exist."
+        )
+        sys.exit(0)
+
+    # --- Write indexd_registry table ---
+    reg_location = (
+        f"{indexd_s3_path.rstrip('/')}/indexd_registry/"
+        if indexd_s3_path
+        else None
+    )
+    logger.info(
+        "Writing indexd registry to %s.%s",
+        reg_database,
+        reg_table,
+    )
+    write_to_glue(
+        df=registry_df,
+        database=reg_database,
+        table=reg_table,
+        athena_s3_output=athena_s3_output,
+        table_location=reg_location,
+        workgroup=workgroup,
+        partition_cols=["study_id", "indexd_endpoint"],
+        merge_cols=["did"],
+        schema_evolution=True,
+        boto3_session=session,
+    )
+
+    logger.info("Done.")
+
+
+if __name__ == "__main__":
+    main()

g3dt/services/k8s_ops/argocd_restart_etl.sh

@@ -0,0 +1,140 @@
+#!/bin/bash
+# Script to restart gen3 etl cronjob
+usage() {
+    echo "Usage: $0 [-e ENVIRONMENT] [-d DOMAIN] [-a APPNAME] [-n NAMESPACE] [-c ETL_CRONJOB] [-t CONTAINER] [-l] [-s]"
+    echo "  -e ENVIRONMENT Environment profile (e.g. test, staging, prod). Config comes from G3DT_* env vars set by the g3dt CLI."
+    echo "  -d DOMAIN      The domain for argocd login (example: cd.cad.test.biocommons.org.au)"
+    echo "  -a APPNAME     The application name (example: uatgen3)"
+    echo "  -n NAMESPACE   The namespace for the resources (example: cad)"
+    echo "  -c ETL_CRONJOB The name of the ETL cronjob to run (default: etl-cronjob)"
+    echo "  -t CONTAINER   The name of the container to check logs from (default: tube)"
+    echo "  -l             Bypass login"
+    echo "  -s             Sync the argocd app before restarting resources"
+    exit 1
+}
+
+set -eo pipefail
+
+# default values
+ETL_CRONJOB="etl-cronjob"
+CONTAINER_TO_CHECK="tube"
+LOGIN_REQUIRED=true
+SYNC_APP=false
+ENVIRONMENT=""
+
+while getopts "e:d:a:n:c:t:hls" opt; do
+    case ${opt} in
+        e ) ENVIRONMENT=$OPTARG ;;
+        d ) DOMAIN=$OPTARG ;;
+        a ) APPNAME=$OPTARG ;;
+        n ) NAMESPACE=$OPTARG ;;
+        c ) ETL_CRONJOB=$OPTARG ;;
+        t ) CONTAINER_TO_CHECK=$OPTARG ;;
+        l ) LOGIN_REQUIRED=false ;;
+        s ) SYNC_APP=true ;;
+        h ) usage ;;
+        \? ) usage ;;
+    esac
+done
+
+if ! command -v argocd &> /dev/null; then
+    echo "argocd CLI could not be found. Please install it to proceed."
+    exit 1
+fi
+
+# If an environment is specified (display-only), set up AWS profile, kubeconfig,
+# and ArgoCD details from G3DT_* env vars exported by the g3dt CLI.
+if [ -n "$ENVIRONMENT" ]; then
+    CLUSTER_NAME="${G3DT_CLUSTER_NAME:?G3DT_CLUSTER_NAME not set — run via the g3dt CLI}"
+    REGION="${G3DT_REGION:-ap-southeast-2}"
+
+    # Set DOMAIN, APPNAME, NAMESPACE from env vars if not already set via flags
+    DOMAIN="${DOMAIN:-${G3DT_DOMAIN:?G3DT_DOMAIN not set — run via the g3dt CLI}}"
+    APPNAME="${APPNAME:-${G3DT_APP_NAME:?G3DT_APP_NAME not set — run via the g3dt CLI}}"
+    NAMESPACE="${NAMESPACE:-${G3DT_NAMESPACE:?G3DT_NAMESPACE not set — run via the g3dt CLI}}"
+
+    # Never export an empty AWS_PROFILE (empty means ambient credentials).
+    if [ -n "${G3DT_AWS_PROFILE:-}" ]; then
+        export AWS_PROFILE="${G3DT_AWS_PROFILE}"
+        echo "==== Configuring AWS PROFILE as '${AWS_PROFILE}' for environment '${ENVIRONMENT}' ===="
+    else
+        echo "==== No AWS profile set for '${ENVIRONMENT}' — using ambient AWS credentials ===="
+    fi
+
+    echo "Updating kubeconfig for cluster: ${CLUSTER_NAME}"
+    if ! aws eks update-kubeconfig --name "${CLUSTER_NAME}" --region "${REGION}"; then
+        echo "ERROR: Failed to update kubeconfig. Check your AWS credentials/profile."
+        exit 1
+    fi
+fi
+
+# Save the current kubectl context before argocd login (which changes it)
+KUBE_CONTEXT=$(kubectl config current-context 2>/dev/null)
+
+if [ "$LOGIN_REQUIRED" = true ]; then
+    echo "logging into argocd via sso"
+    argocd login --sso $DOMAIN
+    echo "login successful"
+else
+    echo "Bypassing login as per user request"
+fi
+
+# Restore the kubectl context after argocd login
+if [ -n "$KUBE_CONTEXT" ]; then
+    echo "Restoring kubectl context to: ${KUBE_CONTEXT}"
+    kubectl config use-context "$KUBE_CONTEXT"
+fi
+
+if [ "$SYNC_APP" = true ]; then
+    echo "Syncing ArgoCD app: $APPNAME"
+    argocd app sync $APPNAME
+    if [ $? -eq 0 ]; then
+        echo "App $APPNAME synced successfully."
+    else
+        echo "Failed to sync app $APPNAME."
+        exit 1
+    fi
+fi
+
+RESOURCE="${ETL_CRONJOB}-$(date +%Y%m%d%H%M)"
+JOB_NAME=$(kubectl create job --from=cronjob/${ETL_CRONJOB} ${RESOURCE} --namespace ${NAMESPACE} -o name | awk -F'/' '{print $2}')
+
+if [ -z "$JOB_NAME" ]; then
+  echo "Failed to create job. Exiting."
+  exit 1
+fi
+
+echo "Job '$JOB_NAME' created. Waiting for it to complete..."
+
+while true; do
+  SUCCEEDED=$(kubectl get job $JOB_NAME -n $NAMESPACE -o jsonpath='{.status.succeeded}')
+  FAILED=$(kubectl get job $JOB_NAME -n $NAMESPACE -o jsonpath='{.status.failed}')
+
+  if [ "$SUCCEEDED" == "1" ]; then
+    echo "✅ Job '$JOB_NAME' completed successfully!"
+    break
+  elif [ "$FAILED" == "1" ]; then
+    echo "⚠️ Job '$JOB_NAME' failed."
+    break
+  else
+    echo "⏳ Job '$JOB_NAME' is still running... (Succeeded: ${SUCCEEDED:-0}, Failed: ${FAILED:-0})"
+    sleep 10
+  fi
+done
+
+
+# Even if pod fails, it may still have passed, need to check logs
+POD_NAME=$(kubectl get pods -n $NAMESPACE -l job-name=$JOB_NAME -o jsonpath='{.items[0].metadata.name}')
+LOGS=$(kubectl logs $POD_NAME -n $NAMESPACE -c $CONTAINER_TO_CHECK)
+
+
+if echo "$LOGS" | grep -q "Exit code: 0"; then
+    echo "✅ ✅ ✅Log check passed! Found 'Exit code: 0'. The job is truly successful."
+else
+    echo "⚠️ Job Failed, as well as the log validation! Please review output:"
+    echo "--------------------- LOGS START ---------------------"
+    echo "$LOGS"
+    echo "---------------------- LOGS END ----------------------"
+fi
+
+echo "Script finished."

g3dt/services/k8s_ops/argocd_restart_ms.sh

@@ -0,0 +1,102 @@
+#!/bin/bash
+# Script to restart argocd microservices for schema redeployment
+
+# Function to display help message
+usage() {
+    echo "Usage: $0 [-d DOMAIN] [-a APPNAME] [-r RESOURCES] [-n NAMESPACE] [-k KIND] [-l] [-s]"
+    echo "  -d DOMAIN      The domain for argocd login (example: cd.cad.test.biocommons.org.au)"
+    echo "  -a APPNAME     The application name (example: uatgen3)"
+    echo "  -r RESOURCES   Comma-separated string of microservice names to restart (example: sheepdog-deployment,peregrine-deployment )"
+    echo "  -n NAMESPACE   The namespace for the resources (example: cad)"
+    echo "  -k KIND        The kind of resource to restart (default: Deployment)"
+    echo "  -l             Bypass login"
+    echo "  -s             Sync the argocd app before restarting resources"
+    exit 1
+}
+
+set -eo pipefail
+
+# Parse command line arguments
+LOGIN_REQUIRED=true
+KIND="Deployment"
+SYNC_APP=false
+while getopts "d:a:r:n:k:hls" opt; do
+    case ${opt} in
+        d )
+            DOMAIN=$OPTARG
+            ;;
+        a )
+            APPNAME=$OPTARG
+            ;;
+        r )
+            IFS=',' read -r -a RESOURCES <<< "$OPTARG"
+            ;;
+        n )
+            NAMESPACE=$OPTARG
+            ;;
+        k )
+            KIND=$OPTARG
+            ;;
+        l )
+            LOGIN_REQUIRED=false
+            ;;
+        s )
+            SYNC_APP=true
+            ;;
+        h )
+            usage
+            ;;
+        \? )
+            usage
+            ;;
+    esac
+done
+
+# Check if argocd is installed
+if ! command -v argocd &> /dev/null
+then
+    echo "argocd CLI could not be found. Please install it to proceed."
+    exit 1
+fi
+
+if [ "$LOGIN_REQUIRED" = true ]; then
+    echo "logging into argocd via sso"
+    argocd login --sso $DOMAIN
+    echo "login successful"
+else
+    echo "Bypassing login as per user request"
+fi
+
+if [ "$SYNC_APP" = true ]; then
+    echo "Syncing ArgoCD app: $APPNAME"
+    argocd app sync $APPNAME
+    if [ $? -eq 0 ]; then
+        echo "App $APPNAME synced successfully."
+    else
+        echo "Failed to sync app $APPNAME."
+        exit 1
+    fi
+fi
+
+# Iterate through resources
+for RESOURCE in "${RESOURCES[@]}"; do
+  echo "Restarting resource: $RESOURCE"
+  
+  # Run the restart action
+  argocd app actions run $APPNAME restart --kind $KIND --resource-name $RESOURCE --namespace $NAMESPACE
+
+  sleep 5
+  
+  # Wait for the resource to complete its restart
+  echo "Waiting for $RESOURCE to finish..."
+  while true; do
+    echo "Checking $RESOURCE status..."
+    STATUS=$(argocd app get $APPNAME -o json | jq --arg RESOURCE "$RESOURCE" '.status.resources[] | select(.name == $RESOURCE) | .health.status')
+    echo "Status: $STATUS"
+    if [ "$STATUS" == "\"Healthy\"" ]; then
+      echo "$RESOURCE restarted successfully"
+      break
+    fi
+    sleep 5 # Check every 5 seconds
+  done
+done

g3dt/services/k8s_ops/argocd_restart_schema.sh

@@ -0,0 +1,106 @@
+#!/bin/bash
+# Script to restart argocd microservices for schema redeployment
+
+# Function to display help message
+usage() {
+    echo "Usage: $0 [-d DOMAIN] [-a APPNAME] [-r RESOURCES] [-n NAMESPACE] [-k KIND] [-l] [-s]"
+    echo "  -d DOMAIN      The domain for argocd login (example: cd.cad.test.biocommons.org.au)"
+    echo "  -a APPNAME     The application name (example: uatgen3)"
+    echo "  -r RESOURCES   Comma-separated string of microservice names to restart (default: \"sheepdog-deployment\" \"peregrine-deployment\" \"guppy-deployment\" \"portal-deployment\")"
+    echo "  -n NAMESPACE   The namespace for the resources (default: cad)"
+    echo "  -k KIND        The kind of resource to restart (default: Deployment)"
+    echo "  -l             Bypass login"
+    echo "  -s             Run 'argocd app sync' before restarts"
+    exit 1
+}
+
+set -eo pipefail
+
+# Set default values
+RESOURCES=("sheepdog-deployment" "peregrine-deployment" "guppy-deployment" "portal-deployment")
+NAMESPACE="cad"
+KIND="Deployment"
+LOGIN_REQUIRED=true
+SYNC_APP=false
+
+# Parse command line arguments
+while getopts "d:a:r:n:k:hls" opt; do
+    case ${opt} in
+        d )
+            DOMAIN=$OPTARG
+            ;;
+        a )
+            APPNAME=$OPTARG
+            ;;
+        r )
+            IFS=',' read -r -a RESOURCES <<< "$OPTARG"
+            ;;
+        n )
+            NAMESPACE=$OPTARG
+            ;;
+        k )
+            KIND=$OPTARG
+            ;;
+        l )
+            LOGIN_REQUIRED=false
+            ;;
+        s )
+            SYNC_APP=true
+            ;;
+        h )
+            usage
+            ;;
+        \? )
+            usage
+            ;;
+    esac
+done
+
+# Check if argocd is installed
+if ! command -v argocd &> /dev/null
+then
+    echo "argocd CLI could not be found. Please install it to proceed."
+    exit 1
+fi
+
+if [ "$LOGIN_REQUIRED" = true ]; then
+    echo "logging into argocd via sso"
+    argocd login --sso $DOMAIN
+    echo "login successful"
+else
+    echo "Bypassing login as per user request"
+fi
+
+if [ "$SYNC_APP" = true ]; then
+    echo "Syncing ArgoCD app: $APPNAME"
+    argocd app sync $APPNAME
+    if [ $? -eq 0 ]; then
+        echo "App $APPNAME synced successfully."
+    else
+        echo "Failed to sync app $APPNAME."
+        exit 1
+    fi
+fi
+
+# Iterate through resources
+for RESOURCE in "${RESOURCES[@]}"; do
+  echo "Restarting resource: $RESOURCE"
+  
+  # Run the restart action
+  argocd app actions run $APPNAME restart --kind $KIND --resource-name $RESOURCE --namespace $NAMESPACE
+
+  sleep 5
+  
+  # Wait for the resource to complete its restart
+  echo "Waiting for $RESOURCE to finish..."
+  while true; do
+    echo "Checking $RESOURCE status..."
+    STATUS=$(argocd app get $APPNAME -o json | jq --arg RESOURCE "$RESOURCE" '.status.resources[] | select(.name == $RESOURCE) | .health.status')
+    echo "Status: $STATUS"
+    if [ "$STATUS" == "\"Healthy\"" ]; then
+      echo "$RESOURCE restarted successfully"
+      break
+    fi
+    sleep 5 # Check every 5 seconds
+  done
+done

g3dt/services/k8s_ops/login_to_pod.sh

@@ -0,0 +1,110 @@
+#!/bin/bash
+
+# login_testenv_pod.sh
+# Script to log into a specified pod in a Kubernetes environment using a grep pattern.
+
+set -e
+
+show_help() {
+    cat << EOF
+Usage: ${0##*/} [OPTIONS]
+
+Log into a running pod on aws. Direct SSO identity is used by default.
+Also make sure you have an aws SSO profile configured on AWS cli.
+
+Defaults come from the G3DT_* environment variables exported by the g3dt CLI
+when set; flags always override them.
+
+Options:
+  -p PROFILE        AWS CLI profile to use (default: \$G3DT_AWS_PROFILE, else ambient credentials)
+  -r REGION         AWS region (default: \$G3DT_REGION, else ap-southeast-2)
+  -c CLUSTER_NAME   EKS cluster name (default: \$G3DT_CLUSTER_NAME)
+  -a ROLE_ARN       Optional AWS IAM role ARN to assume (default: \$G3DT_EKS_ARN)
+  -n NAMESPACE      Kubernetes namespace (default: \$G3DT_NAMESPACE, else cad)
+  -g GREP_PATTERN   Pattern to grep for pod name (default: guppy)
+  -x                Print all resolved commands before running them
+  -h                Show this help message and exit
+
+Example:
+  ${0##*/} -p myprofile -n mynamespace -g sheepdog
+
+EOF
+}
+
+set -eo pipefail
+
+# Default values (taken from G3DT_* env vars when the g3dt CLI exported them)
+PROFILE="${G3DT_AWS_PROFILE:-}"
+AWS_REGION="${G3DT_REGION:-ap-southeast-2}"
+CLUSTER_NAME="${G3DT_CLUSTER_NAME:?G3DT_CLUSTER_NAME not set — run via the g3dt CLI or pass -c}"
+ROLE_ARN="${G3DT_EKS_ARN:-}"
+NAME_SPACE="${G3DT_NAMESPACE:?G3DT_NAMESPACE not set — run via the g3dt CLI or pass -n}"
+GREP_PATTERN="guppy"
+PRINT_COMMANDS=0
+
+while getopts "p:r:c:a:n:g:xh" opt; do
+  case $opt in
+    p) PROFILE="$OPTARG" ;;
+    r) AWS_REGION="$OPTARG" ;;
+    c) CLUSTER_NAME="$OPTARG" ;;
+    a) ROLE_ARN="$OPTARG" ;;
+    n) NAME_SPACE="$OPTARG" ;;
+    g) GREP_PATTERN="$OPTARG" ;;
+    x) PRINT_COMMANDS=1 ;;
+    h)
+      show_help
+      exit 0
+      ;;
+    \?)
+      show_help >&2
+      exit 1
+      ;;
+  esac
+done
+
+# Prepare commands
+CMD_AWS_SSO_LOGIN="aws sso login --profile ${PROFILE}"
+CMD_AWS_EKS_UPDATE="aws eks update-kubeconfig --name ${CLUSTER_NAME} --region ${AWS_REGION} --profile ${PROFILE}"
+if [[ -n "$ROLE_ARN" ]]; then
+    CMD_AWS_EKS_UPDATE="${CMD_AWS_EKS_UPDATE} --role-arn ${ROLE_ARN}"
+fi
+CMD_GET_POD="kubectl get pods -n \"${NAME_SPACE}\" | grep Running | grep \"${GREP_PATTERN}\" | awk '{print \$1}'"
+
+if [[ $PRINT_COMMANDS -eq 1 ]]; then
+    echo "Resolved commands to be run:"
+    echo ""
+    echo "# 1. AWS SSO Login"
+    echo "$CMD_AWS_SSO_LOGIN"
+    echo ""
+    echo "# 2. Update kubeconfig"
+    echo "$CMD_AWS_EKS_UPDATE"
+    echo ""
+    echo "# 3. Get pod name"
+    echo "$CMD_GET_POD"
+    echo ""
+fi
+
+echo "Logging into aws sso with profile: $PROFILE"
+eval "$CMD_AWS_SSO_LOGIN"
+
+echo "Updating Kubernetes context for cluster: $CLUSTER_NAME in region: $AWS_REGION"
+eval "$CMD_AWS_EKS_UPDATE"
+
+# Get pod name using grep pattern
+POD_NAME=$(kubectl get pods -n "${NAME_SPACE}" | grep Running | grep "${GREP_PATTERN}" | awk '{print $1}')
+
+if [[ -z "$POD_NAME" ]]; then
+    echo "Error: No running pod matching pattern '${GREP_PATTERN}' found in namespace '${NAME_SPACE}'." >&2
+    exit 1
+fi
+
+CMD_KUBECTL_EXEC="kubectl exec -it \"$POD_NAME\" -n \"${NAME_SPACE}\" -- bash"
+
+if [[ $PRINT_COMMANDS -eq 1 ]]; then
+    echo "# 4. Exec into pod"
+    echo "$CMD_KUBECTL_EXEC"
+    echo ""
+fi
+
+echo "Logging into pod: $POD_NAME"
+kubectl exec -it "$POD_NAME" -n "${NAME_SPACE}" -- bash