[WIP] Support Cloud-Init in GCP

sky/clouds/service_catalog/constants.py

@@ -1,6 +1,6 @@
 """Constants used for service catalog."""
 import os
 
-HOSTED_CATALOG_DIR_URL = 'https://raw.githubusercontent.com/skypilot-org/skypilot-catalog/master/catalogs'  # pylint: disable=line-too-long
+HOSTED_CATALOG_DIR_URL = 'https://raw.githubusercontent.com/cblmemo/skypilot-catalog/ubuntu-in-gcp/catalogs'  # pylint: disable=line-too-long
 CATALOG_SCHEMA_VERSION = 'v5'
 LOCAL_CATALOG_DIR = os.path.expanduser('~/.sky/catalogs/')

sky/skylet/providers/gcp/config.py

@@ -1,7 +1,9 @@
+import base64
 import copy
 import json
 import logging
 import os
+import textwrap
 import time
 from functools import partial
 from typing import Dict, List, Set, Tuple
@@ -309,6 +311,8 @@ def bootstrap_gcp(config):
     config = _configure_iam_role(config, crm, iam)
     config = _configure_key_pair(config, compute)
     config = _configure_subnet(config, compute)
+    config = _configure_cloud_init(config)
+    print("DEBUG HERE", config)
 
     return config
 
@@ -841,6 +845,41 @@ def _configure_subnet(config, compute):
     return config
 
 
+def _configure_cloud_init(config):
+    """Configure cloud-init for the cluster."""
+    config = copy.deepcopy(config)
+
+    node_configs = [
+        node_type["node_config"]
+        for node_type in config["available_node_types"].values()
+    ]
+
+    cloud_init_encoded_script = base64.b64encode(
+        textwrap.dedent(
+            """\
+        #cloud-config
+        write_files:
+          - path: /etc/apt/apt.conf.d/20auto-upgrades
+            content: |
+              APT::Periodic::Update-Package-Lists "0";
+              APT::Periodic::Download-Upgradeable-Packages "0";
+              APT::Periodic::AutocleanInterval "0";
+              APT::Periodic::Unattended-Upgrade "0";
+          - path: /etc/apt/apt.conf.d/10cloudinit-disable
+            content: |
+              APT::Periodic::Enable "0";
+        """
+        ).encode("utf-8")
+    ).decode("utf-8")
+
+    for node_config in node_configs:
+        node_config["metadata"] = {
+            "items": [{"key": "user-data", "value": cloud_init_encoded_script}]
+        }
+
+    return config
+
+
 def _create_firewall_rule(config, compute, body):
     operation = (
         compute.firewalls()

sky/templates/gcp-ray.yml.j2

@@ -140,7 +140,6 @@ initialization_commands: []
 # Increment the following for catching performance bugs easier:
 #   current num items (num SSH connections): 1  (+1 if tpu_vm)
 setup_commands:
-  # Disable `unattended-upgrades` to prevent apt-get from hanging. It should be called at the beginning before the process started to avoid being blocked. (This is a temporary fix.)
   # Line 'mkdir -p ..': Create ~/.ssh/config file in case the file does not exist in the custom image.
   # Line 'pip3 --v ..': Make sure python3 & pip3 are available on this image.
   # Line 'which conda ..': some images (TPU VM) do not install conda by
@@ -150,17 +149,7 @@ setup_commands:
   # Line 'sudo grep ..': set the number of threads per process to unlimited to avoid ray job submit stucking issue when the number of running ray jobs increase.
   # Line 'mkdir -p ..': disable host key check
   # Line 'python3 -c ..': patch the buggy ray files and enable `-o allow_other` option for `goofys`
-  - function mylsof { p=$(for pid in /proc/{0..9}*; do i=$(basename "$pid"); for file in "$pid"/fd/*; do link=$(readlink -e "$file"); if [ "$link" = "$1" ]; then echo "$i"; fi; done; done); echo "$p"; };
-    sudo systemctl stop unattended-upgrades || true;
-    sudo systemctl disable unattended-upgrades || true;
-    sudo sed -i 's/Unattended-Upgrade "1"/Unattended-Upgrade "0"/g' /etc/apt/apt.conf.d/20auto-upgrades || true;
-    p=$(mylsof "/var/lib/dpkg/lock-frontend"); echo "$p";
-    sudo kill -9 `echo "$p" | tail -n 1` || true;
-    sudo rm /var/lib/dpkg/lock-frontend;
-    sudo pkill -9 dpkg;
-    sudo pkill -9 apt-get;
-    sudo dpkg --configure --force-overwrite -a;
-    mkdir -p ~/.ssh; touch ~/.ssh/config;
+  - mkdir -p ~/.ssh; touch ~/.ssh/config;
     pip3 --version > /dev/null 2>&1 || (curl -sSL https://bootstrap.pypa.io/get-pip.py -o get-pip.py && python3 get-pip.py && echo "PATH=$HOME/.local/bin:$PATH" >> ~/.bashrc);
     (type -a python | grep -q python3) || echo 'alias python=python3' >> ~/.bashrc;
     (type -a pip | grep -q pip3) || echo 'alias pip=pip3' >> ~/.bashrc;