cluster-builder 0.5.2__tar.gz → 0.6.0__tar.gz

This diff represents the content of publicly available package versions that have been released to one of the supported registries. The information contained in this diff is provided for informational purposes only and reflects changes between package versions as they appear in their respective public registries.
Files changed (35) hide show
  1. {cluster_builder-0.5.2 → cluster_builder-0.6.0}/PKG-INFO +12 -14
  2. {cluster_builder-0.5.2 → cluster_builder-0.6.0}/README.md +11 -13
  3. {cluster_builder-0.5.2 → cluster_builder-0.6.0}/cluster_builder/config/__init__.py +2 -2
  4. {cluster_builder-0.5.2 → cluster_builder-0.6.0}/cluster_builder/config/cluster.py +80 -7
  5. {cluster_builder-0.5.2 → cluster_builder-0.6.0}/cluster_builder/infrastructure/executor.py +7 -4
  6. {cluster_builder-0.5.2 → cluster_builder-0.6.0}/cluster_builder/infrastructure/templates.py +2 -33
  7. {cluster_builder-0.5.2 → cluster_builder-0.6.0}/cluster_builder/swarmchestrate.py +233 -39
  8. {cluster_builder-0.5.2 → cluster_builder-0.6.0}/cluster_builder/templates/aws/main.tf +31 -11
  9. {cluster_builder-0.5.2 → cluster_builder-0.6.0}/cluster_builder/templates/deploy_manifest.tf +12 -2
  10. {cluster_builder-0.5.2 → cluster_builder-0.6.0}/cluster_builder/templates/edge/main.tf +32 -11
  11. {cluster_builder-0.5.2 → cluster_builder-0.6.0}/cluster_builder/templates/k3s_provider.tf +1 -1
  12. {cluster_builder-0.5.2 → cluster_builder-0.6.0}/cluster_builder/templates/openstack/main.tf +139 -41
  13. {cluster_builder-0.5.2 → cluster_builder-0.6.0}/cluster_builder/templates/registry_secret.tf +12 -2
  14. cluster_builder-0.6.0/cluster_builder/templates/remove_manifest.tf +43 -0
  15. {cluster_builder-0.5.2 → cluster_builder-0.6.0}/cluster_builder/utils/hcl.py +36 -15
  16. {cluster_builder-0.5.2 → cluster_builder-0.6.0}/cluster_builder/utils/logging.py +1 -4
  17. {cluster_builder-0.5.2 → cluster_builder-0.6.0}/cluster_builder.egg-info/PKG-INFO +12 -14
  18. {cluster_builder-0.5.2 → cluster_builder-0.6.0}/cluster_builder.egg-info/SOURCES.txt +1 -3
  19. {cluster_builder-0.5.2 → cluster_builder-0.6.0}/pyproject.toml +1 -1
  20. {cluster_builder-0.5.2 → cluster_builder-0.6.0}/tests/test_hcl.py +5 -3
  21. cluster_builder-0.5.2/cluster_builder/templates/ha_user_data.sh.tpl +0 -33
  22. cluster_builder-0.5.2/cluster_builder/templates/master_user_data.sh.tpl +0 -37
  23. cluster_builder-0.5.2/cluster_builder/templates/worker_user_data.sh.tpl +0 -34
  24. {cluster_builder-0.5.2 → cluster_builder-0.6.0}/LICENSE +0 -0
  25. {cluster_builder-0.5.2 → cluster_builder-0.6.0}/cluster_builder/__init__.py +0 -0
  26. {cluster_builder-0.5.2 → cluster_builder-0.6.0}/cluster_builder/config/postgres.py +1 -1
  27. {cluster_builder-0.5.2 → cluster_builder-0.6.0}/cluster_builder/infrastructure/__init__.py +0 -0
  28. {cluster_builder-0.5.2 → cluster_builder-0.6.0}/cluster_builder/templates/aws_provider.tf +0 -0
  29. {cluster_builder-0.5.2 → cluster_builder-0.6.0}/cluster_builder/templates/edge_provider.tf +0 -0
  30. {cluster_builder-0.5.2 → cluster_builder-0.6.0}/cluster_builder/templates/openstack_provider.tf +0 -0
  31. {cluster_builder-0.5.2 → cluster_builder-0.6.0}/cluster_builder/utils/__init__.py +0 -0
  32. {cluster_builder-0.5.2 → cluster_builder-0.6.0}/cluster_builder.egg-info/dependency_links.txt +0 -0
  33. {cluster_builder-0.5.2 → cluster_builder-0.6.0}/cluster_builder.egg-info/requires.txt +0 -0
  34. {cluster_builder-0.5.2 → cluster_builder-0.6.0}/cluster_builder.egg-info/top_level.txt +0 -0
  35. {cluster_builder-0.5.2 → cluster_builder-0.6.0}/setup.cfg +0 -0
@@ -1,6 +1,6 @@
1
1
  Metadata-Version: 2.4
2
2
  Name: cluster-builder
3
- Version: 0.5.2
3
+ Version: 0.6.0
4
4
  Summary: Swarmchestrate cluster builder
5
5
  Author-email: Gunjan <G.Kotak@westminster.ac.uk>, Jay <J.Deslauriers@westminster.ac.uk>
6
6
  License: Apache2
@@ -107,8 +107,7 @@ from cluster_builder import Swarmchestrate
107
107
 
108
108
  # Initialise the orchestrator
109
109
  orchestrator = Swarmchestrate(
110
- template_dir="/path/to/templates",
111
- output_dir="/path/to/output"
110
+ template_dir="/path/to/templates", output_dir="/path/to/output"
112
111
  )
113
112
  ```
114
113
 
@@ -122,7 +121,7 @@ You may define it directly as a Python dictionary or load it from a separate fil
122
121
  #### 2. Load the configuration in Python:
123
122
 
124
123
  ```python
125
- config ={
124
+ config = {
126
125
  # your config fields here
127
126
  }
128
127
 
@@ -144,8 +143,8 @@ To remove a specific node from a cluster:
144
143
  ```python
145
144
  # Remove a node by its resource name
146
145
  orchestrator.remove_node(
147
- cluster_name="your-cluster-name",
148
- resource_name="eloquent_feynman" # The resource identifier of the node
146
+ cluster_name="your_cluster_name",
147
+ resource_name="eloquent-feynman", # The resource identifier of the node
149
148
  )
150
149
  ```
151
150
 
@@ -159,9 +158,7 @@ To completely destroy a cluster and all its nodes:
159
158
 
160
159
  ```python
161
160
  # Destroy the entire cluster
162
- orchestrator.destroy(
163
- cluster_name="your-cluster-name"
164
- )
161
+ orchestrator.destroy(cluster_name="your_cluster_name")
165
162
  ```
166
163
 
167
164
  The **destroy** method:
@@ -179,8 +176,10 @@ The deploy_manifests method copies Kubernetes manifests to the target cluster no
179
176
  orchestrator.deploy_manifests(
180
177
  manifest_folder="path/to/manifests",
181
178
  master_ip="MASTER_NODE_IP",
182
- ssh_key_path="path/to/key.pem",
183
- ssh_user="USERNAME"
179
+ ssh_user="USERNAME",
180
+ ssh_key="path/to/key.pem", # for key auth
181
+ ssh_auth_method="key", # or "password"
182
+ ssh_password="", # required when ssh_auth_method="password"
184
183
  )
185
184
  ```
186
185
  ---
@@ -240,7 +239,7 @@ By default, cluster names are generated automatically. To specify a custom name:
240
239
 
241
240
  ```python
242
241
  config = {
243
- "cluster_name": "production-cluster",
242
+ "cluster_name": "production_cluster",
244
243
  # ... other configuration ...
245
244
  }
246
245
  ```
@@ -250,8 +249,7 @@ config = {
250
249
 
251
250
  Templates should be organised as follows:
252
251
  - `templates/` - Base directory for templates
253
- - `templates/{cloud}/` - Terraform modules for each cloud provider
254
- - `templates/{role}_user_data.sh.tpl` - Node initialisation scripts
252
+ - `templates/{cloud}/` - Opentofu modules for each cloud provider
255
253
  - `templates/{cloud}_provider.tf` - Provider configuration templates
256
254
 
257
255
  ---
@@ -90,8 +90,7 @@ from cluster_builder import Swarmchestrate
90
90
 
91
91
  # Initialise the orchestrator
92
92
  orchestrator = Swarmchestrate(
93
- template_dir="/path/to/templates",
94
- output_dir="/path/to/output"
93
+ template_dir="/path/to/templates", output_dir="/path/to/output"
95
94
  )
96
95
  ```
97
96
 
@@ -105,7 +104,7 @@ You may define it directly as a Python dictionary or load it from a separate fil
105
104
  #### 2. Load the configuration in Python:
106
105
 
107
106
  ```python
108
- config ={
107
+ config = {
109
108
  # your config fields here
110
109
  }
111
110
 
@@ -127,8 +126,8 @@ To remove a specific node from a cluster:
127
126
  ```python
128
127
  # Remove a node by its resource name
129
128
  orchestrator.remove_node(
130
- cluster_name="your-cluster-name",
131
- resource_name="eloquent_feynman" # The resource identifier of the node
129
+ cluster_name="your_cluster_name",
130
+ resource_name="eloquent-feynman", # The resource identifier of the node
132
131
  )
133
132
  ```
134
133
 
@@ -142,9 +141,7 @@ To completely destroy a cluster and all its nodes:
142
141
 
143
142
  ```python
144
143
  # Destroy the entire cluster
145
- orchestrator.destroy(
146
- cluster_name="your-cluster-name"
147
- )
144
+ orchestrator.destroy(cluster_name="your_cluster_name")
148
145
  ```
149
146
 
150
147
  The **destroy** method:
@@ -162,8 +159,10 @@ The deploy_manifests method copies Kubernetes manifests to the target cluster no
162
159
  orchestrator.deploy_manifests(
163
160
  manifest_folder="path/to/manifests",
164
161
  master_ip="MASTER_NODE_IP",
165
- ssh_key_path="path/to/key.pem",
166
- ssh_user="USERNAME"
162
+ ssh_user="USERNAME",
163
+ ssh_key="path/to/key.pem", # for key auth
164
+ ssh_auth_method="key", # or "password"
165
+ ssh_password="", # required when ssh_auth_method="password"
167
166
  )
168
167
  ```
169
168
  ---
@@ -223,7 +222,7 @@ By default, cluster names are generated automatically. To specify a custom name:
223
222
 
224
223
  ```python
225
224
  config = {
226
- "cluster_name": "production-cluster",
225
+ "cluster_name": "production_cluster",
227
226
  # ... other configuration ...
228
227
  }
229
228
  ```
@@ -233,8 +232,7 @@ config = {
233
232
 
234
233
  Templates should be organised as follows:
235
234
  - `templates/` - Base directory for templates
236
- - `templates/{cloud}/` - Terraform modules for each cloud provider
237
- - `templates/{role}_user_data.sh.tpl` - Node initialisation scripts
235
+ - `templates/{cloud}/` - Opentofu modules for each cloud provider
238
236
  - `templates/{cloud}_provider.tf` - Provider configuration templates
239
237
 
240
238
  ---
@@ -2,7 +2,7 @@
2
2
  Configuration management for the Cluster Builder.
3
3
  """
4
4
 
5
- from cluster_builder.config.postgres import PostgresConfig
6
5
  from cluster_builder.config.cluster import ClusterConfig
6
+ from cluster_builder.config.postgres import PostgresConfig
7
7
 
8
- __all__ = ["PostgresConfig", "ClusterConfig"]
8
+ __all__ = ["ClusterConfig", "PostgresConfig"]
@@ -2,17 +2,27 @@
2
2
  Cluster configuration management.
3
3
  """
4
4
 
5
- import os
6
5
  import logging
6
+ import os
7
+ import re
7
8
  import secrets
8
9
  import string
10
+ from typing import Any
11
+
9
12
  from names_generator import generate_name
10
13
 
11
- from typing import Any
12
14
  from cluster_builder.infrastructure import TemplateManager
13
15
 
14
16
  logger = logging.getLogger("swarmchestrate")
15
17
 
18
+ # RFC 1123 label: lowercase alphanumerics or '-', starting/ending with an alphanumeric, max 63 chars
19
+ RFC_1123_LABEL_RE = re.compile(r"^[a-z0-9]([-a-z0-9]*[a-z0-9])?$")
20
+ RFC_1123_MAX_LENGTH = 63
21
+
22
+ # Unquoted PostgreSQL identifier: starts with a letter or underscore, followed
23
+ # by letters, digits, or underscores, max 63 characters. No hyphens allowed.
24
+ PG_IDENTIFIER_RE = re.compile(r"^[a-z_][a-z0-9_]{0,62}$")
25
+
16
26
 
17
27
  class ClusterConfig:
18
28
  """Manages cluster configuration and preparation."""
@@ -32,6 +42,55 @@ class ClusterConfig:
32
42
  self.template_manager = template_manager
33
43
  self.output_dir = output_dir
34
44
 
45
+ @staticmethod
46
+ def validate_rfc1123_name(name: str, field_name: str) -> None:
47
+ """
48
+ Validate that a name is RFC 1123 compliant (used for k8s resource names).
49
+
50
+ Args:
51
+ name: The name to validate
52
+ field_name: Name of the field, used in the error message
53
+
54
+ Raises:
55
+ ValueError: If the name is not RFC 1123 compliant
56
+ """
57
+ if (
58
+ not name
59
+ or len(name) > RFC_1123_MAX_LENGTH
60
+ or not RFC_1123_LABEL_RE.match(name)
61
+ ):
62
+ error_msg = (
63
+ f"Invalid {field_name} '{name}': must be RFC 1123 compliant "
64
+ f"(lowercase alphanumeric characters or '-', start/end with an "
65
+ f"alphanumeric character, max {RFC_1123_MAX_LENGTH} characters). "
66
+ "Underscores are not permitted."
67
+ )
68
+ logger.error(error_msg)
69
+ raise ValueError(error_msg)
70
+
71
+ @staticmethod
72
+ def validate_pg_identifier_name(name: str, field_name: str) -> None:
73
+ """
74
+ Validate that a name is a valid, unquoted PostgreSQL identifier (used
75
+ for cluster_name, which doubles as the Terraform state schema name).
76
+
77
+ Args:
78
+ name: The name to validate
79
+ field_name: Name of the field, used in the error message
80
+
81
+ Raises:
82
+ ValueError: If the name is not a valid PostgreSQL identifier
83
+ """
84
+ if not name or not PG_IDENTIFIER_RE.match(name):
85
+ error_msg = (
86
+ f"Invalid {field_name} '{name}': must start with a lowercase "
87
+ "letter or underscore and contain only lowercase letters, "
88
+ "digits, and underscores (max 63 characters). Hyphens are not "
89
+ "permitted."
90
+ )
91
+ logger.error(error_msg)
92
+ raise ValueError(error_msg)
93
+
35
94
  def get_cluster_output_dir(self, cluster_name: str) -> str:
36
95
  """
37
96
  Get the output directory path for a specific cluster.
@@ -46,7 +105,8 @@ class ClusterConfig:
46
105
 
47
106
  def generate_random_name(self) -> str:
48
107
  """
49
- Generate a readable random string using names-generator.
108
+ Generate a readable, RFC 1123 compliant random string (hyphen-separated)
109
+ using names-generator. Used for resource_name.
50
110
 
51
111
  Returns:
52
112
  A randomly generated name
@@ -56,6 +116,18 @@ class ClusterConfig:
56
116
  logger.debug(f"Generated random name: {name}")
57
117
  return name
58
118
 
119
+ def generate_cluster_name(self) -> str:
120
+ """
121
+ Generate a readable, PostgreSQL-safe random string (underscore-separated)
122
+ using names-generator. Used for cluster_name.
123
+
124
+ Returns:
125
+ A randomly generated cluster name
126
+ """
127
+ name = generate_name()
128
+ logger.debug(f"Generated cluster name: {name}")
129
+ return name
130
+
59
131
  def generate_k3s_token(self, length: int = 16) -> str:
60
132
  """
61
133
  Generate a secure random alphanumeric token for K3s.
@@ -120,13 +192,16 @@ class ClusterConfig:
120
192
 
121
193
  # Generate a cluster name if not provided
122
194
  if "cluster_name" not in prepared_config:
123
- cluster_name = self.generate_random_name()
195
+ cluster_name = self.generate_cluster_name()
124
196
  prepared_config["cluster_name"] = cluster_name
125
197
  logger.debug(f"Creating new cluster: {cluster_name}")
126
198
  else:
127
199
  logger.debug(
128
200
  f"Adding node to existing cluster: {prepared_config['cluster_name']}"
129
201
  )
202
+ self.validate_pg_identifier_name(
203
+ prepared_config["cluster_name"], "cluster_name"
204
+ )
130
205
 
131
206
  cluster_dir = self.get_cluster_output_dir(prepared_config["cluster_name"])
132
207
  logger.debug(f"Cluster directory: {cluster_dir}")
@@ -140,6 +215,7 @@ class ClusterConfig:
140
215
  logger.debug(
141
216
  f" Using provided Resource name: {prepared_config['resource_name']}"
142
217
  )
218
+ self.validate_rfc1123_name(prepared_config["resource_name"], "resource_name")
143
219
 
144
220
  # Create the cluster directory
145
221
  try:
@@ -150,7 +226,4 @@ class ClusterConfig:
150
226
  logger.error(error_msg)
151
227
  raise RuntimeError(error_msg)
152
228
 
153
- # Copy user data template
154
- self.template_manager.copy_user_data_template(role, cloud)
155
-
156
229
  return cluster_dir, prepared_config
@@ -2,8 +2,9 @@
2
2
  Command execution utilities for infrastructure management.
3
3
  """
4
4
 
5
- import subprocess
6
5
  import logging
6
+ import subprocess
7
+ import sys
7
8
 
8
9
  from yaspin import yaspin
9
10
  from yaspin.spinners import Spinners
@@ -19,8 +20,8 @@ class CommandExecutor:
19
20
  command: list,
20
21
  cwd: str,
21
22
  description: str = "command",
22
- timeout: int = None,
23
- env: dict = None, # <-- Add optional env param
23
+ timeout: int | None = None,
24
+ env: dict | None = None, # <-- Add optional env param
24
25
  ) -> str:
25
26
  """
26
27
  Execute a shell command with proper logging and error handling.
@@ -62,8 +63,10 @@ class CommandExecutor:
62
63
  pass # Still running → spinner starts
63
64
 
64
65
  # Either timeout <= 5s, or process still running after 5s
66
+ # color is only supported on a real terminal (avoids yaspin UserWarning in Jupyter)
67
+ spinner_kwargs = {"color": "cyan"} if sys.stdout.isatty() else {}
65
68
  spinner = (
66
- yaspin(Spinners.point, text=f"Running {description}...", color="cyan")
69
+ yaspin(Spinners.point, text=f"Running {description}...", **spinner_kwargs)
67
70
  if show_spinner
68
71
  else None
69
72
  )
@@ -3,9 +3,9 @@ Template management for cluster deployments.
3
3
  """
4
4
 
5
5
  import filecmp
6
+ import logging
6
7
  import os
7
8
  import shutil
8
- import logging
9
9
 
10
10
  from cluster_builder.utils.hcl import extract_template_variables
11
11
 
@@ -90,7 +90,7 @@ class TemplateManager:
90
90
  try:
91
91
  shutil.copy2(src_path, dst_path)
92
92
  logger.debug(f"Copied provider template {template_file} to {dst_path}")
93
- except Exception as e:
93
+ except Exception as e: # noqa: BLE001 - wrap any copy failure into a RuntimeError
94
94
  error_msg = f"Failed to copy provider template {template_file}: {e}"
95
95
  logger.error(error_msg)
96
96
  raise RuntimeError(error_msg)
@@ -99,37 +99,6 @@ class TemplateManager:
99
99
  f"✅ Provider configurations ensured in cluster directory {cluster_dir}"
100
100
  )
101
101
 
102
- def copy_user_data_template(self, role: str, cloud: str) -> None:
103
- """
104
- Copy the user data template for a specific role to the cloud provider directory.
105
-
106
- Args:
107
- role: K3s role (master, worker, etc.)
108
- cloud: Cloud provider name
109
-
110
- Raises:
111
- RuntimeError: If the template file doesn't exist or can't be copied
112
- """
113
- user_data_src = os.path.join(self.templates_dir, f"{role}_user_data.sh.tpl")
114
- user_data_dst = os.path.join(
115
- self.templates_dir, cloud, f"{role}_user_data.sh.tpl"
116
- )
117
-
118
- if not os.path.exists(user_data_src):
119
- error_msg = f"User data template not found: {user_data_src}"
120
- logger.error(error_msg)
121
- raise RuntimeError(error_msg)
122
-
123
- try:
124
- shutil.copy2(user_data_src, user_data_dst)
125
- logger.debug(
126
- f"Copied user data template from {user_data_src} to {user_data_dst}"
127
- )
128
- except (OSError, shutil.Error) as e:
129
- error_msg = f"Failed to copy user data template: {e}"
130
- logger.error(error_msg)
131
- raise RuntimeError(error_msg)
132
-
133
102
  def get_required_variables(self, cloud: str) -> dict:
134
103
  """
135
104
  Get the variables required for a specific cloud provider's templates.