1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178 179 180 181 182 183 184 185 186 187 188 189 190 191 192 193 194 195 196 197 198 199 200 201 202 203 204 205 206 207 208 209 210 211 212 213 214 215 216 217 218 219 220 221 222 223 224 225 226 227 228 229 230 231 232 233 234 235 236 237 238 239 240 241 242 243 244 245 246 247 248 249 250 251 252 253 254 255 256 257 258 259 260 261 262 263 264 265 266 267 268 269 270 271 272 273 274 275 276 277 278 279 280 281 282 283 284 285 286 287 288 289 290 291 292 293 294 295 296 297 298 299 300 301 302 303 304 305 306 307 308 309 310 311 312 313 314 315 316 317 318 319 320 321 322 323 324 325 326 327 328
|
# API
While most users will interact with BundleWrap through the `bw` command line utility, you can also use it from your own code to extract data or further automate config management tasks.
Even within BundleWrap itself (e.g. templates, libs, and hooks) you are often given repo and/or node objects to work with. Their methods and attributes are documented below.
Some general notes on using BundleWrap's API:
* There can be an arbitrary amount of `bundlewrap.repo.Repository` objects per process.
* Repositories are read as needed and not re-read when something changes. Modifying files in a repo during the lifetime of the matching Repository object may result in undefined behavior.
<br>
## Example
Here's a short example of how to use BundleWrap to get the uptime for a node.
from bundlewrap.repo import Repository
repo = Repository("/path/to/my/repo")
node = repo.get_node("mynode")
uptime = node.run("uptime")
print(uptime.stdout)
<br>
## Reference
### bundlewrap.repo.Repository(path)
The starting point of any interaction with BundleWrap. An object of this class represents the repository at the given path. `path` can be a subpath of your repository (e.g., `bundles/nginx/`) and will internally be resolved to the root path of said repository.
<br>
**`.branch`**
The current git branch of this repo. `None` if not in a git repo.
<br>
**`.clean`**
Boolean indicating if there are uncommitted changes in git. `None` if not in a git repo.
<br>
**`.groups`**
A list of all groups in the repo (instances of `bundlewrap.group.Group`)
<br>
**`.nodes`**
A list of all nodes in the repo (instances of `bundlewrap.node.Node`)
<br>
**`.revision`**
The current git, hg or bzr revision of this repo. `None` if no SCM was detected.
<br>
**`.get_group(group_name)`**
Returns the Group object for the given name.
<br>
**`.get_node(node_name)`**
Returns the Node object with the given name.
<br>
**`.nodes_in_all_groups(group_names)`**
Returns a list of Node objects where every node is a member of every group name given.
<br>
**`.nodes_in_any_group(group_names)`**
Returns all Node objects that are a member of at least one of the given group names.
<br>
**`.nodes_in_group(group_name)`**
Returns a list of Node objects in the named group.
<br>
**`.nodes_not_in_group(group_name)`**
Returns a list of Node objects not in the named group.
<br>
**`.nodes_with_bundle(bundle_name)`**
Returns a list of Node objects that do have the named bundle.
<br>
**`.nodes_without_bundle(bundle_name)`**
Returns a list of Node objects that do not have the named bundle.
<br>
**`.nodes_matching_lambda(self, lambda_str, lambda_workers=None)`**
Returns a list of Node objects matching the lambda.
Example:
```python
nodes = repo.nodes_matching_lambda("node.metadata.get('foo/magic', 47) < 3")
```
- `lambda_str` is evaluated as python code with `node` being one of the nodes and expected to return a boolean.
- `lambda_workers` is number of parallel workers used to check lambda condition on every node
<br>
**`.nodes_matching(self, target_strings, lambda_workers=None)`**
Returns a list of nodes matching any of the given target-strings. This is the same API that is used by
all the bw commandlines, i.e. `bw items` or `bw apply` to select which nodes to operate on.
Example:
```python
nodes = repo.nodes_matching("lambda:node.metadata.get('foo/magic', 47) < 3")
nodes = repo.nodes_matching("loc.routers")
nodes = repo.nodes_matching("loc.router-1")
```
- `target_strings` is a list of expression to select target nodes
- `lambda_workers` is number of parallel workers used to check lambda condition on every node
The following expressions can be used:
- `my_node` to select a single node
- `my_group` all nodes in this group
- `bundle:my_bundle` all nodes with this bundle
- `!bundle:my_bundle` all nodes without this bundle
` `!group:my_group` all nodes not in this group
- `"lambda:node.metadata_get('foo/magic', 47) < 3"` all nodes whose `metadata["foo"]["magic"]` is less than three
<br>
### bundlewrap.node.Node()
A system managed by BundleWrap.
<br>
**`.bundles`**
A list of all bundles associated with this node (instances of `bundlewrap.bundle.Bundle`)
<br>
**`.groups`**
A list of `bundlewrap.group.Group` objects this node belongs to
<br>
**`.hostname`**
The DNS name BundleWrap uses to connect to this node
<br>
**`.items`**
A list of items on this node (instances of subclasses of `bundlewrap.items.Item`)
<br>
**`.magic_number`**
A large number derived from the node's name. This number is very likely to be unique for your entire repository. You can, for example, use this number to easily "jitter" cronjobs:
'{} {} * * * root /my/script'.format(
node.magic_number % 60,
node.magic_number % 2 + 4,
)
<br>
**`.metadata`**
A dictionary of custom metadata, merged from information in [nodes.py](../repo/nodes.py.md) and [groups.py](../repo/groups.py.md)
<br>
**`.name`**
The internal identifier for this node
<br>
**`.download(remote_path, local_path)`**
Downloads a file from the node.
- `remote_path` Which file to get from the node
- `local_path` Where to put the file
<br>
**`.get_item(item_id)`**
Get the Item object with the given ID (e.g. "file:/etc/motd").
<br>
**`.has_bundle(bundle_name)`**
`True` if the node has a bundle with the given name.
<br>
**`.has_any_bundle(bundle_names)`**
`True` if the node has a bundle with any of the given names.
<br>
**`.in_group(group_name)`**
`True` if the node is in a group with the given name.
<br>
**`.in_any_group(group_names)`**
`True` if the node is in a group with any of the given names.
<br>
**`.run(command, may_fail=False)`**
Runs a command on the node. Returns an instance of `bundlewrap.operations.RunResult`.
- `command` What should be executed on the node
- `may_fail` If `False`, `bundlewrap.exceptions.RemoteException` will be raised if the command does not return 0.
`bundlewrap.exceptions.TransportException` will be raised if there was a transport error while running the command, e.g. the SSH process died unexpectedly.
<br>
**`.run_ipmitool(command)`**
Runs a command on an ipmi interface. Returns an instance of
`bundlewrap.operations.RunResult`. A command is a string of tokens which
gets passed to ipmitool. The command will be split at whitespace
characters and added to the commandline:
# runs `ipmitool -H hostname -U username -P password power status`
node.run_ipmitool('power status')
<br>
**`.upload(local_path, remote_path, mode=None, owner="", group="")`**
Uploads a file to the node.
- `local_path` Which file to upload
- `remote_path` Where to put the file on the target node
- `mode` File mode, e.g. "0644"
- `owner` Username of the file owner
- `group` Group name of the file group
<br>
### bundlewrap.group.Group
A user-defined group of nodes.
<br>
**`.name`**
The name of this group
<br>
**`.nodes`**
A list of all nodes in this group (instances of `bundlewrap.node.Node`, includes subgroup members)
<br>
### bundlewrap.utils.Fault
A Fault acts as a lazy stand-in object for the result of a given callback function. These objects are returned from the "vault" attached to `Repository` objects:
>>> repo.vault.password_for("demo")
<bundlewrap.utils.Fault object at 0x10782b208>
Only when the `value` property of a Fault is accessed or when the Fault is converted to a string, the callback function is executed. In the example above, this means that the password is only generated when it is really required (e.g. when used in a template). This is particularly useful when used in metadata in connection with [secrets](secrets.md). Users will be able to generate metadata with Faults in it, even if they lack the required keys for the decryption operation represented by the Fault. The key will only be required for files etc. that actually use it. If a Fault cannot be resolved (e.g. for lack of the required key), BundleWrap can just skip the item using the Fault, while still allowing other items on the same node to be applied.
Faults also support some rudimentary string operations such as appending a string or another Fault, as well as some string methods:
>>> f = repo.vault.password_for("1") + ":" + repo.vault.password_for("2")
>>> f
<bundlewrap.utils.Fault object at 0x10782b208>
>>> f.value
'VOd5PC:JUgYUb'
>>> f += " "
>>> f.value
'VOd5PC:JUgYUb '
>>> f.strip().value
'VOd5PC:JUgYUb'
>>> repo.vault.password_for("1").format_into("Password: {}").value
'Password: VOd5PC'
>>> repo.vault.password_for("1").b64encode().value
'Vk9kNVA='
>>> repo.vault.password_for("1").as_htpasswd_entry("username").value
'username:$apr1$8be694c7…'
These string methods are supported on Faults: `format`, `lower`, `lstrip`, `replace`, `rstrip`, `strip`, `upper`, `zfill`
|