oxidized 0.31.0 → 0.32.0

checksums.yaml

@@ -1,7 +1,7 @@
 ---
 SHA256:
-  metadata.gz: 59670040bed3f1d863d073d2f87dfaa07f3918eef49c3a0f2bbf44b6b111c3cc
-  data.tar.gz: 7e1c5400732aaf05567e20c635bee4e9bd9d0d4254b1aec78f2cb69fb48c98a2
+  metadata.gz: b47cc2ddbb41778414a86ed387714b91340ff440f125d948b8168c6ab1641f71
+  data.tar.gz: 492d75d181d94cef898e5a687715d81e94dd9139ff6310e4c6b304edb92be39e
 SHA512:
-  metadata.gz: 60fad3a803759b63ee5519d0abb7f60eedd397eef68fa0728009b41553698b0d6b45dbd1c02de235ca9b929849b63b6caff3e751d5c24cdcf7039f1096b8727c
-  data.tar.gz: ea896202d8ba22fae8f64ba596b48fd31bdca016dee62652a1ca3feb27be0c2d07328472c4f8fa96f3b001ffacdc37078555f0c6439dbe1aca9ce85cc32afe3d
+  metadata.gz: c5028d2ae6b9ecb04783381d0804d68014f308e4109627b27d7a4c8225529ea06d3c300a46547fcfd47f1c961acab294cb7abae8200b0aa857a86af27c6f1269
+  data.tar.gz: cbeebeca48db2727432e42d069f3b94d326e150163944f27a21594dcaf73bc0eab7b91518f5d265c5a576e083b749ca733123fa438d247d15d9d2cf365b4a1dc

data/.github/workflows/ruby.yml

@@ -19,7 +19,8 @@ jobs:
     runs-on: ubuntu-latest
     strategy:
       matrix:
-        ruby-version: ['3.1', '3.2', '3.3']
+        ruby-version: ['3.1', '3.2', '3.3', '3.4', 'ruby-head']
+    continue-on-error: ${{ matrix.ruby-version == 'ruby-head' }}
 
     steps:
     - uses: actions/checkout@v4
@@ -38,5 +39,3 @@ jobs:
         reporter: github-pr-review
     - name: Run tests
       run: bundle exec rake
-    - uses: codecov/codecov-action@v3
-      if: ${{ always() }}

data/.rubocop.yml

@@ -1,9 +1,8 @@
 inherit_from: .rubocop_todo.yml
 
-# Do not attempt to police vendored code
 AllCops:
   NewCops: enable
-  TargetRubyVersion: 3.1
+  # Do not attempt to police vendored code
   Exclude:
     - 'vendor/**/*'
 

data/.rubocop_todo.yml

@@ -1,22 +1,22 @@
 # This configuration was generated by
 # `rubocop --auto-gen-config`
-# on 2024-10-15 06:30:40 UTC using RuboCop version 1.66.1.
+# on 2025-02-17 10:13:53 UTC using RuboCop version 1.72.1.
 # The point is for the user to remove these configuration records
 # one by one as the offenses are removed from the code base.
 # Note that changes in the inspected code, or installation of new
 # versions of RuboCop, may require this file to be generated again.
 
-# Offense count: 29
+# Offense count: 30
 # Configuration parameters: AllowedMethods, AllowedPatterns, CountRepeatedAttributes, Max.
 Metrics/AbcSize:
   Enabled: false
 
-# Offense count: 18
+# Offense count: 17
 # Configuration parameters: AllowedMethods, AllowedPatterns.
 Metrics/CyclomaticComplexity:
   Max: 12
 
-# Offense count: 15
+# Offense count: 14
 # Configuration parameters: AllowedMethods, AllowedPatterns.
 Metrics/PerceivedComplexity:
   Max: 12
@@ -72,12 +72,12 @@ Style/OpenStructUse:
     - 'lib/oxidized/node.rb'
     - 'spec/hook/githubrepo_spec.rb'
 
-# Offense count: 31
+# Offense count: 33
 # This cop supports unsafe autocorrection (--autocorrect-all).
 Style/SlicingWithRange:
   Enabled: false
 
-# Offense count: 85
+# Offense count: 94
 # This cop supports unsafe autocorrection (--autocorrect-all).
 # Configuration parameters: Mode.
 Style/StringConcatenation:

data/CHANGELOG.md

@@ -4,6 +4,37 @@ All notable changes to this project will be documented in this file.
 
 The format is based on [Keep a Changelog](https://keepachangelog.com/en/1.1.0/).
 
+## [0.32.0 – 2025-02-17]
+This release fixes a security issue in oxidized-web, which is included in the
+Docker container of oxidized. If you are not using the Docker container but
+use oxidized-web, be sure to update your oxidized-web gem to 0.15.0.
+
+### Added
+- junos: add unit test (@systeembeheerder)
+- apc_aos: support for scp (@robertcheramy)
+- config: allow model_map and group_map keys to be regexp. Fixes #3360 (@ytti)
+- enterprise_sonic: add new model enterprise_sonic (@ohai89)
+- model for Kornfeld Operating System (@yurenkov)
+
+### Changed
+- sonicos: accept policy message. Fixes #3339 (@Steve-M-C, @robertcheramy)
+- input/ssh: change input.debug to dump all characters and include sent commands. (@robertcheramy)
+- cumulus: remove ANSI Escape codes and fix prompt issues. The prompt is more specific now (@alchemyx, @robertcheramy)
+- model unit tests: the tests are automated and simpler to use (@ytti, @robertcheramy)
+- device2yaml.rb: moved to extra/, commands can be specified from the command line or from a file (no cmdsets provided anymore) (@robertcheramy)
+- extra/gitdiff-msteams.sh: honor the 28KB size limit and add an optional link to GitHub (@mopi3456)
+
+### Fixed
+- tplink: send 'enable' before the enable password. Fixes #3271 (@robertcheramy)
+- asyncos: fix prompt for hostnames containing "-" . Fixes #3327 (@robertcheramy)
+- sonicos: fix prompt for hostnames containing "-" . Fixes #3333 (@robertcheramy)
+- xos: Hide radius accounting secret
+- fsos: Hide AAA and SNMP secrets (@RayaneB35)
+- aos7: fix prompt for version 8.8x. Fixes #3351 (@robertcheramy)
+- aosw: Hide power measurements (@rouven0)
+- arubainstant: show version prepends a space to prompt when a core file is present. Fixes #3398 (@robertcheramy)
+
+
 ## [0.31.0 – 2024-11-29]
 
 ### Added
@@ -66,6 +97,7 @@ The format is based on [Keep a Changelog](https://keepachangelog.com/en/1.1.0/).
 - netgear: prompt for gs752tpp. Fixes #3287 (@robertcheramy)
 - aoscx: fixed regex for 6400 switches to hide temperature and power (@steveneppler)
 
+
 ## [0.30.1 – 2024-04-12]
 
 ### Added

data/Dockerfile

@@ -18,6 +18,7 @@ ENV GEM_HOME=/usr/local/bundle
 # Install the x25519 gem
 RUN gem install x25519 --no-document
 
+
 ###################
 # build oxidized
 COPY . /tmp/oxidized/
@@ -76,7 +77,7 @@ RUN apt-get -yq update \
     # Use ubuntu gems where possible
     # Gems needed by oxidized
     ruby-rugged ruby-slop ruby-psych \
-    ruby-net-telnet ruby-net-ssh ruby-net-ftp ruby-net-scp ruby-ed25519 \
+    ruby-net-telnet ruby-net-ssh ruby-net-ftp ruby-ed25519 \
     # Gem dependencies for inputs
     ruby-net-http-persistent ruby-mechanize \
     # Gem dependencies for sources
@@ -101,7 +102,9 @@ RUN gem install --no-document \
     # dependencies for hooks
     slack-ruby-client cisco_spark \
     # dependencies for specific inputs
-    net-tftp
+    net-tftp \
+    # Net scp is needed in Version >= 4.1.0, which is not available in ubuntu
+    net-scp
 
 # install oxidized from prebuilder
 # The Dockerfile ist version-independent, so use oxidized-*.gem to cach the gem

data/Rakefile

@@ -74,6 +74,7 @@ task :chmod do
     extra/oxidized.runit
     extra/syslog.rb
     extra/update-ca-certificates.runit
+    extra/device2yaml.rb
   ]
   dirs = []
   %x(git ls-files -z).split("\x0").reject { |f| f.match(/^(test|spec|features)\//) }.each do |file|
@@ -83,4 +84,31 @@ task :chmod do
   dirs.sort.uniq.each { |dir| File.chmod(0o0755, dir) }
 end
 
+# Build the container image with docker or podman
+def command_available?(command)
+  system("which #{command} > /dev/null 2>&1")
+end
+
+def docker_needs_root?
+  !system('docker info > /dev/null 2>&1')
+end
+
+desc 'Build the container image with docker or podman'
+task :build_container do
+  # Prefer podman if available as it runs rootless
+  if command_available?('podman')
+    sh 'podman build -t oxidized:`git describe --tags` -t oxidized:latest .'
+  elsif command_available?('docker')
+    if docker_needs_root?
+      puts 'docker needs root to build the image. Using sudo...'
+      sh 'sudo docker build -t oxidized:`git describe --tags` -t oxidized:latest .'
+    else
+      sh 'docker build -t oxidized:`git describe --tags` -t oxidized:latest .'
+    end
+  else
+    puts 'You need Podman or Docker to build the container image.'
+    exit 1
+  end
+end
+
 task default: :test

data/docs/Configuration.md

@@ -174,7 +174,16 @@ input:
 
 ## Advanced Configuration
 
-Below is an advanced example configuration. You will be able to (optionally) override options per device. The router.db format used is `hostname:model:username:password:enable_password`. Hostname and model will be the only required options, all others override the global configuration sections.
+Below is an advanced example configuration.
+
+You will be able to (optionally) override options per device.
+The router.db format used is `hostname:model:username:password:enable_password`.
+Hostname and model will be the only required options, all others override the
+global configuration sections.
+
+Custom model names can be mapped to an oxidized model name with a string or
+a regular expression.
+
 
 ```yaml
 ---
@@ -226,6 +235,7 @@ source:
 model_map:
   cisco: ios
   juniper: junos
+  !ruby/regexp /procurve/: procurve
 ```
 
 ## Advanced Group Configuration
@@ -268,7 +278,8 @@ groups:
           ssh_keys: "~/.ssh/id_rsa_bar_vyatta"
 ```
 
-For mapping multiple group values to a common name
+For mapping multiple group values to a common name, you can use strings and
+regular expressions:
 
 ```yaml
 group_map:
@@ -276,6 +287,7 @@ group_map:
   alias2: groupA
   alias3: groupB
   alias4: groupB
+  !ruby/regexp /specialgroup/: groupS
   aliasN: groupZ
   # ...
 ```

data/docs/Creating-Models.md

@@ -54,6 +54,52 @@ The API documentation contains a list of [methods](https://github.com/ytti/oxidi
 
 A more fleshed out example can be found in the `IOS` and `JunOS` models.
 
+### Common task: mechanism for handling 'enable' mode
+The following code snippet demonstrates how to handle sending the 'enable'
+command and an enable password.
+
+This example is taken from the `IOS` model. It covers scenarios where users
+need to enable privileged mode, either without providing a password (by setting
+`enable: true` in the configuration) or with a password.
+
+```ruby
+  cfg :telnet, :ssh do
+    post_login do
+      if vars(:enable) == true
+        cmd "enable"
+      elsif vars(:enable)
+        cmd "enable", /^[pP]assword:/
+        cmd vars(:enable)
+      end
+    end
+  end
+```
+Note: remove `:telnet, ` if your device does not support telnet.
+
+### Common Task: remove ANSI escape codes
+> :warning: This common task is experimental.
+> If it does not work for you, please open an issue so that we can adapt the
+> code snippet.
+
+Some devices produce ANSI escape codes to enhance the appearance of output.
+However, this can make prompt matching difficult and some of these ANSI escape
+codes might end up in the resulting configuration.
+
+You can remove most [ANSI escape codes](https://en.wikipedia.org/wiki/ANSI_escape_code#Control_Sequence_Introducer_commands) using the following Ruby
+code in your model:
+```
+  # Remove ANSI escape codes
+  expect /\e\[[0-?]*[ -\/]*[@-~]\r?/ do |data, re|
+    data.gsub re, ''
+  end
+```
+Explanation of the Regular Expression:
+- `\e\[`   : Control Sequence Introducer (CSI), which starts with "ESC [".
+- `[0-?]*` : "Parameter" bytes (range 0x30–0x3F, corresponding to ASCII `0–9:;<=>?`).
+- `[ -\/]*`: "Intermediate" bytes (range 0x20–0x2F, corresponding to ASCII ` !"#$%&'()*+,-./`).
+- `[@-~]`  : The "final" byte (range 0x40–0x7E, corresponding to ASCII ``@A–Z[\]^_`a–z{|}~).[``).
+- `\r?`    : Some ESC codes include a carriage return, which we do not want in the resulting config.
+
 ## Extending an existing model with a new command
 
 The example below can be used to extend the `JunOS` model to collect the output of `show interfaces diagnostics optics` and append the output to the configuration file as a comment. This command retrieves DOM information on pluggable optics present in a `JunOS`-powered chassis.
@@ -86,32 +132,16 @@ Intuitively, it is also possible to:
 * Create a completely new model, with a new name, for a new operating system type.
 * Testing/validation of an updated model from the [Oxidized GitHub repo models](https://github.com/ytti/oxidized/tree/master/lib/oxidized/model) by placing an updated model in the proper location without disrupting the gem-supplied model files.
 
-## Create unit tests for the model
-> :warning: model unit tests are still a work in progress and need some polishing.
-
-If you want the model to be integrated into oxidized, you can
-[submit a pull request on github](https://github.com/ytti/oxidized/pulls).
+## Create Unit Tests for the Model
+If you want the model to be integrated into Oxidized, you can
+[submit a pull request on GitHub](https://github.com/ytti/oxidized/pulls).
 This is a greatly appreciated submission, as there are probably other users
 using the same network device as you are.
 
 A good (and optional) practice for submissions is to provide a
-[unit test for your model](/spec/model). This reduces the risk that further
-developments could break it, and facilitates debugging issues without having
-access to a physical network device for the model.
-
-In order to simulate the device in the unit test, you need a
-[YAML simulation file](/examples/device-simulation/), have a look at the
-link for an explanation on how to create one.
-
-Creating the unit test itself is explained in
-[README.md in the model unit test directory](/spec/model/README.md).
-
-Remember - producing a YAML simulation file and/or writing a unit test is
-optional.
-The most value comes from the YAML simulation file. The unit
-test can be written by someone else, but you need access to the device for the
-YAML simulation file. If you encounter problems, open an issue or ask for help
-in your pull request.
+[unit test for your model](/docs/ModelUnitTests.md). This reduces the risk that
+further developments could break it, and facilitates debugging issues without
+having access to a physical network device for the model.
 
 ## Advanced features
 

data/docs/DeviceSimulation.md

@@ -0,0 +1,184 @@
+# Device Simulation
+Oxidized supports [150+ devices](/docs/Supported-OS-Types.md).
+
+No developer has access to all of these devices, which makes the task of
+maintaining Oxidized difficult:
+
+- Issues can't be resolved because the developer has no access to the device.
+- Further developments can produce regressions.
+
+In order to address this, we can simulate the devices. An example of a
+simulation is the [model unit tests](/spec/model), but one could also simulate a
+device within an SSH server.
+
+The simulation of devices is currently focused on SSH-based devices. This may be
+extended to other inputs like Telnet or FTP in the future.
+
+## YAML Simulation Data
+The underlying data for the simulation is a [YAML](https://yaml.org/) file in
+which we store all relevant information about the device. The most important
+information is the responses to the commands used in the Oxidized models.
+
+The YAML simulation files are stored under
+[/spec/model/data/](/spec/model/data/), with the naming convention
+`<model>:<description>:simulation.yaml`, where `<model>` is the lowercase name
+of the Oxidized model and `<description>` is the name of the test case.
+`<description>` is generally formatted as `<hardware>_<software>` or
+`<hardware>_<software>_<information>`.
+
+### Creating a YAML Simulation File with device2yaml.rb
+A device does not only output the ASCII text we can see in the console.
+It adds ANSI escape codes for nice colors, bold and underline, \r, and so on.
+These are key factors in prompt issues, so they must be represented in the YAML
+file. We use the Ruby string format with interpolations like \r, \e, and so on.
+Another important point is trailing spaces at the end of lines. Some text
+editors automatically remove trailing spaces, so we code them with \x20.
+
+Although a YAML file could be written by hand, this is quite a tedious task to
+catch all the extra codes and code them into YAML. This can be automated with
+the Ruby script [extra/device2yaml.rb](/extra/device2yaml.rb).
+
+`device2yaml.rb` needs Ruby and the gem
+[net-ssh](https://rubygems.org/gems/net-ssh/) to run. On Debian, you can install
+them with `sudo apt install ruby-net-ssh`.
+
+Run `extra/device2yaml.rb`, the online help tells you the options.
+```
+oxidized$ extra/device2yaml.rb
+Missing a host to connect to...
+
+Usages:
+- device2yaml.rb [user@]host -i file [options]
+- device2yaml.rb [user@]host -c "command1
+  command2
+  command3" [options]
+
+-i and -c are mutualy exclusive, one must be specified
+
+[options]:
+    -c, --commands "command list"    specify the commands to be run
+    -i, --input file                 Specify an input file for commands to be run
+    -o, --output file                Specify an output YAML-file
+    -t, --timeout value              Specify the idle timeout beween commands (default: 5 seconds)
+    -e, --exec-mode                  Run ssh in exec mode (without tty)
+    -h, --help                       Print this help
+```
+
+- `[user@]host` specifies the user and host to connect to the device. The
+password will be prompted interactively by the script. If you do not specify a
+user, it will use the user executing the script.
+- The commands that will be run on the device must be defined in
+`deviceyaml.rb`. You can give the commands online with `-c` or read them from a
+file (one line per command) with `-i`. The commands should match exactly the
+ones of the model (no abbreviations) and include the commands of the
+`post_login` and `pre_logout` sections. When using `-c` and editing the shell
+command line, `CTRL-V CTRL-J` is very useful to add a line break.
+- `device2yaml.rb` waits an idle timeout after the last received data
+before sending the next command. The default is 5 seconds. If your device makes
+a longer pause than 5 seconds before or within a command, you will see that the
+output of the command is shortened or slips into the next command in the YAML
+file. You will have to change the idle timeout to a greater value to address
+this.
+- When run without the output argument, `device2yaml.rb` will only print the SSH
+output to the standard output. You must use `-o <model:HW_SW:simulation.yaml>`
+to store the collected data in a YAML file.
+- If your Oxidized model uses SSH exec mode (look for `exec true` in the model),
+you will have to use the option `-e` to run `device2yaml.rb` in SSH exec mode.
+
+Note that `device2yaml.rb` takes some time to run because of the idle timeout of
+(default) 5 seconds between each command. You can press the "Escape" key if you
+know there is no more data to come for the current command (when you see the
+prompt for the next command), and the script will stop waiting and directly
+process the next command.
+
+
+Running the script against an ios device would look like:
+```shell
+extra/device2yaml.rb oxidized@r61 -c "terminal length 0
+terminal width 0
+show version
+show vtp status
+show inventory
+show running-config
+exit" -o spec/model/data/ios:C8200L_16.12.1:simulation.yaml
+```
+### Publishing the YAML Simulation File to Oxidized
+Publishing the YAML simulation file of your device helps maintain Oxidized. This
+task may take some time, and we are very grateful that you take this time for
+the community!
+
+You should pay attention to removing or replacing anything you don't want to
+share with the rest of the world, for example:
+
+- Passwords
+- IP Addresses
+- Serial numbers
+
+You can also shorten the configuration if you want - we don't need 48 times the
+same configuration for each interface, but it doesn't hurt either.
+
+Take your time, this is an important task: after you have uploaded your file on
+GitHub, it may be impossible to remove it.
+You can use search/replace to make consistent and faster changes, for example
+change the hostname everywhere.
+
+The YAML simulation files are stored under
+[/spec/model/data/](/spec/model/data/), with the naming convention
+`<model>:<description>:simulation.yaml`, where `<model>` is the lowercase name
+of the Oxidized model and `<description>` is the name of the test case.
+`<description>` is generally formatted as `<hardware>_<software>` or
+`<hardware>_<software>_<information>`.
+
+Using a correct name for the file is important to ensure it is included in
+automatic model unit tests.
+
+Examples:
+
+- spec/model/data/aoscx:R0X25A-6410_FL.10.10.1100:simulation.yaml
+- spec/model/data/asa:5512_9.12-4-67_single-context:simulation.yaml
+- spec/model/data/ios:C9200L-24P-4G_17.09.04a:simulation.yaml
+
+When you are finished, commit and push to your forked repository on GitHub, and
+submit a Pull Request. Thank you for your help!
+
+### Interactive Mode
+The `device2yaml.rb` script is basic and sometimes needs some help, especially
+when dealing with a device that sends its output page by page and requires you
+to press space for the next page. `device2yaml.rb` does not know how to handle
+this.
+
+While `device2yaml.rb` is running, you can type anything on the keyboard, and it
+will be sent to the remote device. So you can press space or 'n' to get the next
+page.
+
+You can also use this to enter an enable password.
+
+If you press the "Esc" key, `device2yaml.rb` will not wait for the idle timeout
+and will process the next command right away.
+
+### YAML Format
+The YAML file has two sections:
+- init_prompt: describing the lines sent by the device before we can send a
+command. It usually includes MOTD banners and must include the first prompt.
+- commands: the commands the Oxidized model sends to the network device and
+their outputs.
+
+The outputs are multiline and use YAML block scalars (`|`), with the trailing \n
+removed (`-` after `|`). The outputs include the echo of the given command and
+the next prompt. Escape characters are coded in Ruby style (\n, \r...).
+
+Here is a shortened example of a YAML file:
+```yaml
+---
+init_prompt: |-
+  \e[4m\rLAB-R1234_Garderos#\e[m\x20
+commands:
+  show system version: |-
+    show system version
+    grs-gwuz-armel/003_005_068 (Garderos; 2021-04-30 16:19:35)
+    \e[4m\rLAB-R1234_Garderos#\e[m\x20
+# ...
+  exit: ""
+```
+
+

data/docs/Hooks.md

@@ -259,13 +259,15 @@ gem install slack-ruby-client
 
 ### slackdiff hook configuration example
 
+> Please note that the channel needs to be your Slack channel ID.
+
 ```yaml
 hooks:
   slack:
     type: slackdiff
     events: [post_store]
     token: SLACK_BOT_TOKEN
-    channel: "#network-changes"
+    channel: "CHANNEL_ID"
 ```
 
 The token parameter is a Slack API token that can be generated following [this tutorial](https://api.slack.com/tutorials/tracks/getting-a-token).  Until Slack stops supporting them, legacy tokens can also be used.
@@ -278,13 +280,11 @@ hooks:
     type: slackdiff
     events: [post_store]
     token: SLACK_BOT_TOKEN
-    channel: "#network-changes"
+    channel: "CHANNEL_ID"
     diff: false
     message: "%{node} %{group} %{model} updated https://git.intranet/network-changes/commit/%{commitref}"
 ```
 
-Note the channel name must be in quotes.
-
 A proxy can optionally be specified if needed to reach the Slack API endpoint.
 
 ```yaml
@@ -293,7 +293,7 @@ hooks:
     type: slackdiff
     events: [post_store]
     token: SLACK_BOT_TOKEN
-    channel: "#network-changes"
+    channel: "#CHANNEL_ID"
     proxy: http://myproxy:8080
 ```
 

data/docs/Issues.md

@@ -49,7 +49,7 @@ contributing code via a pull request (PR) or hiring a developer.
 
 ## Sumbit a YAML Simulation File
 To help developers troubleshoot device-specific issues, you may be asked to submit a
-[YAML simulation file](https://github.com/ytti/oxidized/blob/master/examples/device-simulation/README.md#creating-a-yaml-file-with-device2yamlrb) for your device.
+[YAML simulation file](/docs/DeviceSimulation.md#creating-a-yaml-file-with-device2yamlrb) for your device.
 
 Here's a brief overview how to do it, you can find more details in the link
 above.
@@ -63,22 +63,28 @@ sudo apt install git ruby-net-ssh
 ```
 git clone git@github.com:<your github user>/oxidized.git
 ```
-- run the device2yaml.rb script (you’ll be provided with the command set and
-  output filename to use)
+- run the `extra/device2yaml.rb` script (you’ll be provided with the command to
+run) from the repository root:
+
 ```
-cd oxidized/examples/device-simulation
-# Replace user and devicename to appropriate values
-./device2yaml.rb user@devicename -c cmdsets/ios -o yaml/asr900_26.8.1b.yaml
+extra/device2yaml.rb oxidized@r61 -c "terminal length 0
+terminal width 0
+show version
+show vtp status
+show inventory
+show running-config
+exit" -o spec/model/data/ios:C8200L_16.12.1:simulation.yaml
 ```
+
 - The script waits 5 seconds between commands, and outputs the response of the
   device. You can press "ESC" if you see the prompt and want to pass to next
   command without waiting for the timeout.
-- The result will be stored in `oxidized/examples/device-simulation/yaml/`.
+- The result will be stored in `spec/model/data/`.
 - Replace any sensitive information with placeholder values in the output file.
 - Commit & push the file to github
 ```
-git add yaml/asr900_26.8.1b.yaml
-git commit -m "Device simulation for ASR900"
+git add spec/model/data/ios:C8200L_16.12.1:simulation.yaml
+git commit -m "Device simulation for C8200L"
 git push
 ```
 - Create a pull request (PR) in GitHub, referencing the issue number (e.g.,

data/docs/Model-Notes/APC_AOS.md

@@ -1,29 +1,42 @@
 # APC AOS Configuration
 
-Currently, the configuration of APC Network Management Cards can be downloaded with FTP only.
+The configuration of APC Network Management Cards can be downloaded using FTP
+and SCP.
+
+To download with SCP, you need a
+[patch](https://github.com/net-ssh/net-scp/pull/71) to
+[Net::SCP](https://github.com/net-ssh/net-scp, which has been included
+upstream, but there is currently no new release of Net::SCP and its authors are
+unresponsive.
+
+To temporarily solve this,
+[@robertcheramy forked Net::SCP](https://github.com/robertcheramy/net-scp). You
+can build or download the gem there. This gem is already included in the
+oxidized container image (in the release coming after 0.31.0).
 
-A download of the configuration with SCP is [work in progress](https://github.com/ytti/oxidized/issues/1802).
-As the APC has an unusual behavior (the connection is closed without an exit-status), this has to be
-[fixed](https://github.com/net-ssh/net-scp/pull/71) upstream in [Net::SCP](https://github.com/net-ssh/net-scp).
-As soon as there is a release of Net::SCP supporting the behavior of APC OS, we will activate SCP in oxidized.
 
 ## Can I collect more information than just the configuration?
-APC OS does not have the ability to show the config.ini within an SSH-session. As oxidized can only get the
-configuration with one input type at a time, it is not possible to fetch config.ini via FTP/SCP and get the output of
-some commands via SSH at the same time.
+APC OS does not have the ability to show the config.ini within an SSH-session.
+As oxidized can only get the configuration with one input type at a time, it is
+not possible to fetch config.ini via FTP/SCP and get the output of
+some commands via SSH at the same time. Feature request #3334 has been opened
+to support multiple inputs in oxidized.
+
+A ticket has been opened with APC support in order to enable "cat config.ini"
+within an SSH-session, but APC is not willing to support this.
 
-A ticket has been opened with APC support in order to support "cat config.ini" within an SSH-session, but
-the chances it will be supported at some time are not very good, and older versions will still not support it.
 
-## How do I activate FTP input?
-In order to download the configuration with FTP (and in the future with SCP), you have to activate it as an
-input in the oxidized configuration. If you do not activate the input, oxidized will fail for the node with
-a rather unspecific error (`WARN -- : /apc status fail, retry attempt 1`).
+## How do I activate FTP/SCP input?
+In order to download the configuration with FTP or SCP, you have to activate it
+as an input in the oxidized configuration. If you do not activate the input,
+oxidized will fail for the node with a
+[rather unspecific error](https://github.com/ytti/oxidized/issues/3346)
+(`WARN -- : /apc status fail, retry attempt 1`).
 
 The configuration can be done either globally or only for the model apc_aos.
 
-The global configuration would look like this. Note that Oxidized will try every input type in the given order
-until it succeeds, or it will report a failure.
+The global configuration would look like this. Note that Oxidized will try every
+input type in the given order until it succeeds, or it will report a failure.
 ```yaml
 input:
   default: ssh, ftp, scp