Author: Matus Goljer [email protected]
Version: 0.0.1
This project is a fork of https://github.com/emacs-pe/docker-tramp.el which provides similar functionality for Docker containers.
nomad-tramp.el offers a TRAMP method for Docker containers
deployed on HashiCorp Nomad.
NOTE:
nomad-tramp.elrelies on thenomad execcommand and python3. Tested with nomad version 1.2.6+ but should work with any nomad version which supports exec.
Offers the TRAMP method nomad to access running containers
C-x C-f /nomad:[email protected]%node-name:/path/to/file
where
| Part | Description |
|---|---|
| task@ | The task name (optional). Default is the first task of the task group. |
| job | The job name. |
| task-group | The task group name. |
| alloc-index | Allocation index. If count = 1, it is always 0 |
| %node-name | Name of the node where the allocation runs (optional) |
Busyboxes built with the ENABLE_FEATURE_EDITING_ASK_TERMINAL config option
send also escape sequences, which tramp-wait-for-output doesn't ignores
correctly. Tramp upstream fixed in 98a5112 and is available since
Tramp>=2.3.
For older versions of Tramp you can dump docker-tramp-compat.el in your
load-path somewhere and add the following to your init.el, which
overwrites tramp-wait-for-output with the patch applied:
(require 'docker-tramp-compat)
This is a known issue with Tramp, but is not a bug so much as a poor default
setting. Adding tramp-own-remote-path to tramp-remote-path will make
Tramp use the remote's PATH environment varialbe.
(add-to-list 'tramp-remote-path 'tramp-own-remote-path)
Converted from nomad-tramp.el by el2markdown.