This commit adds a subcommand to both the client and server to allow
changing the name of a peer. The peer retains all the same attributes as
before (public keys, IPs, admin/disabled status, etc.).
Closes#87
This subcommand takes a shell as an argument and generates shell
completions for that shell to stdout.
example:
```
$ innernet completions bash
OR
$ innernet-server completions bash
```
This commit adds a `delete-cidr` to both the client and server. It walks
through the prompts just like adding a CIDR.
Only eligible CIDRs are presented to the user. Eligibilty requires:
- CIDR has no child CIDRs
- CIDR has no assigned peers
Closes#23
Based on the conversation from #5 (comment) - this changes innernet's behavior on Linux from automatically falling back to the userspace, instead requiring --backend userspace to be specified.
This should help people avoid weird situations in environments like Docker.
change private keys on client earlier to avoid race conditions,
and attempt the fetch call multiple times to avoid spurious issues,
while also not failing the entire command if fetch doesn't succeed.
Scripts that demonstrate building a network of docker containers, doubling as an integration test for innernet.
Includes a number of improvements to the recent non-interactive CLI changes as well.
This is a stop-gap CSRF protection mechanism from unsophisticated attacks. It's to be considered a temporary solution until a more complete one can be implemented, but it should be sufficient in most cases for the time being.
See https://github.com/tonarino/innernet/issues/38 for further discussion.
Introduces `--hosts-path [PATH]` and `--no-write-hosts` options in `innernet`.
This could be further improved to have a persistent setting in a config file i.e. /etc/innernet.conf (which doesn't currently exist).
Fixes#6
Jake McGinty