exoscale / cli

Command-line tool for everything at Exoscale: compute, storage, dns.

Exoscale CLI

Manage your Exoscale infrastructure easily from the command-line with exo.

Installation

Follow the steps for your platform on our community docs.

Configuration

Running the exo config command will guide you through the initial configuration.

You can create and find API credentials in the IAM section of the Exoscale Console.

The configuration file and all assets created during exo operations will be saved in the following location:

OS	Location
GNU/Linux, *BSD	$HOME/.config/exoscale/
macOS	$HOME/Library/Application Support/exoscale/
Windows	%USERPROFILE%\.exoscale\

The configuration parameters are then saved in a exoscale.toml file with the following minimum format:

defaultaccount = "account_name"

[[accounts]]
  key = "API_KEY"
  name = "account_name"
  secret = "API_SECRET"

The current configuration and configuration file path can be shown with exo config show.

Usage

The exo CLI contains documentation for all of its commands, you can explore them by running exo help. Additional information and tutorials are available on Exoscale's community website.

Integrations

Fig

When using Fig you can run this command to output Fig completion spec:

exo integrations generate-fig-spec

External contributions

• setup-exoscale GitHub action

cli's Issues

defaultSSHKey shouldn't be written to the config

I went throught the config using v1.0.3 and there is a defaultSSHKey = "" in the config now? Like the sosEndpoint it shouldn't be there unless needed.

config: not written due to endpoint failure

Issue by marcaurele
Thursday Sep 06, 2018 at 05:41 GMT
Originally opened as exoscale/egoscale#318

---

When the remote endpoint is not available, it's impossible to add an entry and it crashed the whole import.

1. write each configuration entry in the file as soon as possible
2. don't exit when the endpoint does not reply

[+] Account name [blabla]: 
Get http://localhost:8080/client/api?apikey=blablablablablabl&command=listZones&page=1&pagesize=50&response=json&signature=blablablablabl%3D: dial tcp [::1]:8080: connect: connection refused
<program exits & no config written>

"exo vm template register" usage help

exo vm template register called without any argument tries to register a template – and obviously fails:

$ exo vm template register
Registering the template. failure!
RegisterCustomTemplate.Checksum (string) field is required, got ""

It should print and usage error message and the command usage help.

exo ssh error if instance name matchs multiples instances

exo vm create test
exo vm create test2

exo ssh test
more than one element found: apikey=xxxx, command=listVirtualMachines, expires=2018-12-11T17%3A02%3A33%2B0100, name=test, response=json, signatureversion=3

Add recursive upload capability for `sos upload`

It would be great if sos upload could take an optional flag such as --recursive or -R for uploading directories!

Thanks again for the great work!

Wrong request signature on preprod

$ exo -A preprod zone
API error Unauthorized 401 (CSErrorCode(0) 0): unable to verify user credentials and/or request signature

why create a new SSH key

Issue by sebgoa
Tuesday Sep 04, 2018 at 11:03 GMT
Originally opened as exoscale/egoscale#311

---

I like the simplicity of exo vm create foobar

but it creates a new SSH key:

$ exo vm create barbaz
Creating private SSH key
Deploying "barbaz" 100 % [======================================] done

What to do now?

1. Connect to the machine

> exo ssh barbaz
ssh -i "/Users/sebastiengoasguen/.exoscale/instances/2c5b53cb-c469-4209-8970-18f6ac913b88/id_rsa" [email protected]

2. Put the SSH configuration into ".ssh/config"

> exo ssh barbaz --info
Host barbaz
	HostName 185.19.30.86
	User ubuntu
	IdentityFile /Users/sebastiengoasguen/.exoscale/instances/2c5b53cb-c469-4209-8970-18f6ac913b88/id_rsa

Tip of the day:
	You're the sole owner of the priva

Why not use the default SSH key - if there is one-

Add CORS support to `exo sos`

Great job implementing the SOS interface!

It might be useful to consider adding support to update CORS headers with the cli. The following link provides a helpful overview:

https://docs.aws.amazon.com/AmazonS3/latest/dev/cors.html

The s3cmd tool provides setcors and delcors commands that provide most of the functionality one might need for typical use cases.

Thanks again!

write tests

the cli has close to none... sadly.

Allow VM instance security group modification with exo CLI

Currently it is not possible to modify a VM instance's security groups using the exo CLI – even when the VM is stopped.

I see 2 different ways to implement this (or maybe even both):

# In the "vm" command group:
exo vm firewall add VM SG1 [SG2 SGn...]
exo vm firewall remove VM SG1 [SG2 SGn...]

# In the "firewall" command group:
exo firewall attach SG VM1 [VM2 VMn...]
exo firewall detach SG VM1 [VM2 VMn...]

Any opinion?

exo add doc.go

Issue by greut
Wednesday Jul 04, 2018 at 07:03 GMT
Originally opened as exoscale/egoscale#213

---

https://godoc.org/github.com/exoscale/egoscale/cmd/exo

Package name is too generic

The deb package name is called cli which is too generic and not informative. Something like exocli would be more meaningful.

exoscale.toml default mode is too open

Hi.

After running exo config the file and dir are readable by anyone.
Considering that it contains secrets, that's not ideal ;-)

$ umask                                                                                                                                                                        
022
$ exo version
exo v1.0.0 (egoscale v0.11.6)
$ ls -ald /home/ajacoutot/.exoscale{,/exoscale.toml} 
drwxr-xr-x  2 ajacoutot  ajacoutot   512B Oct  6 17:47 /home/ajacoutot/.exoscale/
-rw-r--r--  1 ajacoutot  ajacoutot   411B Oct  6 17:47 /home/ajacoutot/.exoscale/exoscale.toml

Error while launching a one node kube

Hello, I have the following error when creating a one node kube cluster. Anything I've missed ?

$ exo lab kube create kube
Creating private SSH key
Deploying "kube" ⠼                                                                 done!
Bootstrapping Kubernetes cluster (can take up to several minutes):
Instance system upgrade...
Instance system upgrade: failed
+ sudo -E DEBIAN_FRONTEND=noninteractive apt-get update
+ sudo -E DEBIAN_FRONTEND=noninteractive apt-get upgrade -y
E: Sub-process /usr/bin/dpkg returned an error code (1)
+ sudo -E DEBIAN_FRONTEND=noninteractive apt-get update
+ sudo -E DEBIAN_FRONTEND=noninteractive apt-get upgrade -y
E: Sub-process /usr/bin/dpkg returned an error code (1)
+ sudo -E DEBIAN_FRONTEND=noninteractive apt-get update
+ sudo -E DEBIAN_FRONTEND=noninteractive apt-get upgrade -y
E: Sub-process /usr/bin/dpkg returned an error code (1)
+ sudo -E DEBIAN_FRONTEND=noninteractive apt-get update
+ sudo -E DEBIAN_FRONTEND=noninteractive apt-get upgrade -y
E: Sub-process /usr/bin/dpkg returned an error code (1)
+ sudo -E DEBIAN_FRONTEND=noninteractive apt-get update
+ sudo -E DEBIAN_FRONTEND=noninteractive apt-get upgrade -y
E: Sub-process /usr/bin/dpkg returned an error code (1)
+ sudo -E DEBIAN_FRONTEND=noninteractive apt-get update
+ sudo -E DEBIAN_FRONTEND=noninteractive apt-get upgrade -y
E: Sub-process /usr/bin/dpkg returned an error code (1)

cluster bootstrap failed: Process exited with status 100

`exo version` does not display version if no config file is present

exo version does not display version if no config file is present

• current result:
  exo version Config File "exoscale" Not Found in "[/Users/acr/.exoscale /Users/acr /Users/acr/github/demo/cli]"

• expected result (even if config not set/correct/present):
  exo version exo dev n/a (egoscale 0.12.2)

exo dns add A should show the usage

Issue by greut
Friday Aug 03, 2018 at 13:01 GMT
Originally opened as exoscale/egoscale#265

---

% exo dns add A
required flag(s) "address" not set

since it requires a domain name shouldn't it display the usage at this stage?

exo dns delete has no confirmation

deleting stuff should have confirmations.

Register template

I was play around with the openbsd template.
I was able to successfully upload the template:

$exo vm template add --url http://redacted/exoscale-openbsd-6.5-2019-04-29-ac2a63.qcow2 --checksum ac2a63055f323f4073fc1b4bc726fed1 --description openbsd --name "OpenBSD 6.5 64-bit" --username root
Registering the template.................................... success.
┼──────────┼────────────────────┼──────────────────────────────────────┼
│   ZONE   │        NAME        │                  ID                  │
┼──────────┼────────────────────┼──────────────────────────────────────┼
│ at-vie-1 │ OpenBSD 6.5 64-bit │ f1129815-8b61-49c1-a9a8-6376e1b5fb59 │
┼──────────┼────────────────────┼──────────────────────────────────────┼

but when I try to list --mine I got:

$ exo vm template list --mine
Malformed Windows/OpenBSD. "openbsd"

On credential storage

Issue by Fnux
Friday Sep 21, 2018 at 06:48 GMT
Originally opened as exoscale/egoscale#330

---

Hello,

The exo command line utility seems very convenient, but the way it stores credentials bothers me: it doesn't make sense to use 2FA to authenticate against portal.exoscale.com while a plaintext file containing API credentials working for any services sits on my disk. Is there a way - or is it planned - to sources these credentials from an external command ? I'd like to source the secret from my password store.

Thank you!

Sometimes, custom templates are not visible

I have 2 custom templates. One is named Linux Debian mcorbin 80 64-bit, the other Linux Debian mcorbin 50 64-bit.

cs listTemplates templatefilter="self" returns both templates, but not the CLI:

$ exo -A prod vm template list --mine
┼────────────────────────────────┼──────┼──────────────┼──────────────────────────────────────┼──────────┼
│        OPERATING SYSTEM        │ DISK │ RELEASE DATE │                  ID                  │ CATEGORY │
┼────────────────────────────────┼──────┼──────────────┼──────────────────────────────────────┼──────────┼
│ Linux Debian mcorbin 80 64-bit │      │ 2019-05-27   │ 80056bac-d7cb-4703-a48a-f75cc3d4c1ac │ self     │
┼────────────────────────────────┼──────┼──────────────┼──────────────────────────────────────┼──────────┼

[CH3649]

kube create doesn't handle ^C

It's not possible to interrupt the installation while it's running...

support CAA records

https://support.dnsimple.com/articles/caa-record/

Refacto some CMDs more async

as in this issue #99 and this PRs #101, #119 make the same refactoring for:

use asyncTasks() to make multiple query at the same time

• exo vm start
• exo vm stop
• exo vm reboot
• exo vm reset
• exo vm resize
• exo vm scale

privnet allow cidrmask

allow a more convenient way of specifying the netmask

• --netmask 255.255.255.0
• --cidrmask 24

exo status command

Read and display that page:

https://exoscalestatus.com/api.json

api shouldn't allows raw args

Allowing any args in the API gives the false impression that they do something.

% exo api vm list keyword=bar 

2018/11/28 09:17:18 GET /compute?apikey=EXO9d45a5d9bf68d5689a0eeb55&command=listVirtualMachines&expires=2018-11-28T09%3A27%3A18%2B0100&response=json&signature=...&signatureversion=3 HTTP/1.1

% exo api vm list --keyword=bar

2018/11/28 09:18:14 GET /compute?apikey=EXO9d45a5d9bf68d5689a0eeb55&command=listVirtualMachines&expires=2018-11-28T09%3A28%3A14%2B0100&keyword=bar&response=json&signature=...signatureversion=3 HTTP/1.1

Panic When using Env credentials

Env credentials example:

EXOSCALE_SECRET_KEY=XXXXXXXX...
EXOSCALE_API_KEY=EXOXXXX...
EXOSCALE_COMPUTE_ENDPOINT=https://api.exoscale.ch/compute

Panic on exo vm create test1 because gCurrentAccount is nil

When I verify if it's nil to continue there is an other issue on zone
more than one element found: apikey=EXO5145bd25f0ce3d5cc7ab087a, command=listZones

other issue exo config ask to create new one like with no config file

getMyCIDR(false) returns an IPv6 CIDR if IPv6 connectivity available

The cmd/firewall.go:getMyCIDR() function returns an IPv6 even if called with isIpv6 argument set to false. This is because the net.Dial("udp", ...) call uses OS TCP/IP stack priority which is IPv6 before IPv4, so if the host on which the code runs happens to have IPv6 working the OpenDNS resolution will actually happen over IPv6 and therefore return an IPv6 address instead of IPv4.

A way to fix this would be to call net.Dial() with udp4 instead of udp when isIpv6=false, and udp6 when isIpv6=true. Example for the following code:

package main

import (
	"context"
	"fmt"
	"net"

	"github.com/exoscale/egoscale"
)

func main() {
	fmt.Println(getMyCIDR(false))
}

func getMyCIDR(isIpv6 bool) (*egoscale.CIDR, error) {
	cidrMask := 32
	dnsServer := "resolver1.opendns.com"

	if isIpv6 {
		dnsServer = "resolver2.ipv6-sandbox.opendns.com"
		cidrMask = 128
	}

	resolver := net.Resolver{
		Dial: func(ctx context.Context, network, address string) (net.Conn, error) {
			return net.Dial("udp", dnsServer+":53")
		},
		PreferGo: true,
	}

	ips, err := resolver.LookupIPAddr(context.Background(), "myip.opendns.com")
	if err != nil {
		return nil, err
	}

	if len(ips) < 1 {
		return nil, fmt.Errorf("no IP addresses were found using OpenDNS")
	}

	return egoscale.ParseCIDR(fmt.Sprintf("%s/%d", ips[0].IP, cidrMask))
}

Returns 2a01:cb04::/32 <nil>

By calling net.Dial("udp4", ...), I get 86.247.71.58/32 <nil>.

If you're OK with the suggested fix I can submit a PR.

The cli crashes when adding a DNS entry

The cli crashes when adding a DNS entry when I execute this command:
$ exo dns add A mydomain.com -a ${DEPLOYMENT_SERVER_IP} -n ${MACHINE_NAME}

I get the following error:

panic: runtime error: invalid memory address or nil pointer dereference
[signal SIGSEGV: segmentation violation code=0x1 addr=0x10 pc=0x6a9689]

goroutine 1 [running]:
github.com/exoscale/egoscale.(*Client).dnsRequest(0x0, 0xe81c00, 0xc0001b9680, 0xc000501ac0, 0x16, 0x0, 0x0, 0x0, 0xc93196, 0x3, ...)
	/src/vendor/github.com/exoscale/egoscale/dns.go:314 +0x59
github.com/exoscale/egoscale.(*Client).GetDomain(0x0, 0xe81c00, 0xc0001b9680, 0x7ffd4db36741, 0xa, 0x0, 0x0, 0xc000501ba0)
	/src/vendor/github.com/exoscale/egoscale/dns.go:157 +0xd4
github.com/exoscale/cli/cmd.glob..func12(0x13d0e80, 0xc0005612c0, 0x1, 0x5, 0x0, 0x0)
	/src/cmd/dns_records_add.go:35 +0x19b
github.com/spf13/cobra.(*Command).execute(0x13d0e80, 0xc000561270, 0x5, 0x5, 0x13d0e80, 0xc000561270)
	/src/vendor/github.com/spf13/cobra/command.go:762 +0x473
github.com/spf13/cobra.(*Command).ExecuteC(0x13d7e20, 0x20, 0xc000000300, 0xaf252d)
	/src/vendor/github.com/spf13/cobra/command.go:852 +0x2fd
github.com/spf13/cobra.(*Command).Execute(0x13d7e20, 0xd2e8d8, 0xc000219e00)
	/src/vendor/github.com/spf13/cobra/command.go:800 +0x2b
github.com/exoscale/cli/cmd.Execute(0xe5ed10, 0x5, 0xe7d000, 0x28)
	/src/cmd/root.go:92 +0x1c2
main.main()
	/src/main.go:34 +0x5e

My CLI is running in a docker container and configured with ENV variables. Other commands such as vm list and create work fine.

firewall add my-ip create two rules

Adding the SSH rule with my-ip create two rules when only the one containing my-ip should exist.

% exo firewall show foo
(nothing)

% exo firewall add foo ssh --my-ip
INGRESS CIDR 0.0.0.0/0 tcp 22 (ssh)
        CIDR 80.224.112.21/32 tcp 22 (ssh)

Content-Type sniffing broken on files < 512 bytes

Hello,

CLI doesn't sniff Content Type on files < 512bytes, leading to objects served with a "Content-Type: application/octet-stream".
Please find a PR that should fix the problem : #129

Alex

Create a function to execute non async tasks asynchronousely

Can be useful for #87
PR: #89

• egoscale.command

• egoscale.booleanRequest Doesn't implement command interface

unknown shorthand flag -A in exo api

first:
unknown flag

go run main.go -A preprod api vm ls
unknown shorthand flag: 'A' in -A
exit status 1

and with long flag version

go run main.go --account preprod api vm ls
API error ParamError 431 (InvalidParameterValueException 4350): could not find account preprod in domain XXXX-xXXXXX-XXXX-XXXXX
exit status 1

but without api it's working good

go run main.go --account preprod vm ls
...
....

Add progress bar on exo sos download

Issue by pierre-emmanuelJ
Friday Aug 31, 2018 at 16:11 GMT
Originally opened as exoscale/egoscale#308

---

make like the actual sos upload command

exo sos upload bucket <local path>
testFile done [=================================================] 100 %
Successfully uploaded testFile of size 602 KiB

autocompletion setup

fix and documentation the auto completion build, installation and usage.

interactive config does read using Powershell

The first selection appears, then the question doesn't.

Sending y or N goes further though.

Cannot spin up my little kube with 'exo lab kube'

I've installed the exo cli (1.0.7) on my mac and tried to spin up a one node kube.
The deployment step went fine but it seems the api-server is not started.
Any idea what I'm missing ?

$ exo lab kube create 1ptikub
Creating private SSH key
Deploying "1ptikub" ⠼                                                                                done!
Bootstrapping Kubernetes cluster (can take up to several minutes):
Instance system upgrade... success!
Docker Engine installation... success!
Kubernetes cluster node installation... success!
Kubernetes cluster node initialization... success!

Your Kubernetes cluster is ready. What to do now?

1. Install the "kubectl" command, if you don't have it already:

    https://kubernetes.io/docs/tasks/tools/install-kubectl/

2. Execute the following command:

    eval $(exo lab kube env "1ptikub")

You might want to persist this change by adding it to your shell startup
configuration (e.g. ~/.bashrc, ~/.zshrc).

3. Check that your cluster is reachable:

    kubectl cluster-info

4. When you're done with your cluster, you can either:
* stop it using the "exo lab kube stop" command
* restart it later using the "exo lab kube start" command
* delete it permanently using the "exo lab kube delete" command

The error I got when running a kubectl command

$ eval $(exo lab kube env "1ptikub")
$ kubectl cluster-info
Kubernetes master is running at https://185.19.28.89:6443

To further debug and diagnose cluster problems, use 'kubectl cluster-info dump'.
Unable to connect to the server: dial tcp 185.19.28.89:6443: i/o timeout

sos metadata: some characters are reserved

> exo.exe sos metadata add yoanyoanyoan a%20b.zip "a&b" "c"
Put https://ppsos-ch-dk-2.exo.io/yoanyoanyoan/a%2520b.zip: net/http: invalid header field name "x-amz-meta-a&b"

SOS: add exo sos show cmd

there is already some cmd to show info like:
exo sos acl list
exo sos header list
exo sos metadata list

but why not add exo sos show <bucket> <object name>
to show info like on the web consol (object url) and why not acl, header, meta to have the info quicker

exo lab kube: add support for user-specified k8s version

Sometimes running exo lab kube create fails because the latest version of Kubernetes (retrieved at https://dl.k8s.io/release/stable.txt) is not yet packaged by Ubuntu). For example right now the latest k8s version advertised is 1.13.4, however creating exokube instances fails because the latest packaged version of the kubelet/kubeadm/kubectl utilities is 1.13.3.

Adding an optional --version flag to the exo lab kube create command would allow users to specify a Kubernetes version (e.g. 1.13.3) as a temporary workaround.

Add JSON output

We have standard exo tab output:

% exo vm ls
┼───────┼────────────────┼────────────────┼─────────┼──────────┼──────────────────────────────────────┼
│ NAME  │ SECURITY GROUP │   IP ADDRESS   │ STATUS  │   ZONE   │                  ID                  │
┼───────┼────────────────┼────────────────┼─────────┼──────────┼──────────────────────────────────────┼
│ test3 │ default        │ 89.145.162.134 │ Running │ de-fra-1 │ 1c288f5d-4f21-4f6b-8ff1-31f071e975f7 │
│ test2 │ default        │ 89.145.162.74  │ Running │ de-fra-1 │ 53a75824-4d0f-4a6c-98c2-0603abe2cda8 │
│ test1 │ default        │ 89.145.161.66  │ Running │ de-fra-1 │ d5767a87-7a56-47b5-8712-9efed5e1deb7 │
┼───────┼────────────────┼────────────────┼─────────┼──────────┼──────────────────────────────────────┼

We have Markdown output:

% exo vm ls | cat

NAME	SECURITY GROUP	IP ADDRESS	STATUS	ZONE	ID
test3	default	89.145.162.134	Running	de-fra-1	1c288f5d-4f21-4f6b-8ff1-31f071e975f7
test2	default	89.145.162.74	Running	de-fra-1	53a75824-4d0f-4a6c-98c2-0603abe2cda8
test1	default	89.145.161.66	Running	de-fra-1	d5767a87-7a56-47b5-8712-9efed5e1deb7

Example for Json

{
  "vm": [
    {
      "name": "test1",
      "security-group": "default",
      "ip-address": "89.145.161.66",
      "status": "running",
      "zone": "de-fra-1",
      "id": "d5767a87-7a56"
    },
    {
      "name": "test2",
      "security-group": "default",
      "ip-address": "89.145.161.66",
      "status": "running",
      "zone": "de-fra-1",
      "id": "d5767a87-7a56"
    },
    {
      "name": "test3",
      "security-group": "default",
      "ip-address": "89.145.161.66",
      "status": "running",
      "zone": "de-fra-1",
      "id": "d5767a87-7a56"
    }
  ]
}

SOS: corrupted file on upload

bug on exo sos upload
I have tested the download (okay with file uploaded from the web console)

$ sha1sum big.video.mkv
87431f013c576b6faa45f105fb7f6946891dcc2c big.video.mkv

$ exo sos upload buckettest big.video.mkv
done! 100 %  big.video.mkv

$ exo sos download buckettest  big.video.mkv testdownload.mkv
Successfully downloaded "big.video.mkv" into "testdownload.mkv"

$ sha1sum testdownload.mkv 
f65c9d28f2715034eee0921eb157643100eab74e testdownload.mkv

BUG: checksum are different

Download is okay!

$ exo sos download buckettest  big.video.mkv testdownload1.mkv
Successfully downloaded "big.video.mkv" into "testdownload1.mkv"
$ exo sos download buckettest  big.video.mkv testdownload2.mkv
Successfully downloaded "big.video.mkv" into "testdownload2.mkv"
$ exo sos download buckettest  big.video.mkv testdownload3.mkv
Successfully downloaded "big.video.mkv" into "testdownload3.mkv"

$ sha1sum testdownload*.mkv
f65c9d28f2715034eee0921eb157643100eab74e  testdownload1.mkv
f65c9d28f2715034eee0921eb157643100eab74e  testdownload2.mkv
f65c9d28f2715034eee0921eb157643100eab74e  testdownload3.mkv

all download are valid!

Managed private networks

Issue by mcorbin
Thursday Sep 13, 2018 at 10:37 GMT
Originally opened as exoscale/egoscale#319

---

Add support for managed private network in Egoscale/Exo cli:

• createNetwork: add the startip, endip, netmask parameters.
• updateNetwork: add the startip, endip, netmask parameters.
• listNetworks: show the startip, endip, netmask parameters in the output.
• addNicToVirtualMachine: add the ip parameter.
• updateVmNicIp: this new call should be implemented.

`exo eip delete` only deletes first argument, contrary to other commands

In commands as exo firewall delete is possible to pass multiple params:

$ exo firewall delete A B C
[+] Are you sure you want to delete the security group: "A" [yN]: y
b0e5f8da-d1b7-47cd-933e-6237a28615d7
[+] Are you sure you want to delete the security group: "B" [yN]: y
e07b16e9-5d58-41a4-af4f-47ade67c2e03
[+] Are you sure you want to delete the security group: "C" [yN]: y
1e849d90-0e02-4f26-b17c-7d630ff80319

This is not true for the exo eip delete command.

To reproduce:

$ exo eip delete 89.145.165.78 89.145.165.61
[+] sure you want to delete "89.145.165.78" EIP [yN]: y
0xc00017c5e0

Expected behaviour:

$ exo eip delete 89.145.165.78 89.145.165.61
[+] sure you want to delete "89.145.165.78" EIP [yN]: y
# either return the resource uuid as done for other cases or a human readable confirmation message
[+] sure you want to delete "89.145.165.61" EIP [yN]: y
# either return the resource uuid as done for other cases or a human readable confirmation message

Fix gh-pages website build

Add a `--vanilla` flag to `exo sos list`

Listing objects in buckets is great but gives this kind of output:

 [2018-06-28 12:17:19 UTC] 411 KiB FILENAME_1
 [2018-06-28 12:17:18 UTC] 418 KiB FILENAME_2
 ...

It would be nice to have an option to have the output as filename only, in order to use the list in subsequent commands without having to manipulate the string.

exo lab kube to put kubeconfig into a proper context

https://github.com/ahmetb/kubectx

% kubectx
minikube

% exo lab kube create bar
...

% kubectx
minikube
exo-kube-bar

help. https://kubernetes.io/docs/setup/certificates/#configure-certificates-for-user-accounts

KUBECONFIG=<filename> kubectl config set-cluster default-cluster --server=https://<host ip>:6443 --certificate-authority <path-to-kubernetes-ca> --embed-certs
KUBECONFIG=<filename> kubectl config set-credentials <credential-name> --client-key <path-to-key>.pem --client-certificate <path-to-cert>.pem --embed-certs
KUBECONFIG=<filename> kubectl config set-context default-system --cluster default-cluster --user <credential-name>
KUBECONFIG=<filename> kubectl config use-context default-system

runstatus unmarshalling errors

% EXOSCALE_TRACE= exo runstatus list

Content-Type: application/json

{"detail":"Invalid Authorization header"}

json: cannot unmarshal string into Go value of type []string

SOS: upload freeze on empty file

all is in the title

`exo eip delete` returns 0xc0...

To reproduce:

$ exo eip delete 89.145.165.61
[+] sure you want to delete "89.145.165.61" EIP [yN]: y
0xc00017c5e0

Expected behaviour:

$ exo eip delete 89.145.165.61
[+] sure you want to delete "89.145.165.61" EIP [yN]: y
# either return the resource uuid as done for other cases or a human readable confirmation message