-
Notifications
You must be signed in to change notification settings - Fork 229
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
SSH connection failed #40
Comments
I'm finding that test-kitchen with vagrant-lxc is a better fit. |
@fnichol can we increase the number of SSH attempts for asset scp, or make it configurable? |
Hmm, wait_for_ssh() should be avoiding this issue. |
Isn't there some bit about Ubuntu restarting sshd at startup? Could be why it's |
I'm also running into this issue, running on Jenkins:
|
After looking into this, it looks like the code that establishes an SSH connection in Test Kitchen retries but doesn't pause between attempts (unlike the |
I too am experiencing this issue, and I believe test-kitchen/test-kitchen/pull/454 may offer a workaround and/or solution if it is merged. |
I'm facing same issue, |
Is it possible that this issue is not with SSH'ing but with SCP? Can you try to copy something with SCP into the container or install openssh-server+openssh-client (SCP is in client) on the container. |
+1 |
+1
|
I'm having the same problem and I found a workaround:
|
I've the same
I got rid of it commenting out Follow up problem is, of course no services are running, can't test my stuff properly :-( |
I have the same issue when starting systemd, seems like the vanilla startup-command, which Since i require upstart / systemd |
Here's our kitchen.yml, which include our workaround for this issue, the first provision command. The second one fix a testinfra related issue which may be useful for other tools relying on /sbin/init. ---
driver:
name: docker
use_sudo: false
provision_command:
- rm /lib/systemd/system/ssh.service
- '[ ! -f /sbin/init ] && ln -s /lib/systemd/systemd /sbin/init || true'
run_command: /bin/systemd
privileged: true
volume:
- "/sys/fs/cgroup:/sys/fs/cgroup:ro"
dns:
- 1.1.1.1
- 9.9.9.9
transport:
name: sftp
platforms:
- name: stretch
driver_config:
image: jrei/systemd-debian:9
platform: debian
- name: buster
driver_config:
image: jrei/systemd-debian:10
platform: debian
suites:
- name: nitrogen
provisioner:
salt_bootstrap_options: -X -p git -x python2.7 stable 2017.7
- name: fluorine
provisioner:
salt_bootstrap_options: -X -p git -x python2.7 stable 2019.2
provisioner:
name: salt_solo
salt_install: bootstrap
is_file_root: true
require_chef: true
salt_copy_filter:
- .git
dependencies:
- name: common
repo: git
source: https://gitlab+deploy-token-6:[email protected]/salt/common-formula
branch: dev
state_top:
base:
"*":
- bender
pillars_from_files:
pillar.sls: test/pillar.sls
pillars:
top.sls:
base:
"*":
- pillar
verifier:
name: shell
remote_exec: false
command: pytest --junitxml=test/${KITCHEN_INSTANCE}_test_report.xml --html=test/${KITCHEN_INSTANCE}_test_report.html --self-contained-html --color=yes --host="docker://root@${KITCHEN_CONTAINER_ID}" "test/integration/" |
i fixed it right now by doing
|
I'm running into an issue where is appears test-kitchen isn't waiting long enough to SSH when I'm using /sbin/init for my run_command. If I wait a couple of second, I can then ssh into the instance. Is this something that can be addressed by kitchen-docker or are there any workarounds I might employ?
I'm trying to solve the problem of runit not starting when the package has been installed. On Debian systems, which I'm testing on, it starts via the inittab and the runit cookbook handles this by issuing a telinit once the package is installed.
The text was updated successfully, but these errors were encountered: