gcloud deploy issue - boot2docker 1.3.1 - node.js

This is a follow on from the question here:
gcloud deploy error: Image with tag google/docker-registry was not found which is based on the post here http://www.incrediblemolk.com/running-node-js-on-google-app-engine-managed-vms/ .
I've followed the steps in the second link and succeeded in getting gcloud and docker to run the github project https://github.com/GoogleCloudPlatform/appengine-nodejs-quickstart locally. However I am stuck on deployment.
After downloading the docker-registry container, I found that I had the 'insecure registry' issue, so I followed the steps in the top rated answer in the stack overflow question linked above. However when I tried to restart the daemon, I received a 'Tls message' as follows:
docker#boot2docker:/mnt/sda1/var/lib/boot2docker$ sudo /etc/init.d/docker restart
Need TLS certs for boot2docker,127.0.0.1,10.0.2.15,192.168.59.103
I was wondering if there is something that I need to do, or if this is a boot2docker issue? I noticed that in the boot2docker project on github, people mentioned 'workarounds' for disabling tls, which I managed to do by adding a line
DOCKER_TLS=no
to my /var/lib/boot2docker/profile file within the VM, and this did enable me to restart the docker daemon, however "gcloud preview app deploy --server preview.appengine.google.com ." requires that tls is enabled. So I feel like I'm untying one hand from behind my back and replacing it with the other in this instance!
Is there some fix required for this, or am I missing something important? If so, do I need to use boot2docker 1.3.0 as opposed to 1.3.1?
If anyone could help me in this matter that would be greatly appreciated.
Otherwise, my strategy I think would be to wait for these tools to mature a bit more, as I am after all using the preview version of gcloud, and I think TLS support in boot2docker is relatively new.

Yes, 1.3.1 is not compatible with our SDK. Plese use the latest (as of today) 1.3.2 or 1.3.0.

Related

Please add a #NgModule annotation but I cannot figure why

I have a Node/Angular app which works locally in a Docker instance. For deployment to a Google VM, we have a script which checks out the Git files, compiles them, copies the outputs into an nginx folder and restarts the nginx.
I have (what I thought was) identical setups on two different VMs. I test any changes on the "staging" VM and then if all is working correctly, I run them on the "production" VM.
Yesterday, this system broke. My staging VM pulls, compiles and runs the code without any issues or even warnings. The production VM gives errors during the compilation
The compilation line is
sudo $(npm bin)/ng build --prod --output-path=dist
and the first error I see is
ERROR in : Unexpected value 'AngularFireDatabaseModule in /home/redacted/ng-app/node_modules/#angular/fire/database/angular-fire-database.d.ts' imported by the module 'AppModule in /home/redacted/ng-app/src/app/app.module.ts'. Please add a #NgModule annotation.
So, I saw this question which said I should move import to the declarations section. I did this for AngularFireDatabaseModule and then it would not compile on my (working) staging server, giving me an error
Please add a #Pipe/#Directive/#Component annotation
Which this answer says means I need to move the declaration back to the imports!
My assumption is that my two supposedly identical VMs have something different about them, but I am having difficulty figuring out what.
Is there some kind soul who can help me figure out what the difference is?
Both VMs: Running Ubuntu 20.04.3 LTS (GNU/Linux 5.11.0-1028-gcp x86_64)
Both VMs up to date
Both VMs: node --version
v17.4.0
Both VMs: sudo node --version
v13.14.0
Both VMs: npm -v
8.3.1
Both VMs: sudo npm -v
6.14.4
What else should I be checking for?
OK, I can't be sure what exactly fixed it, but in the end, I had to go into the folder where the compilation was occurring and run
sudo npm update
which seemd to fix the problems. Could've sworn I did that a few times before, but ¯\(ツ)/¯
Leaving this for anyone who is despairing - the simple answers seem to work best

Docker container --tls-min-v1.0 is not allowed in NODE_OPTIONS

I am trying to run a Kibana docker container.
When I launch it I get this error
/usr/share/kibana/bin/../node/bin/node: --tls-min-v1.0 is not allowed
in NODE_OPTIONS
I cannot see any options being specified. I thought I cracked it when I noted that this option is only available in NODE > 12.00
I upgraded the container to use NODE 12.3.1 but the problem persists.
Has anyone come across this before, and if so how did you go about fixing it?

how to update etherpad-lite properly?

I keep totally failing to update an etherpad-lite server. The problem: Even a Google search for the update procedure brings up hardly any information, only that one should run "git pull origin".
I have now tried this in many different ways. The update usually works, but afterwards one of these errors occurs:
Plugins can no longer be installed
The service can no longer be started (TypeError: log4js.replaceConsole is not a function)
The entire admin panel no longer works.
I tried uninstalling or updating all plug-ins before, but both hardly brought any improvements, only other errors. The update of the plugins in the admin console fails, I tried it via the updatePlugins.sh script. Here a message appears that at least etherpad-lite 1.8,6 must be installed. I am currently at version 1.8.4 and would like to update to the latest version 1.8.12. However, some of the plug-ins are still updated. A very strange behaviour.
I would be happy if someone could tell me how to properly update the etherpad-lite instance step by step. (ubuntu 20)
Thank you!
I have recently updated Etherpad-lite from version 1.8.6 to 1.8.13.
For me, executing git pull origin and then checking out the 1.8.13 release tag, with git checkout 1.8.13, made the work.
It is important, despite having Etherpad configured as a service, to run it for the first time using:
src/bin/run.sh.
Node v12.22.1
npm 6.14.12
Ubuntu 20.04.2
I hope it has been useful to you.

Is there a way to enable kubernetes on docker via command line on linux?

I am using docker and trying to enable kubernetes and set CPU and Memory via command line.
I have looked at this answer but unfortunately cannot find this file.
Is there any way to enable Kubernetes on Docker for Mac via terminal?
Docker does not have an app-ified version for Linux that I know of, so there is no relation to the Docker for Mac/Windows app. There are many tools to locally install Kubernetes on Linux so they probably didn't see much reason to make something new. Minikube is the traditional one, but you can also check out microk8s, k3s, KinD, and many others.

how to install ForgeRock Open Identity Stack

I download frstack from
https://github.com/wstrange/frstack#install-the-forgerock-open-identity-stack-ois
After I install Ansible and Vagrant, I don't know how to make it work
Are there any references available?
Thanks
Author of that project here. That is very much a work in progress - so I would not start with that.
Just get a tomcat instance running and deploy OpenAM to it. Once you understand the basics then you can look at an automated deployment using Ansible.

Resources