ph-ee: detailed instructions for ph-ee-env-template and ph-ee-env-labs #12
Conversation
| @@ -36,13 +36,58 @@ Depending on the actual configuration, the payment hub's Helm chart installs the | |||
| * Kibana | |||
| * Zeebe Operate monitoring UI | |||
|
|
|||
| Each of these components has a git repository in [https://github.com/openMF](https://github.com/openMF), containing source code and Dockerfiles. | |||
There was a problem hiding this comment.
Thanks for the taking the Initiative to document your experience.
I have a couple of nits. :)
- Can we change the link point to 'https://github.com/orgs/openMF/repositories?q=ph-ee'? This will help first timers right to the ph-ee repos.
|
|
||
| Most components will also have a Dockerfile. After building the image, you may wish to host these docker images in a local docker registry. This can be done using docker locally per [https://docs.docker.com/registry/deploying/](https://docs.docker.com/registry/deploying/), once you make sure that whatever domain name you use is available via DNS, or a local entry in /etc/hosts. | ||
|
|
||
| If you wish to setup a local maven repository to host ph-ee-connector-common, this page describes how to set one up using nginx: [https://stackoverflow.com/questions/13834141/create-local-maven-repository](https://docs.docker.com/registry/deploying/) |
There was a problem hiding this comment.
- Looks like a typo, Stackoverflow link points the previous docker link.
- Can we have a simpler version for ph-ee-connector-common use? In my practice, I just do 'mvn clean install' in my local and this is enough for other services to use connector-common from local mvn registry while building.
There was a problem hiding this comment.
Oh, that would simplify things. Thanks! I'm a bit new to this as you can probably tell
| Inspect [https://github.com/openMF/ph-ee-env-template/blob/master/helm/package.sh](https://github.com/openMF/ph-ee-env-template/blob/master/helm/package.sh) for build instructions, and customize the commands to suit your installation. Unless you are remotely pushing the files, you will not need to run scp. Only cp is needed in that case, however please ensure the web root directory (/usr/share/nginx/html) is updated to reflect your configuration. If using the nginx file above, it would be `/var/www/jenkinsmifos-helm`. | ||
|
|
||
| Verify your results by doing `curl http://jenkins.mifos.io:8082/index.yaml` to download your index.yaml. If problems occur, check your web server logs. | ||
|
|
There was a problem hiding this comment.
LGTM. I was thinking of cutting out the Jenkins.mifos.io part completely and move the charts to a GitHub Pages based Helm repo. As seen here. We can take this up later with a proof of concept.
|
Addressed two nits above, also added another line clarifying how to host with a local docker registry |
|
I saw this was still open, so added another commit. Instead of the deployment script killing all kubectl instances, it will save the backgrounded instances PID and kill that Is there anything else required for this PR @SubhamPramanik? |
|
LGTM I actually don't have access to this repo to merge your PR. Usually I approve or edit changes in gitbook directly(which is linked to this repo, not sure why my access hasn't propagated to the repo), can you try sending a change request over there? I'll be able to approve it. |
No description provided.