[WIP] Fix for PORTAL_STORAGE_PGHOST in helm deployments #8
Conversation
… and localhost envs
|
|
|
Hi, yes, this is absolutely the right track. Locally, we have all been testing with the sample repository which had those adaptions in place. Thanks a bunch for this! Unfortunately, I am sick after travelling a lot, so I haven't been able to test this yet. There may also be other loose ends regarding the configuration update, but this is definitely one of the more tedious ones. There may also be something regarding the redis configuration, have you experienced any issues there? |
|
Thanks for the feedback, good to know it's the right direction. And sorry that you're ill! The redis container came up just fine, so nothing to worry about there yet. I suppose it remains to be seen whether the application speaks to the various containers correctly, but as far as kubernetes goes it as alright. |
| const localEnv = loadEnv(targetConfig, 'localhost'); | ||
| if (!localEnv.PORTAL_ECHO_URL) { | ||
| localEnv.PORTAL_ECHO_URL = { | ||
| value: "http://${LOCAL_IP}:5432" |
There was a problem hiding this comment.
This can't be right, it has to be PORTAL_STORAGE_PGHOST. not PORTAL_ECHO_URL, and the value should just be ${LOCAL_IP}. The password should probably default to the default environment's setting, and as a fallback to kong.
There was a problem hiding this comment.
Good catch, will fix.
There was a problem hiding this comment.
Now it only writes the password, not the host. And the value for the password is a little off perhaps.
|
Do you mind if I pull this in and polish the rest myself? |
This is just a WIP PR to see if I'm on the right track here.
This is in reference to Issue Haufe-Lexware/wicked.haufe.io#139. Helm deployments are failing because the
PORTAL_STORAGE_PGHOSTis using the default value ofwicked-databaseinstead of the k8s hostname.I noticed that
updateStep12_v1_0_0candupdateStep10_v1_0_0aboth had to explicitly update the k8s environment, so I simply followed that process withupdateStep14_v1_0_0e.I haven't figured out exactly how I'm going to test that the right values are getting set. I'd prefer not to go through the whole process of pushing a new image to a personal docker repo to test this out with helm, but I also can't quite envision how to spin up the docker image by itself and pass it all the variables that get set when I run helm. I thought it might be as simple as setting
envName=k8swhen I spin up the container, but I tried that and it didn't look like what I was expecting.At this point, I'd just like to hear feedback that this PR is on the right track. Feedback very welcome, especially if it's "stop, you're doing it all wrong".