Replies: 31 comments 31 replies
-
|
Hi Amos, Do you have |
Beta Was this translation helpful? Give feedback.
-
|
Apologies for the late response I was on leave this last week. Yes, the I'd up to now followed the instructions n the README
But let me step back a bit, running the exact instructions in the README A number of errors owing to missing parameters. State at failERRORS (each resolved by setting the value): and it is in the go template as indicated RADAR-Kubernetes/charts/management-portal/values.yaml (which lead me to the error I posted in the original post) I noted that error 1-5 are resolved (I think correctly by adding the
|
Beta Was this translation helpful? Give feedback.
-
|
In the base.yaml file the _install key is defined for management portal: Have you changed any file that are tracked by Git? When you run |
Beta Was this translation helpful? Give feedback.
-
I'm a little confused here as production.yaml is just a copy of base.yaml so always contains the _install value |
Beta Was this translation helpful? Give feedback.
-
|
I've not changed those lines, the value is defined unmodified in management_portal:
_install: true
replicaCount: 1 # should be 1
postgres:
host: postgresql
ssl:
enabled: false |
Beta Was this translation helpful? Give feedback.
-
|
I think I spotted the issue in my This gets past the issue. There are new issues now . which I looks like it can be resolved by quoting the key value in MP crashlooping so maybe above not correct? logs from the MP container suggests it's not able to access the keystore... however it is not at least failing and assuming above is fixed the next issue are more familiar param missing issue which I think are readily resolved by adding to |
Beta Was this translation helpful? Give feedback.
-
|
This is how The error from MP suggests that the keystore file haven't been defined. Which should give and output like this: You can further debug this in your machine with this command: And check the output to see how helmfile is reading the variables. You can also remove the |
Beta Was this translation helpful? Give feedback.
-
|
don't know if specific to my setup but it fails... (I tried again from a new clone to check). Also I see no use of |
Beta Was this translation helpful? Give feedback.
-
|
Keystore secret gets created in this file: keystore.p12 file does exist in correct path, please check if the secret is actually there once the MP is installed on the cluster, if the MP is installed and you still can't find the secret it means that it couldn't read the keystore file which means there is an issue in Sorry the command I have you in previous message had a typo as I copied it from my local machine. This on should return some result for you: The |
Beta Was this translation helpful? Give feedback.
-
ok that's helpful 👍 keystore is correctly created after management portal deployed |
Beta Was this translation helpful? Give feedback.
-
|
a few more missing values set then we get here (I think @ThomasKassiotis also hit this one). fails now here. |
Beta Was this translation helpful? Give feedback.
-
|
Please change the last line of To: And try again. |
Beta Was this translation helpful? Give feedback.
-
|
It's working thanks ! |
Beta Was this translation helpful? Give feedback.
-
resolved by adding the next break I think correctly fixed by adding to finally breaking here -- not sure if you are able to shed some light on this? container logs for |
Beta Was this translation helpful? Give feedback.
-
|
Looking at this line: It says that the upload backend app can't connect to the PostgreSQL. In 20-upload.yaml helmfile the first app that should be installed is |
Beta Was this translation helpful? Give feedback.
-
|
I didn't make changes
|
Beta Was this translation helpful? Give feedback.
-
|
Are you running helmfile sync with |
Beta Was this translation helpful? Give feedback.
-
|
Logs from release=radar-upload-connect-backend, as it hangs before it gets removed Not sure if this is missing Log File: |
Beta Was this translation helpful? Give feedback.
-
|
Could you send the logs from upload connector PostgreSQL as well? |
Beta Was this translation helpful? Give feedback.
-
|
here you go |
Beta Was this translation helpful? Give feedback.
-
|
sorry above postgres not correct one... but I can't see the postgres pod you want |
Beta Was this translation helpful? Give feedback.
-
|
so that did fix the radar-upload-postgres issue next error This above issue relates to bases:
- ../environments.yaml
---
helmDefaults:
atomic: true
timeout: 600
kubeContext: {{ .Values.kubeContext }}
releases:
- name: cc-schema-registry-proxy
chart: ../charts/cc-schema-registry-proxy
installed: {{ .Values.ccSchemaRegistryProxy._install }}
values:
- {{ .Values.ccSchemaRegistryProxy | toYaml | indent 8 | trim }}
- {{ .Values.confluent_cloud| toYaml | indent 8 | trim }}
set:
- name: service.externalName
value: {{ .Values.ccSchemaRegistryProxy.externalName }}
- name: ingress.hosts
values: [{{ .Values.server_name }}]The |
Beta Was this translation helpful? Give feedback.
-
|
Regarding radar-output: please ensure that the access key / secret of minio matches the one in radar-output and run minio:
accessKey: EXAMPLEA
secretKey: EXAMPLESECRETKEYA
radar_output:
source:
s3:
accessToken: EXAMPLEA
secretKey: EXAMPLESECRETKEYA
target:
s3:
accessToken: EXAMPLEA
secretKey: EXAMPLESECRETKEYA |
Beta Was this translation helpful? Give feedback.
-
|
do you mean in I can only see reference to these e.g. |
Beta Was this translation helpful? Give feedback.
-
|
updating the NAME READY STATUS RESTARTS AGE
svclb-nginx-ingress-ingress-nginx-controller-cthvn 0/2 Pending 0 28m
nginx-ingress-ingress-nginx-controller-68df68bc48-pz5vc 1/1 Running 0 28m
cp-zookeeper-0 2/2 Running 0 28m
cp-zookeeper-1 2/2 Running 0 27m
cp-zookeeper-2 2/2 Running 0 27m
cp-kafka-0 2/2 Running 0 26m
cp-kafka-1 2/2 Running 0 26m
cp-kafka-2 2/2 Running 0 25m
cp-schema-registry-6fdf97d6f7-dtq46 2/2 Running 0 24m
postgresql-postgresql-0 3/3 Running 0 23m
app-config-frontend-7b99d549f7-lczp8 1/1 Running 0 23m
radar-grafana-56bcf6f5b4-4pzmr 1/1 Running 0 23m
timescaledb-postgresql-0 3/3 Running 0 23m
catalog-server-6799869c49-lgjmc 1/1 Running 0 23m
app-config-7b5964657d-x7ph4 1/1 Running 0 23m
radar-jdbc-connector-549dbb86b7-z454x 1/1 Running 0 23m
management-portal-684d555d55-9srr8 1/1 Running 1 23m
cp-kafka-rest-5665d4f67c-2j82m 2/2 Running 0 23m
radar-gateway-bcd998964-djk6l 2/2 Running 0 21m
minio-1 1/1 Running 0 19m
minio-3 1/1 Running 0 19m
minio-0 1/1 Running 0 19m
minio-2 1/1 Running 0 19m
redis-master-0 2/2 Running 0 19m
radar-output-647ff5cfbd-fqh9g 1/1 Running 0 19m
radar-upload-postgresql-postgresql-0 3/3 Running 0 19m
radar-upload-connect-backend-5d8cdc6797-lfjtr 1/1 Running 0 19m
radar-upload-connect-frontend-74f5d98887-695l4 1/1 Running 0 19m
radar-s3-connector-8575bf48c9-l4472 1/1 Running 0 19m
radar-upload-source-connector-85ffb5c6f5-j4zwh 1/1 Running 0 19mThe only issue remains the pending ingress controller. I think I just need to noodle around with the port-forward settings note name of the service in the README is different to what I have Note the default testing with unprivileged ports ok for testing e.g. point browser to --> https://example.com:4443/managementportal seems to be tentatively working now -- again only tested very superficially. the MP the upload connector |
Beta Was this translation helpful? Give feedback.
-
|
No the chart values should remain as they were: those should not be pre-specified since minio is exposed to the internet. You have to give those access keys etc. via the latest file in your environments.yaml, so if you’re using production.yaml as latest file, that’s where the key should go. Helmfile values should override the chart values. |
Beta Was this translation helpful? Give feedback.
-
|
ok yes that worked. |
Beta Was this translation helpful? Give feedback.
-
|
Hey, on a fresh install of microk8s, the default context does not exist. Therefore the install fails: |
Beta Was this translation helpful? Give feedback.
-
|
What do you get with $ kubectl config view
apiVersion: v1
clusters:
- cluster:
certificate-authority-data: DATA+OMITTED
server: https://127.0.0.1:6443
name: default
contexts:
- context:
cluster: default
user: default
name: default
current-context: default
kind: Config
preferences: {}
users:
- name: default
user:
client-certificate-data: REDACTED
client-key-data: REDACTEDthen edit the kubeContext: default
server_name: example.com
maintainer_email: [email protected] |
Beta Was this translation helpful? Give feedback.
-
|
I have since uninstalled microk8s. I'm trying with k3s. On microk8s the context created by default is called microk8s. I know how to fix this error, I'm just pointing it out so you can document it in the README. |
Beta Was this translation helpful? Give feedback.
-
|
On a fresh install of k3s, I get this error: I'm not sure what's the cause, but unfortunately, the PR does not work as-is for me. The output of |
Beta Was this translation helpful? Give feedback.
-
|
There is never anything in the logs. It always stays pending then gets uninstalled! |
Beta Was this translation helpful? Give feedback.
-
|
@blootsvoets / @K1Hyve would you re @iDmple's issue in |
Beta Was this translation helpful? Give feedback.
-
|
Is it possible that another service isn't coming up and that the gateway is waiting for that? There are several services pending in the list above. |
Beta Was this translation helpful? Give feedback.
-
|
I've tried your suggestion (atomic: false and timeout increased to 1200) and get this log: get pods shows it as running. How can we fix this so it works without the changes? |
Beta Was this translation helpful? Give feedback.
-
|
Maybe rather than Also, I just now tried re-running the stack from scratch -- seems to work. |
Beta Was this translation helpful? Give feedback.
-
|
My next problem is minio-update-prometheus-secret gets into timeout: How could I fix this? Thanks. |
Beta Was this translation helpful? Give feedback.
-
|
Please run |
Beta Was this translation helpful? Give feedback.
-
|
I have 12 cpus with 32GB of RAM. Will execute the command and let you know. It would be good to include these requirements on the README. |
Beta Was this translation helpful? Give feedback.
-
|
In the meantime I found out that in This provokes an authentication failing in the app-config pod: Thomas also found out that in Is base.secrets.yaml needed or not? This is unclear why such a file exists if not needed? This isn't documented. |
Beta Was this translation helpful? Give feedback.
-
|
This is the result of the command you asked: |
Beta Was this translation helpful? Give feedback.
-
|
Okay, I found out that /var is full (unfortunately it's in its own partition and I have no control over that). I will fix this and come back to you. I guess we can consider the minio error was due to this. |
Beta Was this translation helpful? Give feedback.
-
|
I solved the previous issues and reached further into the installation process. Now, I have another issue: svclb-nginx-ingress-ingress-nginx-controller is pending because: I have 80 and 443 open on the machine and nothing else is using them: What is causing this? Thanks! |
Beta Was this translation helpful? Give feedback.
-
|
For the external ip try where SERVICENAME -> nginx-ingress-ingress-nginx-controller and 192.168.0.X -> your external IP |
Beta Was this translation helpful? Give feedback.
-
|
works thanks! |
Beta Was this translation helpful? Give feedback.
-
|
I now get the following when trying to access the management portal: Trying to debug this I can provide these logs: and Thanks for your help. |
Beta Was this translation helpful? Give feedback.
-
|
What URL you are trying to access to? |
Beta Was this translation helpful? Give feedback.
-
|
I'm indeed accessing that URL. This appears in the logs of the ingress controller: |
Beta Was this translation helpful? Give feedback.
-
|
I found out that k3s deploys its own traefik ingress and loadbalancer in another namespace, and those were taking the port that the nginx-ingress-controller needed. I installed k3s without traefik and servicelb, and now I finally have all pods running: However when accessing the Management Portal I get Can you please advise? Thanks! |
Beta Was this translation helpful? Give feedback.
-
|
This seems like that there is still issues with ingress. Please check if there aren't any extra ingress rules or controllers defined in the cluster. |
Beta Was this translation helpful? Give feedback.
-
|
Here are they: |
Beta Was this translation helpful? Give feedback.
-
|
This looks fine, have you changed the radar.domain when you were pasting it here or that's how it has been configured? |
Beta Was this translation helpful? Give feedback.
-
|
I changed the domain indeed. I changed a few files: The passwords were hardcoded as well as the maintainer email. Without changing the passwords I had authentication errors and that is solved now. The server name is also hardcoded to localhost, I thought I modified it but I seem to have forgotten to do so. |
Beta Was this translation helpful? Give feedback.
-
|
Since you had made many changes, I decided to start over from scratch. I have errors and will open a new thread. |
Beta Was this translation helpful? Give feedback.
-
@K1Hyve Looks like the values in
base.secrets.yamlare needed, but not presently referenced -- is this the proper place for this at the end ofenviromnent.yamlmaking the assumption of the above adding the
base.secrets.yamltoenvironment.yamlI think has resolved the issues with various unset default values.Values.postgres_password.Values.management_portal.oauth_clients.xxxxx.client_secret*in
production.yaml.gotmpl#uncomment:Still failing here though:
in helmfile.d/10-managementportal.yaml: error during 10-managementportal.yaml.part.1 parsing: template: stringTemplate:27:25: executing "stringTemplate" at <.Values.management_portal._install>: map has no entry for key "_install"
Beta Was this translation helpful? Give feedback.
All reactions