In Orion chart 1.0.3, when I set multitenacy option the installation fails.

multiserviceEnabled: true results in:

Error: template: orion/templates/deployment.yaml:233:61: executing "orion/templates/deployment.yaml" at <upper>: wrong type for value; expected string; got bool

multiserviceEnabled: 'true' results in:

creating 2 resource(s)
--
Thu, May 5 2022 2:24:46 pm | Error: Deployment in version "v1" cannot be handled as a Deployment: v1.Deployment.Spec: v1.DeploymentSpec.Template: v1.PodTemplateSpec.Spec: v1.PodSpec.Containers: []v1.Container: v1.Container.Env: []v1.EnvVar: v1.EnvVar.Value: ReadString: expects " or n, but found t, error found in #10 byte of ...\|,"value":true},{"nam\|..., bigger context ...\|"orion"},{"name":"ORIONLD_MULTI_SERVICE","value":true},{"name":"ORIONLD_MONGO_AUTH_SOURCE","value":"\|...

Dear all,

Is it possible to add Wilma PEP proxy helm chart?

rpc error: code = Unknown desc = Manifest generation error (cached): `helm dependency build` failed exit status 1: Error: can't get a valid version for repositories mongodb. Try changing the version constraint in Chart.yaml

I get this error when using the latest version of the orion helm chart with mongodb version 4.0.24.

Tried to fix the error by changing the mongodb version tag to 4.0.20 in the values.yaml but the error still remains.

Using kubernetes version 1.26.3.

Any solution? Thanks!

It seems that the last change broke the helm chart for orion. Error message is

option '-lmtmp' not recognized

Change was introduced with this commit: 698869e

api-umbrella comes with a build in mongodb. The chart is missing a persistence volume to mitigate data lost from the db.

It would be also nice to have the mongodb as a dependent Chart. This would enable independent scalability of the main deployment and the database.

I'm ready to help with the implementation :)

We would like to install orion via helm chart. Currently there is only a orion-ld helm chart. Are there any plans to provide a helm chart for orion?

The Orion-LD Helm Chart include ORIONLD_MONGO_SSL. Currently this environment variable is not supported by Orion-LD version 1.0.1 (will be included in future versions). A comment in values.yml mentioning, that SSL-secured connections are currently just supported by Orion (not LD), would prevent misconceptions.

The helm charts you guys have mentioned are not getting deployed there may be some issues with charts i guess. You should clearly mention what all are dependencies for ecah Fiware GE's so that according to that we can make changes into Yaml file. But as of now no mention and how we can test weather our deployed charts are ready to use or not ?
https://github.com/FIWARE/helm-charts

As Domibus was added again as third party in the Fiware catalog do you plan to add a Domibus chart?

I found an helm chart example in:
https://github.com/dladlk/delis/tree/dev/delis-test-server/helm

I am trying to deploy orion latest version using helm chart: https://github.com/FIWARE/helm-charts/tree/main/charts/orion. For this, I have created a pod of mongo:3.6 and updated image and db details in values.yaml file of orion as:

...
  image:
    # -- orion image name
    #ref: https://hub.docker.com/r/fiware/orion/
    # ref: https://hub.docker.com/r/fiware/orion-ld
    #repository: fiware/orion-ld
    repository: fiware/orion
    # -- tag of the image to be used
    tag: latest
    # -- specification of the image pull policy
    pullPolicy: IfNotPresent
...
  db:
    # -- configuration of the mongo-db hosts. if multiple hosts are inserted, its assumed that mongo is running as a replica set
    hosts: [mongo]
    # - mongodb
    # -- the db to use. if running in multiservice mode, its used as a prefix.
    name: orion
...

after this change, orion pod is created and is in running state: kubectl get pods

but when I try to hit localhost:1026/verion api from pod's shell it shows error: error

I have tried to see logs of the pod and docker container created at worker node but nothing is showing.

At master node: kubectl logs

At worker node: docker ps docker logs docker logs

Please correct me if I am doing anything wrong and suggest me the way in which I can proceed to deploy orion using helm chart.

Hi, we are currently trying to deploy the FIWARE components in our Kubernetes cluster utilitizing the most recent versions of the components. However, using the helm charts for the IoT Agent JSON, we run into the following issue: Since the helm charts were created for version 1.14, the IoT Agent comes with a discovery for the mongoDB - and with a hint that this mechanism would be depracated in the future. Now, when trying to deploy the IoT Agent with a newer version, e.g. 1.19 or the most recent 2.4.2, this mechanism seems not be integrated any longer, which results in the IoT Agent not connecting properly to the mongoDB. We are using mongoDB 6.0.9 from bitnami.

In the values.yaml we specified the mongoDB host name as "mongodb-headless.default.svc.cluster.local". So, we are wondering how the new mechanism is currently working and if there are any more recent helm charts available to deploy the IoT Agent.

You can find the error log from the IoT Agent below:

{"time":"2023-10-23T13:01:40.218Z","lvl":"ERROR","msg":"EISDIR: illegal operation on a directory, read"}
2023-10-23T13:01:40.532553333Z {"op":"IoTAgentNGSI.JEXL","time":"2023-10-23T13:01:40.532Z","lvl":"INFO","msg":"Trasformations can be added to JEXL parser"}
 `node --trace-warnings ...` to show where the warning was created)
2023-10-23T13:01:40.657637540Z time=2023-10-23T13:01:40.656Z | lvl=ERROR | corr=92747a9f-3924-4d84-9440-27c8bab08c8e | trans=92747a9f-3924-4d84-9440-27c8bab08c8e | op=IoTAgentNGSI.DbConn | from=n/a | srv=n/a | subsrv=n/a | msg=MONGODB-001: Error trying to connect to MongoDB: MongoNetworkError: failed to connect to server [localhost:27017] on first connect [Error: connect ECONNREFUSED 127.0.0.1:27017
    at TCPConnectWrap.afterConnect [as oncomplete] (node:net:1278:16)
2023-10-23T13:01:40.657698591Z     at TCPConnectWrap.callbackTrampoline (node:internal/async_hooks:130:17) {
2023-10-23T13:01:40.657705123Z   name: 'MongoNetworkError'
    at Socket.emit (node:domain:489:12)
2023-10-23T13:01:40.657776123Z     at emitErrorNT (node:internal/streams/destroy:157:8)
    at emitErrorCloseNT (node:internal/streams/destroy:122:3)
2023-10-23T13:01:40.657786800Z     at processTicksAndRejections (node:internal/process/task_queues:83:21) | comp=IoTAgent
2023-10-23T13:01:40.657803496Z time=2023-10-23T13:01:40.657Z | lvl=INFO | corr=92747a9f-3924-4d84-9440-27c8bab08c8e | trans=92747a9f-3924-4d84-9440-27c8bab08c8e | op=IoTAgentNGSI.DbConn | from=n/a | srv=n/a | subsrv=n/a | msg=Waiting 5 seconds before attempting again. | comp=IoTAgent
2023-10-23T13:01:45.665291645Z time=2023-10-23T13:01:45.664Z | lvl=INFO | corr=92747a9f-3924-4d84-9440-27c8bab08c8e | trans=92747a9f-3924-4d84-9440-27c8bab08c8e | op=IoTAgentNGSI.DbConn | from=n/a | srv=n/a | subsrv=n/a | msg=Attempting to connect to MongoDB instance with url "mongodb://localhost:27017/iotagentjson". Attempt 2 | comp=IoTAgent
2023-10-23T13:01:45.672053231Z time=2023-10-23T13:01:45.671Z | lvl=ERROR | corr=92747a9f-3924-4d84-9440-27c8bab08c8e | trans=92747a9f-3924-4d84-9440-27c8bab08c8e | op=IoTAgentNGSI.DbConn | from=n/a | srv=n/a | subsrv=n/a | msg=MONGODB-001: Error trying to connect to MongoDB: MongoNetworkError: failed to connect to server [localhost:27017] on first connect [Error: connect ECONNREFUSED 127.0.0.1:27017
2023-10-23T13:01:45.672086861Z     at TCPConnectWrap.afterConnect [as oncomplete] (node:net:1278:16)
    at TCPConnectWrap.callbackTrampoline (node:internal/async_hooks:130:17)

We appreciate your help. Thank you in advance!

Hi, I received {"name":"BAD_REQUEST","message":"Request error connecting to the Context Broker: 501"} error when when I send a POST to iotagent-ul deployed via iotagent-ul helm chart.

I get the following log via kubectl logs pod/iotagent-ul-xxxxxx:

time=2022-10-08T09:03:18.182Z | lvl=DEBUG | corr=d7ed414d-a132-45ea-b4a3-38ee14a828dc | trans=d7ed414d-a132-45ea-b4a3-38ee14a828dc | op=IoTAgentNGSI.Request | from=n/a | srv=openiot | subsrv=/ | msg=Response {
    "error": "NotImplemented",
    "description": "Only NGSIv1-based forwarding supported at the present moment. Set explictely legacyForwarding to true"
}

Following the iotagent-ul documentation to enable NGSIv2 it must be set ngsiVersion in contextBroker:

{
    host: '192.168.56.101',
    port: '1026',
    ngsiVersion: 'v2'
}

So, I have modified:

• Chart.yaml

{
  contextBroker:
    # -- host of the broker
    host: 192.168.49.2
    # -- port of the broker
    port: 30005
    # to use NGSI v2
    ngsiVersion: v2
}

• configmap.yaml

contextBroker: {
            /**
             * Host where the Context Broker is located.
             */
            host: '{{ .Values.iota.contextBroker.host }}',

            /**
             * Port where the Context Broker is listening.
             */
            port: '{{ .Values.iota.contextBroker.port }}'
            
            /**
             * If you want to use NGSI v2.
             */
            ngsiVersion: '{{ .Values.iota.contextBroker.ngsiVersion }}'
        }

However, the new pod status is CrashLoopBackOff and the log received via kubectl logs pod/iotagent-ul-xxxx is:

{"op":"IoTAgentNGSI.DomainControl","time":"2022-10-08T09:15:31.479Z","lvl":"ERROR","msg":"/opt/iotaul/config.js:55\n        ngsiVersion: 'v2'\n        ^^^^^^^^^^^\n\nSyntaxError: Unexpected identifier\n    at Object.compileFunction (node:vm:360:18)\n    at wrapSafe (node:internal/modules/cjs/loader:1055:15)\n    at Module._compile (node:internal/modules/cjs/loader:1090:27)\n    at Object.Module._extensions..js (node:internal/modules/cjs/loader:1180:10)\n    at Module.load (node:internal/modules/cjs/loader:1004:32)\n    at Function.Module._load (node:internal/modules/cjs/loader:839:12)\n    at Module.require (node:internal/modules/cjs/loader:1028:19)\n    at require (node:internal/modules/cjs/helpers:102:18)\n    at start (/opt/iotaul/bin/iotagent-ul:41:18)\n    at Object.<anonymous> (/opt/iotaul/bin/iotagent-ul:56:1)\n    at Module._compile (node:internal/modules/cjs/loader:1126:14)\n    at Object.Module._extensions..js (node:internal/modules/cjs/loader:1180:10)\n    at Module.load (node:internal/modules/cjs/loader:1004:32)\n    at Function.Module._load (node:internal/modules/cjs/loader:839:12)\n    at Function.executeUserEntryPoint [as runMain] (node:internal/modules/run_main:81:12)\n    at node:internal/main/run_main_module:17:47 {\n  domainThrown: true\n}"}

Any suggestions on how to properly configure NGSIv2?

When deploying with helm and using an Ingress, various charts report the following warning:
annotation "kubernetes.io/ingress.class" is deprecated, please use 'spec.ingressClassName' instead

It should be possible to set this parameter instead of using annotations.

I am trying to deploy orion broker through helm and I am facing this issue :-

Error: INSTALLATION FAILED: template: orion/templates/test/orion-test.yaml:1:14: executing "orion/templates/test/orion-test.yaml" at <.Values.test.enabled>: nil pointer evaluating interface {}.enabled

The command which I am using to install it is :-
sudo helm repo add fiware https://fiware.github.io/helm-charts
sudo helm install fiware/orion --generate-name

I wasn't facing this issue earlier.
Can you please help me to resolve it?

Thanks & Regards,
Keshav

I'm deploying mongodb in ReplicaSet mode with a headless service, so my mongodb host is mongodb-headless. I can't set de RS value because of this restriction. It would be great to set RS independently to the number of hosts.

Currently, all PVCs in Keyrock only use the default-storage class. Thats unhandy for some more custom K8s clusters and should be configurable

Orion, Mongo and IoT-Agent have been deployed with the help of the Orion helm chart.

After creating a service, when trying to post a new device, I get the following message:

{"name":"ENTITY_GENERIC_ERROR","message":"Error accesing entity data for device: device1 of type: Thing"}

The logs from the IoT Agent look like this:

time=2022-03-01T10:53:42.928Z | lvl=ERROR | corr=1d437c8d-1c9b-4cba-838d-78ee85ff06eb | trans=1d437c8d-1c9b-4cba-838d-78ee85ff06eb | op=IoTAgentNGSI.Devices-v2 | from=n/a | srv=openiot | subsrv=/ | msg=Protocol error connecting to the Context Broker [500]: {
  error: 'InternalServerError',
  description: 'Database Error (collection: db-openiot.entities - count(): { _id.id: "urn:ngsi-ld:Device:1", _id.type: "Thing", _id.servicePath: { $in: [ null, /^/$/ ] } } - exception: count fails:{ ok: 0.0, errmsg: "not authorized on db-openiot to execute command { count: "entities", query: { _id.id: "urn:ngsi-ld:Device:1", _id.type: "Thing", _id.servicePath: {...", code: 13, codeName: "Unauthorized" })'
} | comp=IoTAgent

I have configured in the deviceRegistry.mongodb the user, password and authSource. I use Orion-LD (latest tag), with the following info:

orionld version: "post-v1.0.0"
orion version: "1.15.0-next"
uptime: "11 d, 18 h, 39 m, 43 s"
git_hash: "nogitversion"
compile_time: "Mon Jan 24 15:03:06 UTC 2022"
compiled_by: "root"
compiled_in: ""
release_date: "Mon Jan 24 15:03:06 UTC 2022"
doc: "https://fiware-orion.readthedocs.org/en/master/"

Is there some other configuration required?

in Keyrock helm chart v0.1.3/v0.1.4 enabling certs PVC with certs.pvc.enabled = true creates the PVC but then it is not mounted as volume in the pod if themes are not enabled at the same time.

It seems that the following code in the statefulset template prevents the certs PVC mount, i.e. the {{- if .Values.theme.enabled }} condition:

          {{- if .Values.theme.enabled }}
          volumeMounts:
            - name: themes-theme
              mountPath: {{ .Values.theme.mountTheme }}
            - name: themes-img
              mountPath: {{ .Values.theme.mountImg }}
            {{- if .Values.theme.sourceFonts }}
            - name: themes-fonts
              mountPath: {{ .Values.theme.mountFonts }}
            {{- end }}
            {{- if .Values.theme.sourceFavicon }}
            - name: themes-favicon
              mountPath: {{ .Values.theme.mountFavicon }}
              subPath: "favicon.ico"
            {{- end }}
            {{- if .Values.certs.pvc.enabled }}
            - name: certs
              mountPath: /opt/fiware-idm/certs
            {{- end }}
          {{- end }}

In order to deploy Canis Major to Kuberentes clusters, a Helm-Chart should be provided.
See: https://github.com/FIWARE-Ops/fiware-gitops/tree/master/aws/token/canis-major as a starting point.