IEM Pro Install errors

0
Dear Community,    i tried to install IEM-Pro with K3s on a Ubuntu WSL2 instance. Ubuntu: 24.04 Industrial Edge Provisioning CLI - Ubuntu: 1.0.29 - Latest   I followed the instructions in the IEM-Pro documentation but encountered some issues during the setup. I tried both options for the installation:  ieprovision CLI direct helm   The ieprovision app finished successfully with the following output: NOTES: CHART NAME: application-management-service CHART VERSION: v1.14.9 APP VERSION: v1.14.9 ** Please be patient while the chart is being deployed ** IEM: https://local.iem/ NAMESPACE: iem-anlagenmodell KEYCLOAK_USER: iem_user KEYCLOAK_USER_PASSWORD: *** KEYCLOAK_CUSTOMER_REALM_ADMIN: *** KEYCLOAK_CUSTOMER_REALM_ADMIN_PASSWORD: *** INFO[0692] Installation complete Helm chart deployed   However, the IEM is not reachable at `https://local.iem/`.   A deeper look into the pods shows that some pods are not running. The problematic pods are: wfx-c7bd888b4-pgql6 0/1 CrashLoopBackOff wfx-qx-594dc887c9-klf49 0/1 CrashLoopBackOff   The logs of these pods shows that the postgresql database is not reachable. My next step was to check the status of the postgresql pod, because this pod was listed as running. The analysis of the postgresql pod shows that the different databases are not successfully created. INFO ==> ** Starting PostgreSQL ** pgaudit extension initialized starting PostgreSQL 14.14 on x86_64-pc-linux-gnu, compiled by gcc (Debian 12.2.0-14) 12.2.0, 64-bit listening on IPv4 address "0.0.0.0", port 5432 listening on IPv6 address "::", port 5432 listening on Unix socket "/tmp/.s.PGSQL.5432" LOG: database system was shut down at 2025-08-04 09:36:55 GMT LOG: database system is ready to accept connections FATAL: database "platformportal" does not exist FATAL: database "authdb" does not exist ERROR: relation "migration_model" does not exist at character 25 STATEMENT: SELECT ID, VERSION FROM MIGRATION_MODEL ORDER BY UPDATE_TIME DESC ERROR: relation "public.databasechangelog" does not exist at character 22 STATEMENT: SELECT COUNT(*) FROM public.databasechangelog ERROR: relation "public.databasechangeloglock" does not exist at character 22 STATEMENT: SELECT COUNT(*) FROM public.databasechangeloglock ERROR: relation "public.databasechangelog" does not exist at character 22 STATEMENT: SELECT COUNT(*) FROM public.databasechangelog ERROR: database "platformportal" already exists STATEMENT: CREATE DATABASE platformportal FATAL: database "ams" does not exist I would expect that the postgresql pod is responsible for creating the databases and tables, but it seems that this is not happening.   Does anyone know if there is a troubleshooting guide for the Industrial Edge Platform? Does anyone have an idea how to fix this issue? Thanks in advance for your help!
asked
4 answers
0

Hello Oliver,

 

Could I ask for more information?

- Status of other objects in the cluster: PVCs, replicaSets, etc

- Was the ingress applied?

- The gateway pod is running, correct? All other pods are fine, from what I understand.

answered
0

Hello Carlos,

 

thank you for your quick response and helpful hints!

 

Regarding the ingress: You were absolutely right! The ingress-nginx-controller pod was indeed not running. I've increased the allocated memory, and it is now running correctly.

 

Here is the current status of all pods in the cluster:

NAME                                            READY   STATUS             RESTARTS      
app-config-service-6f858dc4b7-l29vd             1/1     Running            1 (145m ago)  
app-manager-768cfd7b65-sr62g                    1/1     Running            1 (145m ago)  
auth-service-7dd4967b85-mcfn9                   1/1     Running            5 (145m ago)  
auth-service-v2-775449d4b9-cqxrg                1/1     Running            1 (145m ago)  
chartmuseum-0                                   1/1     Running            1 (145m ago)  
edge-manager-0                                  1/1     Running            2 (145m ago)  
edgeeye-config-service-0                        1/1     Running            1 (145m ago)  
edgeeye-fluentbit-collector-6kqq2               0/1     Init:0/1           0             
edgeeye-fluentbit-watcher-7b676fd4cc-q8bht      0/1     Unknown            0             
firmwaremanagement-minio-76459c6f8-v6dqc        1/1     Running            1 (145m ago)  
firmwaremanager-ui-bdbf6dc7d-vzbjm              1/1     Running            1 (145m ago)  
ie2328-gateway-56885c9768-9x9cz                 2/2     Running            2 (145m ago)  
ie2328-gateway-init-migrations-wdj9c            0/1     Completed          0             
iels-deployment-7957565c69-2gpwl                1/1     Running            1 (145m ago)  
iess-s3-statefulset-0                           5/5     Running            5 (145m ago)  
iess-statefulset-0                              1/1     Running            1 (145m ago)  
influxdb-0                                      1/1     Running            1 (145m ago)  
ingress-nginx-controller-klcxs                  1/1     Running            0             
keycloak-0                                      1/1     Running            1 (145m ago)  
launchpad-c499cbf-hgnws                         1/1     Running            1 (145m ago)  
notification-service-0                          1/1     Running            1 (145m ago)  
onpremisedevicetypemanagement-cc4cf8746-c6mjz   0/1     Init:1/2           0             
onpremisefirmwaremanagement-68cb6b4845-8l9r2    0/1     Init:0/3           0             
portal-service-0                                1/1     Running            1 (145m ago)  
portal-ui-78c546bbcf-5fs7v                      1/1     Running            1 (145m ago)  
portal-wss-7b5d456df4-568fp                     1/1     Running            1 (145m ago)  
portalhub-0                                     1/1     Running            1 (145m ago)  
postgres-0                                      1/1     Running            1 (145m ago)  
postgres-iedevicecatalog-0                      1/1     Running            1 (145m ago)  
regsecgen-29242680-bz4cf                        0/1     Completed          0             
tunnel-server-0                                 1/1     Running            1 (145m ago)  
twin-service-546fbdb579-mrdl9                   0/1     Init:0/1           0             
twin-service-546fbdb579-rvr9t                   0/1     Init:0/1           0             
wfx-7cdddb8fc4-j72m6                            0/1     CrashLoopBackOff   68 (4m7s ago) 
wfx-qx-54b958c76f-lfrbz                         0/1     CrashLoopBackOff   56 (3m50s ago)
workflowexecutor-6db5f4984d-qzpnv               1/1     Running            4 (35m ago)

As you can see, the ie2328-gateway pod is running (2/2 Ready), and my IEM Instance is indeed shown as online in the Hub.

 

However, the core issue persists: the wfx pods (wfx-7cdddb8fc4-j72m6 and wfx-qx-54b958c76f-lfrbz) are still in a CrashLoopBackOff state. Their logs continue to show problems connecting to the PostgreSQL databases.

For example, a typical error from their logs is:

{"level":"error","module":"pgx","err":"failed to connect to `host=postgres-iedevicecatalog.iem-anlagenmodell.svc.cluster.local user=postgres database=wfx`: hostname resolving error (lookup postgres-iedevicecatalog.iem-anlagenmodell.svc.cluster.local: Try again)","time":"2025-08-07T13:27:40Z","caller":"github.com/jackc/pgx/v4@v4.14.1/log/zerologadapter/adapter.go:100","message":"connect failed"}
{"level":"error","error":"failed to connect to `host=postgres-iedevicecatalog.iem-anlagenmodell.svc.cluster.local user=postgres database=wfx`: hostname resolving error (lookup postgres-iedevicecatalog.iem-anlagenmodell.svc.cluster.local: Try again)","time":"2025-08-07T13:27:40Z","caller":"code.siemens.com/swupdate/wfx/cmd/wfx-server/main.go:132","message":"Failed to create backend"}
{"level":"fatal","error":"failed to connect to `host=postgres-iedevicecatalog.iem-anlagenmodell.svc.cluster.local user=postgres database=wfx`: hostname resolving error (lookup postgres-iedevicecatalog.iem-anlagenmodell.svc.cluster.local: Try again)","time":"2025-08-07T13:27:40Z","caller":"code.siemens.com/swupdate/wfx/cmd/wfx-server/main.go:104","message":"Fatal error, bailing out"}

 

Do you have any further suggestions on how to troubleshoot the database initialization process within the PostgreSQL pod, or how to ensure the wfx pods can correctly connect and migrate their schemas?

 

Thanks again for your continued support!

answered
0

Hello Oliver,

 

Good to know that it has improved.

 

Regarding your other points, could I ask that you verify that all the general requirements are met?

https://docs.eu1.edge.siemens.cloud/get_started_and_operate/industrial_edge_management/how_to_setup_operate/pro/overview/general-reqs.html

 

I suspect that pods might be getting stuck due to a lack of resources.

 

Other than that, could you also check the logs for the other problematic pods? What type of error do they display?

answered
0

Hello Carlos,

 

Thanks for the link. My device does meet the general requirements, but I'm not entirely sure if these are actually being applied in the k3s environment.

To verify, I re-ran the installation using Minikube instead of k3s, and in that setup everything worked successfully.

To me, it looks like this was a specific issue with my environment when using k3s.Thanks a lot for your support!

 

Best regards,

Oliver

answered