Hi Sidrah,
Here is the description of one of the pending pods :
[root@icp4bsc-master3 InstallPackage]# kubectl describe pods discovery-watson-discovery-elastic-0
Name: discovery-watson-discovery-elastic-0
Namespace: default
Priority: 0
PriorityClassName: <none>
Node: <none>
Labels: app=watson-discovery
chart=elastic
component=elastic
controller-revision-hash=discovery-watson-discovery-elastic-dbdd68546
heritage=Tiller
release=discovery
statefulset.kubernetes.io/pod-name=discovery-watson-discovery-elastic-0
Annotations: kubernetes.io/psp: ibm-privileged-psp
productID: ICP4D-addon-5802bdd0bc424e2488698c0a8f82cfb8-discovery
productName: IBM Watson Discovery for IBM Cloud Private for Data
productVersion: 2.0.0
Status: Pending
IP:
Controlled By: StatefulSet/discovery-watson-discovery-elastic
Containers:
elastic:
Image: icp4bsc.nice.icp:8500/default/ibm-wex-ee:12.0.3.927
Ports: 9200/TCP, 9300/TCP
Host Ports: 0/TCP, 0/TCP
Limits:
cpu: 99
memory: 4Gi
Requests:
cpu: 750m
memory: 2Gi
Liveness: exec [bash -c ${WEX_HOME}/sbin/healthcheck.sh -w -t 5] delay=210s timeout=5s period=15s #success=1 #failure=3
Readiness: exec [bash -c ${WEX_HOME}/sbin/healthcheck.sh -w -t 5] delay=30s timeout=5s period=15s #success=1 #failure=12
Environment:
WEX_ALL_COMPONENTS: elastic
ELASTIC_REPLICAS: 1
ELASTIC_BACKUP_DIR: /backup-data
ELASTIC_ENDPOINT:
https://localhost:9200 ZEN_MARK_TLS: true
ELASTIC_CERT_FILE: /opt/tls/elastic/tls.crt
ELASTIC_KEY_FILE: /opt/tls/elastic/tls.key
ELASTIC_CACERT: /opt/tls/elastic/ca.crt
ELASTIC_USER: <set to the key 'username' in secret 'discovery-watson-discovery-elastic-secret'> Optional: false
ELASTIC_PASSWORD: <set to the key 'password' in secret 'discovery-watson-discovery-elastic-secret'> Optional: false
MANAGEMENT_SERVER_PORT: 9080
MANAGEMENT_SERVER_SECURE_PORT: 9443
MANAGEMENT_SERVER_ENDPOINT:
http://discovery-watson-discovery-gateway:$(MANAGEMENT_SERVER_PORT)/wex/api/v1 MANAGEMENT_SERVER_SECURE_ENDPOINT:
https://discovery-watson-discovery-gateway:$(MANAGEMENT_SERVER_SECURE_PORT)/wex/api/v1 RANKER_MODEL_MESH: discovery-watson-discovery-serve-ranker.default.svc.cluster.local:8033
TRAINING_DATA_CRUD: discovery-watson-discovery-training-data-crud.default.svc.cluster.local:50051
Mounts:
/backup-data from backup-data (rw)
/opt/tls/elastic from certificate (ro)
/var/run/secrets/kubernetes.io/serviceaccount from discovery-watson-discovery-shared-svc-acc-token-48g2k (ro)
/wexdata from data (rw)
Conditions:
Type Status
PodScheduled False
Volumes:
data:
Type: PersistentVolumeClaim (a reference to a PersistentVolumeClaim in the same namespace)
ClaimName: data-discovery-watson-discovery-elastic-0
ReadOnly: false
certificate:
Type: Secret (a volume populated by a Secret)
SecretName: discovery-watson-discovery-tls-secret
Optional: false
backup-data:
Type: PersistentVolumeClaim (a reference to a PersistentVolumeClaim in the same namespace)
ClaimName: discovery-watson-discovery-elastic-backup-data
ReadOnly: false
discovery-watson-discovery-shared-svc-acc-token-48g2k:
Type: Secret (a volume populated by a Secret)
SecretName: discovery-watson-discovery-shared-svc-acc-token-48g2k
Optional: false
QoS Class: Burstable
Node-Selectors: <none>
Tolerations: node.kubernetes.io/memory-pressure:NoSchedule
node.kubernetes.io/not-ready:NoExecute for 300s
node.kubernetes.io/unreachable:NoExecute for 300s
Events: <none>
[root@icp4bsc-master3 InstallPackage]#
Thanks for your help !
------------------------------
Valerie Le Roy
------------------------------
Original Message:
Sent: Fri September 06, 2019 10:57 AM
From: Sidrah Junaid
Subject: IBM Watson Discovery add-on installation on Cloud Pak for Data
Hi Valerie,
Can you please share the description of one of the Pending pod?
kubectl describe pods <pod-name> -n <namespace>
------------------------------
Sidrah Junaid
Original Message:
Sent: Fri September 06, 2019 02:53 AM
From: Valerie Le Roy
Subject: IBM Watson Discovery add-on installation on Cloud Pak for Data
Hi Sidrah,
Again, thanks a lot for your answer !
Please find hereafter the results of the kubectl get pods command:
Thanks for your support !
I also ran again the command by using mycluster.icp for the hostname and IP address of the master (after the appropriate modifications in the /etc/hosts file). I got again the Error: watch closed before Until timeout error message.
------------------------------
Valerie Le Roy
Original Message:
Sent: Thu September 05, 2019 11:42 AM
From: Sidrah Junaid
Subject: IBM Watson Discovery add-on installation on Cloud Pak for Data
Hi Valerie,
MasterHostname should be mycluster.icp. masterIP should be the master hostIP (not the load balancer).
In addition to that, could you please run
kubectl get pods -l release=${name}
using the same value for name as you used in that install command and share the result .
------------------------------
Sidrah Junaid
Original Message:
Sent: Thu September 05, 2019 08:52 AM
From: Valerie Le Roy
Subject: IBM Watson Discovery add-on installation on Cloud Pak for Data
Hi Sidrah,
Thanks a lot for your answer !
I confirm that some resources were created after the helm install command. So I deleted them by using the commands you provided.
Then I ran again the install command with the timeout parameter:
helm install --name ${name} zenrepos/ibm-watson-discovery-prod --tls --set global.license=accept --set core.common.persistence.storageClassName=oketi-nfs --set core.ingestion.mount.storageClassName=oketi-nfs --set elastic.persistence.storageClassName=oketi-nfs --set etcd.persistence.storageClassName=oketi-nfs --set hdp.persistence.storageClassName=oketi-nfs --set minio.persistence.storageClass=oketi-nfs --set postgresql.persistence.storageClassName=oketi-nfs --set elastic.backup.persistence.storageClassName=oketi-nfs --set global.icp.masterHostname=icp4bsc.nice.icp --set global.icp.masterIP=192.168.66.150 --timeout 1200
I got the following error message : Error: watch closed before Until timeout
Please note that I have a high-availability configuration, with 3 master nodes. So the global.icp.masterIP parameter is set to the IP address of the cluster VIP. And in the /etc/hosts file, I have added the following entry (as described in the installation procedure) :
192.168.66.150 mycluster.icp
------------------------------
Valerie Le Roy
Original Message:
Sent: Wed September 04, 2019 11:07 AM
From: Sidrah Junaid
Subject: IBM Watson Discovery add-on installation on Cloud Pak for Data
Hi Valerie,
Could you please check if there is any pod, deployment and job created for that chart
kubectl get --namespace=<my-namespace> configmaps,jobs,pods,statefulsets,deployments,roles,rolebindings,secrets,serviceaccounts --selector=release=<my-release>
I am curious if it's just a time-out issue where the install just took too long and so the post-install jobs never ran. So, for that we can add --timeout 1200 in helm install command.
So first clean up the previous install by running the following commands in order:
helm delete --purge --tls <my-release>
kubectl delete --namespace=<my-namespace> configmaps,jobs,pods,statefulsets,deployments,roles,rolebindings,secrets,serviceaccounts --selector=release=<my-release>
kubectl delete --namespace=<my-namespace> persistentvolumeclaims --selector=release=<my-release>
Remove persistent volumes ,
kubectl delete persistentvolumes $(kubectl get persistentvolumes \
--output=jsonpath='{range .items[*]}{@.metadata.name}:{@.status.phase}:{@.spec.claimRef.name}{"\n"}{end}' \
| grep ":Released:" \
| grep "<my-release>-" \
| cut -d ':' -f 1)
------------------------------
Sidrah Junaid
Original Message:
Sent: Wed September 04, 2019 09:36 AM
From: Valerie Le Roy
Subject: IBM Watson Discovery add-on installation on Cloud Pak for Data
Hello,
I have a Cloud Pak for Data V2.1 installation and I would like to install the Watson Discovery V2.0.0 add-on.
I have followed the instructions found at https://www.ibm.com/support/knowledgecenter/en/SSQNUZ_2.1.0/com.ibm.icpdata.doc/watson/discovery-install.html
I followed each step of the installation procedure. Everything was OK, until the Helm installation command.
Here is the command I ran :
helm install --name ${name} zenrepos/ibm-watson-discovery-prod --tls --set global.license=accept --set core.common.persistence.storageClassName=oketi-nfs --set core.ingestion.mount.storageClassName=oketi-nfs --set elastic.persistence.storageClassName=oketi-nfs --set etcd.persistence.storageClassName=oketi-nfs --set hdp.persistence.storageClassName=oketi-nfs --set minio.persistence.storageClass=oketi-nfs --set postgresql.persistence.storageClassName=oketi-nfs --set elastic.backup.persistence.storageClassName=oketi-nfs --set global.icp.masterHostname=icp4bsc.nice.icp --set global.icp.masterIP=192.168.66.150
(I initialized the name variable to discovery)
Here is the error I get : Error: timed out waiting for the condition
Would you have an idea on how to avoid this error ?
I could not find any information related to this error message in the documentation.
Many thanks in advance for any help on this !
------------------------------
Valerie Le Roy
------------------------------
#CloudPakforDataGroup