search

IBM / Db2

Db2 Universal Container deployment
Apache License 2.0
9 stars 34 forks source link

Unable to apply db2 license #14

Closed xfabre closed 4 years ago

xfabre commented 4 years ago

with version 11.5.1.0-CN1, the <release-name>-db2u-0 pod never reaches a ready state. In the pod logs, I see:

[17987911.557092] db2u_root_entrypoint.sh[20]: + sudo /opt/ibm/db2/V11.5.0.0/adm/db2licm -a /db2u/license/db2u-lic
[17987911.646054] db2u_root_entrypoint.sh[20]: LIC1416N  The license could not be added automatically.  Return code: "-100".
[17987911.649329] db2u_root_entrypoint.sh[20]: + [[ 156 -ne 0 ]]
[17987911.649459] db2u_root_entrypoint.sh[20]: + echo '(*) Unable to apply db2 license.'
[17987911.649556] db2u_root_entrypoint.sh[20]: (*) Unable to apply db2 license.
[17987911.649648] db2u_root_entrypoint.sh[20]: + exit 1

I've accepted the license when I've run

./db2u-install --db-type db2oltp --namespace test-db2 --release-name testdb2 --tiller-namespace tiller --storage-class managed-nfs-storage

and the license is here in a configuration map with sensible dates

> kubectl get cm testdb2-db2u-lic -o yaml
apiVersion: v1
data:
  db2u-lic: |
    [LicenseCertificate]
    CheckSum=07DFC9F9DC7216727C19F837243E1B6E
<..>
    ProductName=IBM DB2 Developer-C Edition
    ProductID=1432
    ProductVersion=11.5
<...>
    LicenseStartDate=05/09/2019
    LicenseDuration=6812
    LicenseEndDate=12/31/2037
<...>
Doout commented 4 years ago

@xfabre was there any deployment of db2 before on your cluster? If v11.5.1.0 was deployed, it might have not clean up the semaphores it created — the easiest way to fix this will to reboot the systems.

xfabre commented 4 years ago

@Doout Yep, I'm in a situation where there is another db2 deployment with v11.5.1.0 in another namespace on the same cluster. If it's not a supported case, I'm fine, you can close the ticket. Thanks.

Doout commented 4 years ago

@xfabre At the moment we don't support more than one deployment on the same node. Is this a single node openshift cluster? If it is not you can try to reboot the node that db2 11.5.1.0-CN1 was deployed on and check if it faces the same issue.

You can check what node db2 was deployed one with oc get po -o wide

xfabre commented 4 years ago

@Doout It's a multi-node cluster but never-mind I can pick up a new one to install db2 11.5.1.0-CN1. Thanks for your explanations.