Multiinstance Qmgr - Standby Qmgr not existing

[schmiuwe]

Hi Callum,

we tried deploy a multiinstance qmgr on Azure. Used included sample configuration and deployment works fine, however the second pod does not seem to have a qmgr installed, also failover testing did not really work. Is this the way it is intended to be?

Please see here:

> kubectl get pods -n t01n03a
NAME               READY   STATUS    RESTARTS   AGE
t01n03a-ibm-mq-0   1/1     Running   0          11m
t01n03a-ibm-mq-1   0/1     Running   0          38s

> kubectl exec --stdin --tty t01n03a-ibm-mq-0 --namespace t01n03a -- /bin/bash
bash-4.4$ runmqsc
5724-H72 (C) Copyright IBM Corp. 1994, 2024.
Starting MQSC for queue manager T01N03A.

No MQSC commands read.
bash-4.4$ exit
exit

> kubectl exec --stdin --tty t01n03a-ibm-mq-1 --namespace t01n03a -- /bin/bash
bash-4.4$ runmqsc
5724-H72 (C) Copyright IBM Corp. 1994, 2024.
AMQ8146E: IBM MQ queue manager not available.

No MQSC commands read.
bash-4.4$ exit
exit
command terminated with exit code 20

> kubectl delete pod t01n03a-ibm-mq-0 -n t01n03a
pod "t01n03a-ibm-mq-0" deleted
> kubectl get pods -n t01n03a
NAME               READY   STATUS    RESTARTS   AGE
t01n03a-ibm-mq-0   0/1     Running   0          45s
t01n03a-ibm-mq-1   1/1     Running   0          6m34s

> kubectl exec --stdin --tty t01n03a-ibm-mq-0 --namespace t01n03a -- /bin/bash
bash-4.4$ runmqsc
5724-H72 (C) Copyright IBM Corp. 1994, 2024.
AMQ8478E: Standby queue manager.

No MQSC commands read.
bash-4.4$ exit
exit
command terminated with exit code 20

> kubectl exec --stdin --tty t01n03a-ibm-mq-1 --namespace t01n03a -- /bin/bash
bash-4.4$ runmqsc
5724-H72 (C) Copyright IBM Corp. 1994, 2024.
AMQ8146E: IBM MQ queue manager not available.

No MQSC commands read.
bash-4.4$ exit
exit
command terminated with exit code 20

> kubectl get pvc -n t01n03a
NAME                    STATUS   VOLUME                                     CAPACITY   ACCESS MODES   STORAGECLASS   AGE
data-t01n03a-ibm-mq     Bound    pvc-385e0594-a6a8-4ccc-b298-5c6730cdc94d   32Gi       RWX            mq-azurefile   24m
log-t01n03a-ibm-mq      Bound    pvc-6695dc64-168d-479f-8480-2517748e3ddb   32Gi       RWX            mq-azurefile   24m
qmgr-t01n03a-ibm-mq-0   Bound    pvc-82eb47b3-a736-451d-8a23-a4f342e96b67   32Gi       RWO            managed        24m
qmgr-t01n03a-ibm-mq-1   Bound    pvc-e8dcf5cb-8430-4461-a2bc-0a2f44605e1a   32Gi       RWO            managed        23m

Thank you, Uwe

[arthurbarr]

Based on the name of your StorageClass, I wonder if you are using Azure Files directly. Azure Files does not implement file locking well, and is explicitly not supported with IBM MQ. You can use it via NFS V4.1, but not via the default mechanisms. Can you please confirm the filesystem type?

[schmiuwe]

Hi Arthur,

we just used the configuration out of the provided sample on GitHub, please see here:

https://github.com/ibm-messaging/mq-helm/tree/main/samples/AzureAKSMultiInstance/deploy

• file: customStorageClass.yaml

For us the mount options in this file where needed/intesting to know ...

I am not sure what and where to configure differently, do you know?

Thank you, Uwe

From: Arthur Barr @.> Sent: Tuesday, April 9, 2024 4:35 PM To: ibm-messaging/mq-helm @.> Cc: Schmiedel Uwe, FG-232 @.>; Author @.> Subject: Re: [ibm-messaging/mq-helm] Multiinstance Qmgr - Standby Qmgr not existing (Issue #79)

Sent from outside the BMW organization - be CAUTIOUS, particularly with links and attachments. Absender außerhalb der BMW Organisation - Bitte VORSICHT beim Öffnen von Links und Anhängen.

---

Based on the name of your StorageClass, I wonder if you are using Azure Files directly. Azure Files does not implement file locking well, and is explicitly not supportedhttps://www.ibm.com/support/pages/testing-statement-ibm-mq-multi-instance-queue-manager-file-systems with IBM MQ. You can use it via NFS V4.1, but not via the default mechanisms. Can you please confirm the filesystem type?

- Reply to this email directly, view it on GitHubhttps://github.com/ibm-messaging/mq-helm/issues/79#issuecomment-2045337994, or unsubscribehttps://github.com/notifications/unsubscribe-auth/A6UGXEKIXVAX5LINTLGRWH3Y4P37TAVCNFSM6AAAAABF6MIGASVHI2DSMVQWIX3LMV43OSLTON2WKQ3PNVWWK3TUHMZDANBVGMZTOOJZGQ. You are receiving this because you authored the thread.Message ID: @.**@.>>

[arthurbarr]

I'd suggest using dspmq for your liveness check, as that might give you back more information. Additionally, you can refer to the queue manager's error logs (and any FDCs) to determine if the secondary instance is attempting to start, and what's going on if so. If there's not an obvious error about locking or similar (which might indicate a change in Azure Files), then I suggest that the best option would be to raise a support ticket including the output of runmqras.

[callumpjackson]

Closing due to inactivity and answers provided by Arthur.