IBM Integration Community Come for answers. Stay for best practices. All we’re missing is you. Join / Log in Ask a question
I have couple of questions about RDQM Cluster Setup;
1) Is Korn shell still a requirement to install and configure RDQM? and2) When I test the failover in RDQM environment using MQ Explorer, I get disconnected when I test the failover by suspending the queuemanager on one of the node or power-off one node. I had to connect again in MQ Explorer - it connects no issue with that. Is there setting that can be configured so that it stays connected/re-connect as soon as queuemanger failed over to next node.Thank you,Regards,
You can use Ansible to install RDQM.First, when a queue manager is failed over in HA; there is a restart happening. So the behaviour you see of disconenct and connect is working as expected.It is recommended to use CCDT for client connection; and setup the auto-reconnect option.This way. you would not need to connect again manually.MQ Exproer has option to connect using CCDT; explore the option.
Hello Om Prakash, Thank you very much.Regards,Rajesh
Hello Om Prakash, I am unable to create the RDQM queuemanager. After creating the cluster using rdqmadm -c, when I try to create the queuemanager on the secondary node;[root@txulmqdv2 sbin]# /opt/mqm/bin/crtmqm -sxs -fs 16384M QM.MQTAMQ3818E: The replicated data subsystem has not been configured.AMQ3812E: Failed to create replicated data queue manager configuration.[root@txulmqdv2 sbin]#
Please advise how to fix the drbd key issue.
Thank you.RajeshIt might have to do with drbd module, the lsmod|grep drbd command does not show anything. When I try to load modprobe drbd, I get the following error message;
[root@dv2 sbin]# modprobe drbdmodprobe: ERROR: could not insert 'drbd': Required key not available
U r missing few required steps.
and then finally create the queue manager. You issue command on a single node, and secondary nodes will auto get teh config.follow these steps:https://www.ibm.com/docs/en/ibm-mq/9.2?topic=multiplatforms-installing-rdqm-replicated-data-queue-managersthis is for mq 9.2; but 9.3 also is same.
Did it work? What actions were taken after validating.
Hello Om Prakash,No it did not work.Thanks.
This is not specific to RDQM but on Linux, the primary group of the mqm user must be the mqm group. See https://www.ibm.com/docs/en/ibm-mq/9.3?topic=linux-setting-up-user-group
Hi John, Thank you very much for the link, I already provision this as I followed the iBM guidelines/steps.Rajesh
Hi Om Prakash, As you followed up on this with me so I am updating you that I am able to create RDQM Queuemanager.
I had a ticket opened with IBM last week, but there was no response so I had to escalate it on Wednesday so yesterday I've got a chance to show the Support Specialist (Robert) and we worked it out. He told me that he has never seen this kind of situation so he has invited many of his team member while I share the screen. So this is what the issue, VMWare cluster was upgraded recently at my work and since I worked all my life on AIX so was not aware and Robert said he also has not see this. The VMWare upgrade made the secure-boot enabled on VMs being created on the VMware cluster. So in order for drbd to work, we need an extra step to enroll the keys otherwise it fails to start the drbd service while creating the cluster. Once we did that then I was able to create the cluster/queuemanager. Now I am able to connect to queuemanager from MQ Explorer as well.
Thank you for following up once again.Rajesh
Hi Om, I had configured the RDQM successfully last week. Today while testing, I suspended the Queuemanager on all the three nodes using rdqmadm command, now I am unable to start the queuemanager. All the resources are in STOPPED state, please help;
You should not suspend all services across the nodes.Start them one by one, and also issue crm_resource --cleanup and see if it works.