MQSeries.net :: View topic - Advice on moving queue managers to cloud environment

jamesb

Hi all,
We've been running our queue managers on 2 node hardware clusters for many years, typically in an active passive configuration where we have a common set of disks that are mounted on the active node and and then get moved across and mounted up on the other node when it fails over due to hardware failure, or more likely patching. All are MQ client apps, running either in Kubernetes or standalone hosts.

Our company has a strategy of migrating from our own data centres to AWS cloud for hosting and and I'm looking at the options available to me. i.e should I try to move the 2 node cluster config as it, or is this now outdated and I need to move the MQ servers to containers or something more cloud-friendly? We have a Kubernetes environment here.

I'd be interested in hearing from anyone here what you company may have done in this situation, anonymously, of course. My thoughts are:

Does this sound like a situation where our MQ servers should be moved to Kubernetes and in the event of a failure, MQ would be restarted in another node?

Should it be moved to a container I build myself?

Should it be moved to an IBM certified container? I'd prefer not to get into the process to buy more IBM kit which will be a big hassle for me, even if I can get it approved.

Your thoughts would be appreciated.

Cheers, James.