Comment 10 for bug 1835637

Sorry about reviving this thread but we are experiencing this problem on a Rocky and a Stein installation.
I have found a work around that might work for you or not (or destroy something), ymmv "you have been warned".
I stop the rabbitmq services on all of the the controlplane nodes, and *IMPORTANT* restart them in the same order i.e the last one to go down is the first one to start up.
On to a question, how can we fix this issue in our installations.
Seem to me a rabbitmq + mnesia issue, could we just update those packages to versions that is not affected by this? Thanks in advance for any pointers!