I've been able to get all of my lab environment nodes to upgrade from 7.5.1 to 7.8.0 except for a single master node that got allocated onto transitional Allocator, one that isn't designated hot warm or frozen and now it wont move off of that Allocator. When I try to move it to the proper AZ I get the error below. It seems like if I use the failover.force=true it will allow this master node to move off of the Allocator. Overall trying to upgrade the deployment and move this node have failed every time.
[validate-plan-prerequisites]: [no.found.constructor.validation.ValidationException: 1. Unsupported downgrade attempted from [List(7.5.1, 7.8.0, 7.8.0, 7.8.0, 7.8.0, 7.8.0, 7.8.0, 7.5.1, 7.8.0, 7.8.0, 7.8.0, 7.5.1, 7.8.0, 7.5.1, 7.8.0)] to [7.5.1]. You can override this check with failover.force=true]"```
How are you moving it? Using the "Move" UI element for a single node should result in the flag
move_only being set (you can see in the "Details" section of the "Activity" page), which should avoid that error
If it isn't setting that flag (maybe if you're running an older ECE?), then I would apply the
move_instances by hand and set
move_only at the same time, via the advanced editor)
(The error itself is because you are declaring that you want the end state of the cluster to be 7.5.1 but it already has 7.8.0 nodes in it - the other option would be to manually up the
version to 7.8.0 in the advanced plan editor, together with the
move_instances ... that would have the advantage of fixing your cluster at the same time, but
move_only is the quickest way of getting your node off the allocator in question)
This topic was automatically closed 14 days after the last reply. New replies are no longer allowed.