[dev] always use the first node to generate descriptions

Some clusters are heterogeneous. We sort nodes before picking the first node to avoid cases where we would use different nodes to generate descriptions. Apparently different ruby versions sort hashes differently.

assigned to @lnussbau

added 1 commit

• 6c23f734 - [dev] use split_cluster_node everywhere relevant to avoid duplicating this logic

Compare with previous version

approved this merge request

added 1 commit

• 7c81037e - [dev] Improve extraction to handle case of gros-1 and gros1-18 for example

Compare with previous version

resolved all threads

enabled an automatic merge when all merge checks for 7c81037e pass

merged

mentioned in commit 05bab805