You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
Now that we have a uniquely identifying Cluster ID (gh-6702), we could use it to prevent the joining of members participatory of another cluster. When a member is [re]joined, it should either have no existing Cluster ID, or a matching Cluster ID to the cluster it is being joined with. A non-matching Cluster ID is an indication that the member was participating in another cluster, and will have a non-sensical raft log & state store, likely causing some problems.
The text was updated successfully, but these errors were encountered:
Hey @adhaamehab , of course contributions are welcome!
This particular feature is definitely going to require some spelunking through Nomad's source as it ties together elements of the CLI, API, Gossip, Raft, and leadership. It also needs to be well tested at each level to avoid breaking any existing functionality. To get started though, there's no better place than with the join CLI command, which is kind of the entry point of the request flow.
Now that we have a uniquely identifying Cluster ID (gh-6702), we could use it to prevent the joining of members participatory of another cluster. When a member is
[re]join
ed, it should either have no existing Cluster ID, or a matching Cluster ID to the cluster it is being joined with. A non-matching Cluster ID is an indication that the member was participating in another cluster, and will have a non-sensical raft log & state store, likely causing some problems.The text was updated successfully, but these errors were encountered: