-
Notifications
You must be signed in to change notification settings - Fork 1.2k
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
Cannot replace a node after changing its IP address #13775
Comments
I believe that the issue is pre-existing. |
Also, we need to test this was raft topology operations enabled. |
should @kostja and team own this? |
@kostja can you guys please own this issue? |
@bhalevy is dtest running with consisten_cluster_management: true? |
Copying my comment from the other PR: The exception comes from I guess it regressed when we started reusing Host ID of replaced node. A potential fix would be to resolve the Host ID conflict inside |
But we stopped reusing the replaced node id. |
We do need to resolve the conflict by dropping the endpoint state of the node that changed its ip address. |
Right, sorry, brain fart. |
When a host changes its ip address we should force remove the previous endpoint state since we want only one endpoint to refer to this host_id. If the new node that changed its ip address is decommissioned, the previous node seems as a normal token owner, just in shutdown status, but it is not longer in the cluster. Refs scylladb#14468 Fixes scylladb#13775 Signed-off-by: Benny Halevy <bhalevy@scylladb.com>
When a host changes its ip address we should force remove the previous endpoint state since we want only one endpoint to refer to this host_id. If the new node that changed its ip address is decommissioned, the previous node seems as a normal token owner, just in shutdown status, but it is not longer in the cluster. Refs scylladb#14468 Fixes scylladb#13775 Signed-off-by: Benny Halevy <bhalevy@scylladb.com>
When a host changes its ip address we should force remove the previous endpoint state since we want only one endpoint to refer to this host_id. If the new node that changed its ip address is decommissioned, the previous node seems as a normal token owner, just in shutdown status, but it is not longer in the cluster. Fixes scylladb#13775 Signed-off-by: Benny Halevy <bhalevy@scylladb.com>
with this PR the test passes with |
Nice. I'm moving this issue out of the raft topology required backlog then, as the issue remains in gossiper-mode but is not a blocker for raft-topology |
How about the dtest for this issue? |
cc @temichus @aleksbykov |
ping |
I've created pr https://github.com/scylladb/scylla-dtest/pull/4043 |
This is a follow up on #13066
The following dtest reproduces the issue:
node4 fails to start, reporting:
The text was updated successfully, but these errors were encountered: