Description
When running a rebalance, the XDCR tab in the UI shows a number of errors (see attached screenshot below). However as rebalance proceeds, some of these errors disappear to be replaced later with new ones. The advertising and subsequent disappearance of these "Errors" can be very confusing ("is something broken or not?"). I think we should either take efforts not to show such temporary errors or, if they do become resolved, inform the user that they have been rectified automatically.
Logs:
https://s3.amazonaws.com/cb-engineering/davidH/collectinfo-2017-06-27T125416-n_0%40192.168.4.39.zip
https://s3.amazonaws.com/cb-engineering/davidH/collectinfo-2017-06-27T125416-n_1%40127.0.0.1.zip
https://s3.amazonaws.com/cb-engineering/davidH/collectinfo-2017-06-27T125416-n_2%40127.0.0.1.zip
https://s3.amazonaws.com/cb-engineering/davidH/collectinfo-2017-06-27T125416-n_3%40127.0.0.1.zip
Experienced via creating a 4 node cluster run, 2 ephemeral buckets and a loopback xdcr from one bucket to the other then running perpetual rebalances.
Example errors:
Attachments
Issue Links
- is duplicated by
-
MB-15158 GoXDCR tries to start replication on new node before rebalance-in is complete ("Invalid configuration. No source nozzle can be constructed...")
-
- Closed
-