Details
Major Description
Summary:
Rebalance button doesn't get disabled even after rebalancing the cluster many times as seen in the video. I think it is an indexer issue as I see some error messages related to indexer in error.log
Steps:
1. I initially had a 1 node KV cluster (.215). I enabled enforce-tls and disabled it a few times, and finally turned off the n2n encryption
2. Added .217 node with all services on it. Rebalance finished, but the cluster never got balanced.
+Observations
+on .217 error.log
[ns_server:error,2021-08-17T20:36:59.042-07:00,ns_1@172.23.105.217:service_status_keeper_worker<0.284.6>:rest_utils:get_json:57]Request to (indexer) getIndexStatus with headers [] failed: {ok,
|
{{500,
|
"Internal Server Error"},
|
[{"Content-Length",
|
"122"},
|
{"Date",
|
"Wed, 18 Aug 2021 03:36:59 GMT"},
|
{"Content-Type",
|
"application/json"}],
|
<<"{\"code\":\"error\",\"error\":\"Fail to retrieve cluster-wide metadata from index service\",\"failedNodes\":[\"172.23.105.217:8091\"]}">>}}
|
[ns_server:error,2021-08-17T20:36:59.042-07:00,ns_1@172.23.105.217:service_status_keeper-index<0.285.6>:service_status_keeper:handle_cast:103]Service service_index returned incorrect status
|
Had seen a similar issue with FTS some time ago
https://issues.couchbase.com/browse/MB-46424
But this doesn't seem to be an FTS issue since the workaround mentioned on that bug didn't help with this cluster
Attachments
| For Gerrit Dashboard: MB-48001 | ||||||
|---|---|---|---|---|---|---|
| # | Subject | Branch | Project | Status | CR | V |
| 163613,4 | MB-48001 Delay registering for topology awareness until part of cluster. | master | query | Status: MERGED | +2 | +1 |