Details
-
Bug
-
Resolution: Fixed
-
Blocker
-
2.0-developer-preview-4
-
Security Level: Public
-
None
-
build 639, .deb
6 node cluster
Description
While loading data, and querying the view, an attempt to rebalance out 2 nodes fails. It is now impossible to add the node back to the cluster as the rest-api throws:
curl -u Administrator:password http://10.1.2.108:8091/nodes/self
"Node is unknown to this cluster."
If I go to the UI, the node shows 1 cluster that isn't itself (10.1.2.105), even though I expected it to be reinitialized.
I will leave this the state that it's in for the next hour or so if needed:
2012-02-03 06:29:54.360 menelaus_web:19:warning:server error during request processing(ns_1@10.1.2.104) - Server error during processing: ["web request failed",
,
,
{what,{case_clause,rebalance_running}},
{trace,
[
,
,
,
]}]
2012-02-03 06:29:54.431 ns_orchestrator:2:info:message(ns_1@10.1.2.104) - Rebalance exited with reason stopped
Attachments
For Gerrit Dashboard: MB-4760 | ||||||
---|---|---|---|---|---|---|
# | Subject | Branch | Project | Status | CR | V |
13015,1 | handle rebalance_running return from bucket deletion call | master | ns_server | Status: MERGED | +2 | +1 |
13017,1 | delete /master db first when deleting bucket. | master | ns_server | Status: MERGED | +2 | +1 |
13032,8 | Avoid deadlock when deleting/shutdown database | master | couchdb | Status: MERGED | +2 | +1 |
13067,4 | Fix indexing of multiple values for the same Key/DocID pair | master | couchdb | Status: MERGED | +2 | +1 |
13071,1 | don't wait for EXIT from cleaner when it's dead.MB-4760 | master | couchdb | Status: MERGED | +2 | +1 |
13278,1 | bp: handle rebalance_running return from bucket deletion call | branch-18 | ns_server | Status: MERGED | +2 | +1 |