Details
-
Bug
-
Resolution: Duplicate
-
Critical
-
5.5.0
-
E5-2680 v3 (48 vCPU)
64 GB
Samsung PM863a (SATA SSD)
-
Untriaged
-
Centos 64-bit
-
-
No
Description
Test case:
- 7 nodes
- 1 bucket, 1 replica
- 10B items (~10TB before compression)
- 10K ops/sec (read-heavy), 10% cache miss ratio
- Swap rebalance of one of the nodes (172.23.96.108 -> 172.23.96.109)
http://cbmonitor.sc.couchbase.com/reports/html/?snapshot=titan_510-1323_rebalance_d902
I tried to restart rebalance after the test but it failed:
Rebalance exited with reason {{badmatch,
|
{error,
|
{failed_nodes,['ns_1@172.23.96.109']}}},
|
[{ns_janitor,cleanup_with_states,6,
|
[{file,"src/ns_janitor.erl"},{line,136}]},
|
{ns_rebalancer,do_run_janitor_pre_rebalance,1,
|
[{file,"src/ns_rebalancer.erl"},
|
{line,766}]}]}
|
Attachments
Issue Links
- duplicates
-
MB-26633 memcached crash during initial data load (10TB)
-
- Closed
-