Couchbase
Table of Contents
Overview
Reference
internalSettings
$ curl -u Administrator:password -v http://localhost:8091/internalSettings
{
"indexAwareRebalanceDisabled": false,
"rebalanceIndexWaitingDisabled": false,
"rebalanceIndexPausingDisabled": false,
"rebalanceIgnoreViewCompactions": false,
"rebalanceMovesPerNode": 1,
"rebalanceMovesBeforeCompaction": 64,
"maxParallelIndexers": 4,
"maxParallelReplicaIndexers": 2,
"maxBucketCount": 10,
"gotraceback": "crash",
"indexAutoFailoverDisabled": true,
"certUseSha1": false
}
Terminology
- Node
- A single Couchbase Server instance
- Cluster
- A collection of nodes that are accessed and managed as a single group.
- Bucket
- A logical container for a related set of items
- Item
- The basic unit of data such as JSON documents
- vBucket
- Physical partitions of the bucket data.
- Cluster map
- A mapping of which services belong to which nodes at a given point in time.
- vBucket map
- A mapping of vBuckets to nodes at a given point in time.
- Replication
- The process of creating additional copies of active data on alternate nodes.
- Rebalance
- Used to redistribute the load and adapt to the new topology of nodes.
- Failover
To divert traffic away from failing nodes to the remaining healthy nodes.
- Node lifecycle
Topics
Document Expiration
TTL
in seconds if it's 30 days or lessTTL
in Unix time if it's over 30 days- Couchbase runs a process named
expiry pager
periodically to clean up expired items Expire pager
runs every60
minutes by default(can be configured through exppagerstime)
How-to
Failover
- Click
Fail Over
button next to the failing over node - Select
Fail Over
strategy
- After failing over process, the node becomes pending.
- After shutting down the node
- After putting back the node, select one of recovery strategy and click
Rebalance
.