A caching layer supporting Redis and filesystem caching, used in core DADI products
Removing the complexity involved in setting up two separate cache handlers for every project, DADI Cache can use either the filesystem or a Redis instance to store and retrieve content.
npm install @dadi/cache --save// require the moduleconstCache=require('@dadi/cache')// setup the options for caching// defaults if nothing specified: // {// directory: {// enabled: true,// path: './cache'// },// redis: {// enabled: false// }// }constoptions={"ttl": 3600,"directory": {"enabled": false,"path": "./cache/"},"redis": {"enabled": true,"host": "127.0.0.1","port": 6379}}constcache=newCache(options)Returns a Promise that returns an empty String if successful, otherwise an Error.
The data argument can be a String, Buffer or Stream.
letkey='test-cached-item'letdata='test data'cache.set(key,data).then(()=>{// do something}).catch(err=>{// Error})Returns a Promise that returns a Stream of the cached data if the key exists or an Error if it does not exist. The error message returned is "The specified key does not exist".
letkey='test-cached-item'cache.get(key).then(stream=>{// do something with the stream}).catch(err=>{// err === "The specified key does not exist"})constexpress=require('express')constapp=express()constdadiCache=require('@dadi/cache')constcache=newdadiCache()app.get(function(req,res){letkey=req.urlcache.get(key).then(stream=>{// cached data found for req.urlres.setHeader('X-Cache','HIT')stream.pipe(res)}).catch(err=>{// cached data not found for req.urlletcontent=fetchContent()// cache the contentcache.set(key,content).then(()=>{res.setHeader('X-Cache','MISS')res.end(content)})})})| Property | Description | Default | Example |
|---|---|---|---|
| ttl | The time, in seconds, after which cached data is considered stale | true | 3600 |
A DADI Cache instance can be created with no options, in which case the following options will be used:
{"directory": {"enabled": true,"path": "./cache"},"redis": {"enabled": false}}| Property | Description | Default | Example |
|---|---|---|---|
| enabled | If true, caching is enabled using the following settings | true | |
| path | The absolute or relative path to the directory for cache files | "./cache" | "/tmp/dadi-cache/" |
| extension | (optional) The extension to use for cache files | none | "json" |
| directoryChunkSize | (optional) If set, cache files are stored in a series of subdirectories based on the cache key | 0 | 5 |
| autoFlush | If true, DADI Cache will clear cache files that are older than the specified TTL setting, at the interval specified by autoFlushInterval | false | true |
| autoFlushInterval | The period of time between clearing cache files (in seconds) | 300 | 1800 |
A set of options for both file and Redis caching must be provided if you intend to use Redis as the cache store. This allows DADI Cache to fallback to file caching in the event of a Redis connection failure.
| Property | Description | Default | Example |
|---|---|---|---|
| enabled | If true, caching is enabled using the following settings | false | true |
| host | The hostname or IP address of the Redis server | "127.0.0.1" | "drum.redistogo.com" |
| port | The port of the Redis server | 6379 | 9092 |
{
"directory": {
"enabled": true,
"path": "./cache"
},
"redis": {
"enabled": true,
"host": "127.0.0.1",
"port": 6379
}
}| Property | Description | Default | Example |
|---|---|---|---|
| enabled | If true, caching is enabled using the following settings | false | true |
| cluster | If true, DADI Cache will connect caching is enabled using the following settings | false | true |
| scaleReads | Specify where to send queries, to the masters, slaves, or a combination. See Read-Write Splitting | "master" | "all" |
| hosts | When cluster: true, Cache uses this array of hosts to connect. Each array item must contain a host and port. | [{"host":"127.0.0.1", "port": 6379}, {"host":"127.0.0.1", "port": 6380}] |
To connect to a Redis cluster an array of hosts must be specified, rather than a single host and port.
The array does not need to contain all your cluster nodes, but a few so that if one is unreachable the next one will be tried. DADI Cache will discover other nodes automatically when at least one node is connnected.
{
"directory": {
"enabled": true,
"path": "./cache"
},
"redis": {
"enabled": true,
"cluster": true,
"scaleReads": "all",
"hosts": [
{
"host": "127.0.0.1",
"port": 6379
},
{
"host": "127.0.0.1",
"port": 6383
}
]
}
}A typical Redis cluster contains three or more masters and several slaves for each master. It's possible to scale out Redis cluster by sending read queries to slaves and write queries to masters by setting the scaleReads option.
scaleReads is "master" by default, which means no queries will be sent to slaves. The other available options:
- "all": Send write queries to masters and read queries to masters or slaves randomly.
- "slave": Send write queries to masters and read queries to slaves.
For example, with scaleReads: "slave":
cache.set('foo','bar')// This query will be sent to one of the masters.cache.get('foo',(err,res)=>{// This query will be sent to one of the slaves.})Note: In the code snippet above, the result may not be equal to "bar" because of the lag of replication between the master and slaves.
In the case of a Redis connection failure, DADI Cache will attempt to reconnect four times before switching to file caching. After a configurable period (default 5 minutes), an attempt will be made to reconnect to Redis and if successful DADI Cache will resume using Redis as the cache store.
- Add LATENCY HISTORY to assist in determining Redis performance
- Add FLUSH
- Add authentication options for Redis