InfluxDB Proxy

This project adds a basic high availability and consistent hash layer to InfluxDB.

NOTE: influx-proxy must be built with Go 1.21+ with Go module support, don't implement udp.

NOTE: InfluxDB Cluster - open source alternative to InfluxDB Enterprise has been released, which is better than InfluxDB Proxy.

Why

We used InfluxDB Relay before, but it doesn't support some demands. We use grafana for visualizing time series data, so we need add datasource for grafana. We need change the datasource config when influxdb is down. We need transfer data across idc, but Relay doesn't support gzip. It's inconvenient to analyse data with connecting different influxdb. Therefore, we made InfluxDB Proxy. More details please visit https://github.com/shell909090/influx-proxy.

Forked from the above InfluxDB Proxy, after many improvements and optimizations, InfluxDB Proxy v1 has released, which no longer depends on python and redis, and supports more features.

Since the InfluxDB Proxy v1 is limited by the only ONE database and the KEYMAPS configuration, we refactored InfluxDB Proxy v2 with high availability and consistent hash, which supports multiple databases and tools to rebalance, recovery, resync and cleanup.

Features

• Support query and write.
• Support /api/v2 endpoints.
• Support flux language query.
• Support some cluster influxql.
• Filter some dangerous influxql.
• Transparent for client, like cluster for client.
• Cache data to file when write failed, then rewrite.
• Support multiple databases to create and store.
• Support database sharding with consistent hash.
• Support custom hash key and shard key of database sharding.
• Support tools to rebalance, recovery, resync and cleanup.
• Load config file and no longer depend on python and redis.
• Support both rp and precision parameter when writing data.
• Support influxdb-java, influxdb shell and grafana.
• Support prometheus remote read and write.
• Support prometheus monitor with /metrics.
• Support authentication and https.
• Support authentication encryption.
• Support health status check.
• Support database whitelist.
• Support version display.
• Support gzip.

Requirements

• Golang >= 1.21 with Go module support
• InfluxDB 1.2 - 1.8 (For InfluxDB 2.x, please visit branch influxdb-v2)

Usage

Quickstart by Docker

Download docker-compose.yml and proxy.json from docker/quick

$ docker-compose up -d

An influx-proxy container (port: 7076) and 4 influxdb containers will start.

Quickstart

$ git clone https://github.com/chengshiwen/influx-proxy.git
$ cd influx-proxy
$ make
$ ./bin/influx-proxy -config proxy.json

Usage

$ ./bin/influx-proxy -h
Usage of ./bin/influx-proxy:
  -config string
        proxy config file with json/yaml/toml format (default "proxy.json")
  -version
        proxy version

Build Release

$ # build current platform
$ make build
$ # build linux amd64
$ make linux
$ # cross-build all platforms
$ make release

Development

Before developing, you need to install and run Docker

$ ./script/setup.sh  # start 4 influxdb instances by docker
$ make run
$ ./script/write.sh  # write data
$ ./script/query.sh  # query data
$ ./script/remove.sh # remove 4 influxdb instances

Description

The architecture is fairly simple, one InfluxDB Proxy instance and two consistent hash circles with two InfluxDB instances respectively. The Proxy should point HTTP requests with db and measurement to the two circles and the four InfluxDB servers.

The setup should look like this:

        ┌──────────────────┐
        │ writes & queries │
        └──────────────────┘
                 │
                 ▼
        ┌──────────────────┐
        │                  │
        │  InfluxDB Proxy  │
        │   (only http)    │
        │                  │
        └──────────────────┘
                 │
                 ▼
        ┌──────────────────┐
        │  db,measurement  │
        │ consistent hash  │
        └──────────────────┘
          |              |
        ┌─┼──────────────┘
        │ └────────────────┐
        ▼                  ▼
     Circle 1          Circle 2
  ┌────────────┐    ┌────────────┐
  │            │    │            │
  │ InfluxDB 1 │    │ InfluxDB 3 │
  │ InfluxDB 2 │    │ InfluxDB 4 │
  │            │    │            │
  └────────────┘    └────────────┘

Proxy Configuration

The configuration file supports format json, yaml and toml, such as proxy.json, proxy.yaml and proxy.toml.

The configuration settings are as follows:

• circles: circle list
  • name: circle name, required
  • backends: backend list belong to the circle, required
    • name: backend name, required
    • url: influxdb addr or other http backend which supports influxdb line protocol, required
    • username: influxdb username, with encryption if auth_encrypt is enabled, default is empty which means no auth
    • password: influxdb password, with encryption if auth_encrypt is enabled, default is empty which means no auth
    • auth_encrypt: whether to encrypt auth (username/password), default is false
    • write_only: whether to write only on the influxdb, default is false
• listen_addr: proxy listen addr, default is :7076
• db_list: database list permitted to access, default is []
• data_dir: data dir to save .dat .rec, default is data
• tlog_dir: transfer log dir to rebalance, recovery, resync or cleanup, default is log
• hash_key: backend key for consistent hash, including idx, exi, name, url or template containing %idx, like backend-%idx, default is idx, once changed rebalance operation or influx-tool transfer is necessary
• shard_key: data shard key template for hash, which containing %db or %mm, like shard-%db-%mm, default is %db,%mm which means database,measurement, once changed rebalance operation or influx-tool transfer is necessary
• flush_size: default is 10000, wait 10000 points write
• flush_time: default is 1, wait 1 second write whether point count has bigger than flush_size config
• check_interval: default is 1, check backend active every 1 second
• rewrite_interval: default is 10, rewrite every 10 seconds
• rewrite_threads: default is 5, rewrite under 5 threads
• conn_pool_size: default is 20, create a connection pool which size is 20
• write_timeout: default is 10, write timeout until 10 seconds
• idle_timeout: default is 10, keep-alives wait time until 10 seconds
• username: proxy username, with encryption if auth_encrypt is enabled, default is empty which means no auth
• password: proxy password, with encryption if auth_encrypt is enabled, default is empty which means no auth
• auth_encrypt: whether to encrypt auth (username/password), default is false
• ping_auth_enabled: enable authentication on /ping and /metrics, default is false
• write_tracing: enable logging for the write, default is false
• query_tracing: enable logging for the query, default is false
• pprof_enabled: enable /debug/pprof HTTP endpoint, default is false
• https_enabled: enable https, default is false
• https_cert: the ssl certificate to use when https is enabled, default is empty
• https_key: use a separate private key location, default is empty
• tls: configuration settings for tls
  • ciphers: set of cipher suite IDs to negotiate when https is enabled, referring to ciphersMap, default is []
  • min_version: minimum version of the tls protocol when https is enabled, including tls1.0, tls1.1, tls1.2 and tls1.3, default is empty
  • max_version: maximum version of the tls protocol when https is enabled, including tls1.0, tls1.1, tls1.2 and tls1.3, default is empty

Hash and Shard Key

hash_key and shard_key together control which influxdb instance the data should be written to. hash_key is backend key for consistent hash, including idx, exi, name, url or template containing %idx (%idx is the index of the influxdb backend in the circle), like backend-%idx. shard_key is data shard key template for hash, which containing %db or %mm (%db and %mm are the database and measurement in the data respectively), like shard-%db-%mm.

To avoid data skew (i.e. uneven data distribution), both hash_key and shard_key need to be set appropriately. Before setting, influx-tool hashdist can help simulate and test the hash distribution. For example, execute

$ head -n 3 table.csv
db1,cpu1
db1,cpu2
db2,cpu3
$ influx-tool hashdist -n 6 -k backend-%idx -K shard-%db-%mm -f table.csv -D -
node total: 6, hash key: backend-%idx, shard key: shard-%db-%mm, total hits: 40
node index: 0, hits: 5, percent: 12.5%, expect: 16.7%
node index: 1, hits: 5, percent: 12.5%, expect: 16.7%
node index: 2, hits: 9, percent: 22.5%, expect: 16.7%
node index: 3, hits: 7, percent: 17.5%, expect: 16.7%
node index: 4, hits: 7, percent: 17.5%, expect: 16.7%
node index: 5, hits: 7, percent: 17.5%, expect: 16.7%

NOTE: Once one of hash_key and shard_key is changed, rebalance operation or influx-tool transfer is necessary.

Query Commands

Unsupported commands

The following commands are forbid.

• GRANT
• REVOKE
• KILL
• EXPLAIN
• SELECT INTO
• CONTINUOUS QUERY
• Multiple queries delimited by semicolon ;
• Multiple measurements delimited by comma ,
• Regexp measurement

Supported commands

Only support match the following commands.

• select from
• show from
• show measurements
• show series
• show field keys
• show tag keys
• show tag values
• show stats
• show databases
• create database
• drop database
• show retention policies
• create retention policy
• alter retention policy
• drop retention policy
• delete from
• drop series from
• drop measurement
• on clause
• from clause like from <db>.<rp>.<measurement>

HTTP Endpoints

HTTP Endpoints

Benchmark

There are three tools for benchmarking InfluxDB, which can also be applied to InfluxDB Proxy:

• influx-stress is a stress tool for generating artificial load on InfluxDB.
• influxdb-comparisons contains code for benchmarking InfluxDB against other databases and time series solutions.
• tsbs (Time Series Benchmark Suite) is a tool for comparing and evaluating databases for time series data.

Tool

There is a tool for InfluxDB and InfluxDB Proxy:

• influx-tool: high performance tool to rebalance, recovery, resync, cleanup and compact. most commands do not require InfluxDB to start

License

MIT.