Telemetry Collected by InfluxDB 2.x
InfluxData collects information about the usage of InfluxDB. This “phone home” capability is intended to provide us with information about the broader user community and an understanding of how the platform is being used. These statistics are never shared with anyone else and help us make better decisions about product features, supportability and more. Ultimately, sharing these metrics with us can help us better understand how you use our product and can help us improve the product over time.
Some people may feel uncomfortable about such telemetry data being gathered, so we want to be as transparent as possible about the information being captured and about how to ‘opt-out’, if you so choose.
Disabling Telemetry
Telemetry reporting is enabled by default. However, we’ve made it as easy as possible to ‘opt-out’ of the reporting. Simply start the InfluxDB process with the --reporting-disabled flag.
You can see the specific command in our documentation.
Metrics Collection
For each InfluxDB 2.x installation, we collect the following at startup and then every 8 hours:
| Tags | Description |
|---|---|
| arch | Microarchitecture InfluxDB was compiled for |
| build date | Date associated with the InfluxDB build |
| commit | SHA of commit associated with the InfluxDB build |
| cpus | Number of CPUs running InfluxDB |
| functions | Flux functions |
| id | Snowflake identifier for the InfluxDB instance |
| Index partition | Identifies the index partition used by the underlying InfluxDB storage engine |
| ip | IP Address of the inbound connection which reports the statistics. This is not the specific IP Address of the machine running InfluxDB unless it is exposed directly on the public Internet. |
| org | Identifier for an organization. Allows for grouping of statistics by organization within the InfluxDB instance |
| os | Operating System InfluxDB is running on |
| result | Text allowing grouping of Flux query invocations results |
| series file partition | Identifies the series files in use for the underlying InfluxDB storage engine. This is not the metadata about series. |
| status | Status of write ahead log (associated to number of successful /failed writes) |
| user_agent | Typically, this is set by the browser, InfluxDB client libraries (includes the language [Go, JavaScript, Java, C#, Ruby, Python, etc.] and version), and other technologies [such as third-party dashboarding applications, etc.]. |
| version | InfluxDB version |
With those tag elements, we then leverage a combination of the unique combination of id, ip, and storage system specifics (where applicable) to capture usage counts of the various subsystems within InfluxDB.
| Fields | Description |
|---|---|
| buckets total counter | Total number of buckets present within the InfluxDB instance |
| bytes written counter | Total number of bytes written |
| bytes scanned counter | Total number of bytes scanned within the storage system via queries and tasks |
| dashboards total counter | Total number of dashboards present within the InfluxDB instance |
| flux function total counter | Total number of calls by function invoked within Flux |
| http api requests counter | Total number of API invocations by each API path |
| query duration histogram | Histogram counting duration of queries into bins |
| organizations total counter | Total number of organizations present within the InfluxDB instance |
| scrapers total counter | Total number of scrapers configured within the InfluxDB instance |
| series total counter | Total number of series present within the InfluxDB instance |
| storage total counter | Total number of bytes stored within the InfluxDB instance |
| task scheduler gauge | Number of tasks running within the InfluxDB instance |
| telegrafs total counter | Total number of Telegraf configurations within the InfluxDB instance |
| tokens total counter | Total number of tokens present within the InfluxDB instance |
| uptime gauge | Number of seconds InfluxDB has been continuously running |
| users total counter | Total number of users present within the InfluxDB instance |
| wal current segment gauge | Number of bytes in the current segments for the write ahead log |
| wal writes total counter | Total number of writes to the write ahead log by status (ok, fail, etc.) |
