Document the "no_sync" write option for InfluxDB

[MeelahMe]

Describe your proposal here.

A user in Kapa.ai asked about the no_sync write option. Kapa.ai could not answer the question due to not being able to find any documentation covering the topic in the InfluxDB docs.

Kapa.ai response did indicate that there is some related documentation that exists for the write consistency levels and batching options but nothing explicitly for no_sync.

My request would be to:

• add a explanation of the no_sync write option
• Describe its purpose, behavior, and use cases
• Include any relevant configuration details

Relevant URLs

• Provide relevant URLs

https://app.kapa.ai/720100f9-ce93-4305-88fd-5fcf71effad7/conversations/96609cd4-23fc-4956-8408-fa64bc247d74

Extra info:

issue #5825: Document no_sync (HTTP API) and no-sync (CLI) in write guides and CLI reference.
Enterprise-only?

[jstirnaman]

I talked to Pete briefly and researched this for the API spec. Here's what I came up with:

      type: boolean
      default: false
      description: |
        Do not sync writes.
            
        By default, InfluxDB writes ingested data to the WAL file and then waits for the WAL to be persistend to disk before acknowledging the write.

        If `no_sync=true`, InfluxDB writes ingested data to the WAL file and the server does not wait for the write to be persisted before acknowledging the write.

@peterbarnett03 Can you confirm (or tag someone else to confirm) this is correct?

[jstirnaman]

@MeelahMe Nicely done! This is a great example that we can share with the team for how to create issues from Kapa conversation feedback.