feat: Adds gzip compression level configuration to backups for v2

[devanbenz]

No description provided.

[Copilot]

Pull request overview

Adds support for configuring server-side gzip compression level when downloading backup snapshots (v2), wiring a new CLI flag through the backup client and OpenAPI-generated client.

Changes:

• Add --gzip-compression-level CLI flag and plumb it into backup request parameters.
• Send Gzip-Compression-Level header on backup metadata and shard download requests when configured.
• Update OpenAPI contract and regenerated models/clients (including InfluxQL response/model updates).

Reviewed changes

Copilot reviewed 9 out of 9 changed files in this pull request and generated 5 comments.

Show a summary per file

File	Description
cmd/influx/backup.go	Adds CLI flag to configure server-side gzip compression level.
clients/backup/backup.go	Passes configured gzip compression level into API requests.
api/contract/overrides/paths/backup_metadata.yml	Documents the new request header in the OpenAPI overrides.
api/api_backup.gen.go	Adds generated request field + header plumbing for Gzip-Compression-Level.
api/model_script_language.gen.go	Adds influxql as a supported ScriptLanguage value.
api/model_influxql_json_response*.gen.go	Improves InfluxQL response model docs and adds partial to results.
api/contract/openapi	Bumps OpenAPI submodule pointer for regenerated output.

---

💡 Add Copilot custom instructions for smarter, more guided reviews. Learn how to get started.

[devanbenz]

Analysis of change here:

The main consumer is the v1 shell client (clients/v1_shell/v1_shell.go). It calls the legacy /query endpoint, deserializes the JSON into api.InfluxqlJsonResponse via json.Unmarshal(), then renders results. The schema change adds descriptions and a Partial boolean field, since json.Unmarshal ignores unknown/missing fields by default, existing code won't break from the addition.

[davidby-influx]

LGTM, but @gwossum should review as well.

[gwossum]

@devanbenz This doesn't seem to work. I ran this branch of influx-cli against the influxdb db/2.x-oss-backup-perf branch. I expected to get an error when I specified --gzip-compression-level asdf, but I did not.

I captured the HTTP headers from influx-cli and saw:

Gzip-Compression-Level: asdf

It looks like influxdb is looking for a different header (gzip_compression_level):
https://github.com/influxdata/influxdb/blob/5efa0f051d784d35f5c0f165be65e7b6960a2799/http/backup_service.go#L100

[gwossum]

@devanbenz In 1.x, it looks like the TSM file compression is done on the backup client side, but in 2.x the TSM file compression is done on the server side. Is that correct?

Assuming that's true, one thing to note is that the 2.x influx-cli currently passes along any compression level without checking if it's valid, then relies on the server to generate the error about an invalid compression level. Overall, I don't think that's a bad thing. Iif we add a new compression level you could use an outdated client and still take advantage of new compression levels on the server.

[devanbenz]

@devanbenz This doesn't seem to work. I ran this branch of influx-cli against the influxdb db/2.x-oss-backup-perf branch. I expected to get an error when I specified --gzip-compression-level asdf, but I did not.

I captured the HTTP headers from influx-cli and saw:

Gzip-Compression-Level: asdf

It looks like influxdb is looking for a different header (gzip_compression_level): influxdata/influxdb@5efa0f0/http/backup_service.go#L100

Sorry, it is very possible you need to be testing against my latest changes to influxdb: influxdata/influxdb#27297 which are not yet ready.