filer: switch workspace upload from import-file to /workspace/import by shreyas-goenka · Pull Request #5165 · databricks/cli

shreyas-goenka · 2026-05-04T08:31:02Z

Summary

Replace POST /api/2.0/workspace-files/import-file/{path} with the multipart variant of POST /api/2.0/workspace/import (via the SDK's Workspace.Upload + format=AUTO). The previous endpoint is being deprecated and the new endpoint is the strategic replacement.

Why

The import-file file API is being deprecated.
In practice /workspace/import is ~2x faster for typical bundle deployments (see benchmarks below).
The import API is well supported, has a higher rate limit (30 vs 20 rps/org), and returns structured errors.

Why multipart

/workspace/import's JSON body caps at 10 MiB (databricks.webapp.autoExportFormatLimitBytes). The multipart variant accepts the same sizes import-file did — verified up to 450 MB for regular files against a real workspace.

End-to-end `bundle deploy` benchmark (aws-prod-ucws, direct engine)

Fresh upload of all files on every deploy (local sync state wiped between runs), 4 KiB files, binaries interleaved per iteration, 3 timed runs each:

Files	import-file (median, range)	workspace/import (median, range)	Speedup
100	15.2 s (13.5–18.7)	8.9 s (8.7–8.9)	1.7×
400	44.4 s (44.0–51.6)	28.9 s (28.9–28.9)	1.5×

Each deploy includes ~3–4 s of fixed non-upload overhead, so the upload phase itself improves by more than the totals suggest. Also note the run-to-run variance: ±0.1 s on the new endpoint vs ±5–8 s on the legacy one (dedicated workspace-import RLv2 group + stateless WCS vs the shared Projects limiter).

File-level throughput (aws-prod-ucws, 10 workers, 4 KiB files)

Endpoint	Files	Wall	files/s	Speedup
new	10	1.01 s	9.94	0.86× (legacy faster)
legacy	10	0.86 s	11.58	—
new	20	1.37 s	14.61	2.22×
legacy	20	3.04 s	6.58	—
new	50	2.72 s	18.40	1.74×
legacy	50	4.73 s	10.57	—
new	100	6.12 s	16.34	2.34×
legacy	100	14.31 s	6.99	—
new	300	19.39 s	15.48	1.81×
legacy	300	35.16 s	8.53	—
new	500	34.69 s	14.41	2.27×
legacy	500	1 m 18.51 s	6.37	—

Crossover at ~15 files. New endpoint sustains ~15 files/s; legacy degrades from 17 → 6 files/s as count grows.

Size caps (verified)

	Files (no header)	Notebooks (`# Databricks notebook source`)
`/workspace/import` (this PR, multipart)	OK at 450 MB	hard 10 MiB cap
`/workspace-files/import-file` (previous)	400 MB; 450 MB → 305 s server timeout	hard 10 MiB cap (same `databricks.notebook.maxNotebookSizeBytes`)

The 10 MiB notebook cap is server-side and applies to both endpoints — switching does not regress maximum notebook upload size.

Error mapping

errors.Is against SDK sentinels rather than raw status/error_code dispatch. All "path is occupied" surfaces map to fileAlreadyExistsError:

400 RESOURCE_ALREADY_EXISTS — sequential conflict, no overwrite flag.
409 ALREADY_EXISTS — concurrent contention (observed in TestLock).
400 INVALID_PARAMETER_VALUE — overwrite=true on a path where the existing object's node type differs from the upload. Branches on the AIP-193 ErrorInfo reason WORKSPACE_OBJECT_TYPE_MISMATCH that WCS now attaches to import path collisions (universe PR #2019174, WP-6031), with a message-substring fallback ("type mismatch" / "node type") for workspaces where that change has not landed.

Note on the fallback: a probe against aws-prod-ucws on 2026-06-12 showed all four collision shapes still return errors without details (no ErrorInfo), i.e. the WP-6031 rollout has not reached that workspace (or /workspace/import there is not yet served by WCS). The fallback is therefore load-bearing for now; it can be removed once the rollout is confirmed.

Plus ErrNotFound → noSuchDirectoryError (or mkdir+retry under CreateParentDirectories); ErrPermissionDenied → permissionError.

Test plan

Unit tests cover all three already-exists shapes + 403 / 500, and the new ErrorInfo-reason branch via a real HTTP roundtrip (the SDK only populates error details when parsing a response).
libs/testserver extended with multipart /workspace/import (now attaching the AIP-193 ErrorInfo detail on collisions), notebook header detection for .py/.sql/.scala/.r, and /workspace/list.
acceptance/bundle/sync-upload-edge-cases exercises 12 MiB binary, empty file, 3 notebook languages, plain .py (no header), .lvdash.json dashboard, non-ASCII filename, spaced filename. Now Git-Bash-safe on Windows (MSYS_NO_PATHCONV=1).
~70 acceptance fixtures regenerated; rebased onto current main and re-regenerated fixtures that changed there (user_agent/simple, cmd/sync/dryrun-missing-remote, force_pull_commands).
All cross-type collision shapes (4) re-verified by direct probe against a real workspace on 2026-06-12.
Load tested vs the previous endpoint at concurrencies 1/10/20/30 (1 KB) and at sizes up to 450 MB; end-to-end deploy benchmark above.
CI green on this PR.

This pull request and its description were written by Isaac, an AI coding assistant.

shreyas-goenka · 2026-05-08T06:49:58Z

+	//    Example: "Node with name /Users/me/.bundle/.../deploy.lock already
+	//    exists. Please pass overwrite=true to update it."
+	//
+	//  - 400 INVALID_PARAMETER_VALUE — overwrite=true on a path where the


This error is not good. This then requires us to do regex on the error to figure out whether it's "already exists". I'm asking the API team to change this.

shreyas-goenka · 2026-05-08T06:51:26Z

Benchmark results on the new API vs old API, the new API is 2x better when a lot of files are being uploaded:

  ┌──────────┬───────┬─────────────┬─────────┬───────────────────────┐
  │ Endpoint │ Files │    Wall     │ files/s │   Speedup vs legacy   │
  ├──────────┼───────┼─────────────┼─────────┼───────────────────────┤
  │ new      │    10 │      1.19 s │    8.37 │ 0.50× (legacy faster) │
  ├──────────┼───────┼─────────────┼─────────┼───────────────────────┤
  │ legacy   │    10 │      0.61 s │   16.45 │                     — │
  ├──────────┼───────┼─────────────┼─────────┼───────────────────────┤
  │ new      │   100 │      6.04 s │   16.56 │                  3.9× │
  ├──────────┼───────┼─────────────┼─────────┼───────────────────────┤
  │ legacy   │   100 │     23.55 s │    4.25 │                     — │
  ├──────────┼───────┼─────────────┼─────────┼───────────────────────┤
  │ new      │   300 │     19.36 s │   15.49 │                  3.2× │
  ├──────────┼───────┼─────────────┼─────────┼───────────────────────┤
  │ legacy   │   300 │ 1 m 01.62 s │    4.87 │                     — │
  ├──────────┼───────┼─────────────┼─────────┼───────────────────────┤
  │ new      │   500 │     32.74 s │   15.27 │                  3.0× │
  ├──────────┼───────┼─────────────┼─────────┼───────────────────────┤
  │ legacy   │   500 │ 1 m 38.62 s │    5.07 │                     — │
  └──────────┴───────┴─────────────┴─────────┴───────────────────────┘

  (10 workers, 4 KiB files, aws-prod-ucws.)

  Findings:
  - Crossover at ~50 files: legacy beats new for ≤ 10 files; new is 3-4× faster from 100+ on.
  - Legacy throughput collapses: 17 → 4-5 files/s as N grows, server applies escalating back-pressure.
  - New throughput is flat at ~15 files/s regardless of count — no degradation up through 500 files.
  - For typical bundle deploys (~100-1000 files), expect a 3-4× wall-clock speedup.

A future reader of this comment can't tell what "this migration" or "the endpoint we are migrating away from" refers to once the PR commit is in the past. Spell out /workspace-files/import-file → /workspace/import where it matters, and re-anchor the no-regression note on those endpoint names rather than "this migration". Co-authored-by: Isaac

…xisting sentinels) Verified against a real workspace by uploading varied content into pre-existing paths of mismatched node types: every cross-type collision the server can produce surfaces as either: - 400 RESOURCE_ALREADY_EXISTS, or - 409 ALREADY_EXISTS both already caught by errors.Is(err, ErrResourceAlreadyExists) / errors.Is(err, ErrAlreadyExists). The "400 INVALID_PARAMETER_VALUE with 'Requested node type [X] is different from the existing node type [Y]'" shape was hypothesized in the original PR but does not occur in practice on /workspace/import — the branch was dead code. Replace the corresponding unit test with one that pins the actual cross-type collision shape (409 ALREADY_EXISTS with "Node with name X already exists"). Co-authored-by: Isaac

…rwrite=true) Previous commit (687c731) called the INVALID_PARAMETER_VALUE branch dead code based on conflict scenarios that didn't pass overwrite=true. Re-probing with overwrite=true (the mode bundle deploy uses) shows the server does return 400 INVALID_PARAMETER_VALUE for cross-type collisions, in two distinct message shapes: "Cannot overwrite the asset at X due to type mismatch (asked: ..., actual: ...)" "Requested node type [...] is different from the existing node type [...]" Both fire only when the existing object's node type differs from the upload (NOTEBOOK at /a/foo, regular-content overwrite-upload to /a/foo or /a/foo.py). Without overwrite=true, the same scenario surfaces as RESOURCE_ALREADY_EXISTS or ALREADY_EXISTS — which is why the no-overwrite probe missed this branch. Restore the mapping (returning fileAlreadyExistsError so the caller's errors.Is(err, fs.ErrExist) check still fires) and broaden the message pattern from "Requested node type" to also match "type mismatch", so both real-workspace messages are caught. Add unit tests for both shapes. Co-authored-by: Isaac

…amples Drop the abstract "Sources for the third bullet" footer and inline a concrete example message for each of the three error shapes. Each bullet now shows exactly what the server returns and the scenario that triggers it, so a future reader doesn't have to re-derive which path/state combination produces each shape. Co-authored-by: Isaac

The previous regen sweep widened the filter from '^//import-file/' '^//workspace/delete' (two narrow exclusions) to '^//workspace/' (catch-all under /workspace/*). That hid the workspace/mkdirs calls the test was previously asserting on. Restore the original intent: only exclude the upload + delete requests, preserving the mkdirs assertions. Also pick up the out.test.toml regeneration for the force_pull_commands test that landed on main (5028) to switch the [Section] format to dotted keys. Co-authored-by: Isaac

…l + scala notebooks Extend the testserver to detect notebook headers for .sql, .scala, .r in addition to .py (each with its language-appropriate line-comment) and to serve GET /api/2.0/workspace/list — so acceptance tests can assert what the server actually stored, not just what was sent on the wire. Extend acceptance/bundle/sync-upload-edge-cases to upload pyNb.py, sqlNb.sql, scalaNb.scala (all with the source header), plus plain-script.py (no header), and assert via "databricks workspace list" that: - pyNb → NOTEBOOK / PYTHON (extension stripped) - sqlNb → NOTEBOOK / SQL (extension stripped) - scalaNb → NOTEBOOK / SCALA (extension stripped) - plain-script.py → FILE (header missing, extension preserved) - dashboard.lvdash.json, large.bin, empty.txt, héllo.txt, with spaces.txt all → FILE Co-authored-by: Isaac

…dge-cases The fact that the test produces deterministic output already proves the 12 MiB binary isn't recorded byte-for-byte (otherwise the recording would contain literal megabytes of payload). The dedicated "12 MiB binary upload is summarized" assertion was implementation-detail noise. Co-authored-by: Isaac

github-actions · 2026-05-08T08:07:23Z

Approval status: pending

`/acceptance/bundle/` - needs approval

71 files changed
Suggested: @denik
Also eligible: @pietern, @andrewnester, @janniklasrose, @anton-107, @lennartkats-db

`/libs/filer/` - needs approval

Files: libs/filer/workspace_files_client.go, libs/filer/workspace_files_client_test.go
Suggested: @simonfaltum
Also eligible: @Divyansh-db, @renaudhartert-db, @hectorcast-db, @parthban-db, @tanmay-db, @tejaskochar-db, @mihaimitrea-db, @chrisst, @rauchy

General files (require maintainer)

4 files changed
Based on git history:

@denik -- recent work in libs/testserver/, libs/filer/, acceptance/internal/

_{Any maintainer (@andrewnester, @anton-107, @denik, @pietern, @simonfaltum, @renaudhartert-db) can approve all areas.

See OWNERS for ownership rules.}

Squashed rebase of shreyas-goenka/import-api onto main. Replaces the POST /api/2.0/workspace-files/import-file/{path} call in WorkspaceFilesClient.Write with the SDK's Workspace.Upload (multipart POST /api/2.0/workspace/import, format=AUTO), keeping main's errors.AsType/workspaceIDHeaders refactors in the untouched methods. Co-authored-by: Isaac

…llisions WCS now attaches structured ErrorInfo (reason + metadata) to /workspace/import path-collision errors (universe PR #2019174, WP-6031). Branch on reason WORKSPACE_OBJECT_TYPE_MISMATCH instead of only string-matching the message. The message match is kept as a fallback: a probe on 2026-06-12 showed aws-prod-ucws still returns these errors without details, so the rollout has not reached all workspaces yet. The testserver now attaches the same ErrorInfo detail to its translated RESOURCE_ALREADY_EXISTS response to mirror the server behavior. Co-authored-by: Isaac

- Drop the trailing blank line in bind/pipelines/update/script that failed the whitespace linter. - Prefix workspace list with MSYS_NO_PATHCONV=1 in sync-upload-edge-cases so Git Bash on Windows does not rewrite the /Workspace/... argument into a C:/Program Files/... path. - Regenerate the user_agent/simple terraform fixture against current main (provider 1.117.0, go 1.25.8, provider_config workspace_id). Co-authored-by: Isaac

dryrun-missing-remote (added on main) records upload requests and still expected the legacy import-file endpoint; force_pull_commands out.test.toml no longer dumps GOOS keys with the current harness. Co-authored-by: Isaac

The previous branch history is superseded by the rebased commits; this merge keeps the rebased tree exactly and preserves the old head as an ancestor so the PR branch can be updated without a force push. Co-authored-by: Isaac

Co-authored-by: Isaac

eng-dev-ecosystem-bot · 2026-06-12T23:34:36Z

Integration test report

Commit: 5a38b31

Run: 27448021211

	Env	🟨KNOWN	🔄flaky	💚RECOVERED	🙈SKIP	✅pass	🙈skip	Time
🟨	aws linux	7			15	264	978	8:29
🟨	aws windows	7			15	266	976	14:41
💚	aws-ucws linux			7	15	360	892	7:03
💚	aws-ucws windows			7	15	362	890	13:58
💚	azure linux			1	17	267	976	6:28
💚	azure windows			1	17	269	974	12:12
💚	azure-ucws linux			1	17	365	888	7:34
💚	azure-ucws windows			1	17	367	886	12:23
💚	gcp linux			1	17	263	979	8:27
🔄	gcp windows		2	1	17	263	977	14:54

24 interesting tests: 15 SKIP, 7 KNOWN, 2 flaky

	Test Name	aws linux	aws windows	aws-ucws linux	aws-ucws windows	azure linux	azure windows	azure-ucws linux	azure-ucws windows	gcp linux	gcp windows
🟨	TestAccept	🟨K	🟨K	💚R	💚R	💚R	💚R	💚R	💚R	💚R	💚R
🙈	TestAccept/bundle/invariant/no_drift	🙈S	🙈S	🙈S	🙈S	🙈S	🙈S	🙈S	🙈S	🙈S	🙈S
🙈	TestAccept/bundle/resources/permissions	🙈S	🙈S	🙈S	🙈S	🙈S	🙈S	🙈S	🙈S	🙈S	🙈S
🟨	TestAccept/bundle/resources/permissions/jobs/destroy_without_mgmtperms/with_permissions	🟨K	🟨K	💚R	💚R	🙈S	🙈S	🙈S	🙈S	🙈S	🙈S
🟨	TestAccept/bundle/resources/permissions/jobs/destroy_without_mgmtperms/with_permissions/DATABRICKS_BUNDLE_ENGINE=direct	🟨K	🟨K	💚R	💚R
🟨	TestAccept/bundle/resources/permissions/jobs/destroy_without_mgmtperms/with_permissions/DATABRICKS_BUNDLE_ENGINE=terraform	🟨K	🟨K	💚R	💚R
🟨	TestAccept/bundle/resources/permissions/jobs/destroy_without_mgmtperms/without_permissions	🟨K	🟨K	💚R	💚R	🙈S	🙈S	🙈S	🙈S	🙈S	🙈S
🟨	TestAccept/bundle/resources/permissions/jobs/destroy_without_mgmtperms/without_permissions/DATABRICKS_BUNDLE_ENGINE=direct	🟨K	🟨K	💚R	💚R
🟨	TestAccept/bundle/resources/permissions/jobs/destroy_without_mgmtperms/without_permissions/DATABRICKS_BUNDLE_ENGINE=terraform	🟨K	🟨K	💚R	💚R
🙈	TestAccept/bundle/resources/postgres_branches/basic	🙈S	🙈S	🙈S	🙈S	🙈S	🙈S	🙈S	🙈S	🙈S	🙈S
🙈	TestAccept/bundle/resources/postgres_branches/recreate	🙈S	🙈S	🙈S	🙈S	🙈S	🙈S	🙈S	🙈S	🙈S	🙈S
🙈	TestAccept/bundle/resources/postgres_branches/replace_existing	🙈S	🙈S	🙈S	🙈S	🙈S	🙈S	🙈S	🙈S	🙈S	🙈S
🙈	TestAccept/bundle/resources/postgres_branches/update_protected	🙈S	🙈S	🙈S	🙈S	🙈S	🙈S	🙈S	🙈S	🙈S	🙈S
🙈	TestAccept/bundle/resources/postgres_branches/without_branch_id	🙈S	🙈S	🙈S	🙈S	🙈S	🙈S	🙈S	🙈S	🙈S	🙈S
🙈	TestAccept/bundle/resources/postgres_endpoints/basic	🙈S	🙈S	🙈S	🙈S	🙈S	🙈S	🙈S	🙈S	🙈S	🙈S
🙈	TestAccept/bundle/resources/postgres_endpoints/recreate	🙈S	🙈S	🙈S	🙈S	🙈S	🙈S	🙈S	🙈S	🙈S	🙈S
🙈	TestAccept/bundle/resources/postgres_projects/update_display_name	🙈S	🙈S	🙈S	🙈S	🙈S	🙈S	🙈S	🙈S	🙈S	🙈S
🙈	TestAccept/bundle/resources/synced_database_tables/basic	🙈S	🙈S	🙈S	🙈S	🙈S	🙈S	🙈S	🙈S	🙈S	🙈S
🙈	TestAccept/bundle/resources/vector_search_endpoints/drift/recreated_same_name	🙈S	🙈S	🙈S	🙈S	🙈S	🙈S	🙈S	🙈S	🙈S	🙈S
🙈	TestAccept/bundle/resources/vector_search_indexes/basic	🙈S	🙈S	🙈S	🙈S	🙈S	🙈S	🙈S	🙈S	🙈S	🙈S
🙈	TestAccept/bundle/resources/vector_search_indexes/grants/select	🙈S	🙈S	🙈S	🙈S	🙈S	🙈S	🙈S	🙈S	🙈S	🙈S
🙈	TestAccept/ssh/connection	🙈S	🙈S	🙈S	🙈S	🙈S	🙈S	🙈S	🙈S	🙈S	🙈S
🔄	TestFetchRepositoryInfoAPI_FromRepo/root	✅p	✅p	✅p	✅p	✅p	✅p	✅p	✅p	✅p	🔄f
🔄	TestFetchRepositoryInfoAPI_FromRepo/subdir	✅p	✅p	✅p	✅p	✅p	✅p	✅p	✅p	✅p	🔄f

Top 29 slowest tests (at least 2 minutes):

duration	env	testname
6:15	aws-ucws windows	TestAccept
6:14	azure windows	TestAccept
6:08	gcp windows	TestAccept
6:05	azure-ucws windows	TestAccept
4:39	gcp windows	TestAccept/bundle/resources/apps/inline_config/DATABRICKS_BUNDLE_ENGINE=terraform
4:21	gcp linux	TestAccept/bundle/resources/apps/inline_config/DATABRICKS_BUNDLE_ENGINE=terraform
4:14	gcp linux	TestAccept/bundle/resources/apps/inline_config/DATABRICKS_BUNDLE_ENGINE=direct
4:06	gcp windows	TestAccept/bundle/resources/apps/inline_config/DATABRICKS_BUNDLE_ENGINE=direct
3:24	aws windows	TestAccept/bundle/resources/apps/inline_config/DATABRICKS_BUNDLE_ENGINE=terraform
3:08	azure windows	TestAccept/bundle/resources/apps/inline_config/DATABRICKS_BUNDLE_ENGINE=terraform
3:08	azure-ucws windows	TestAccept/bundle/resources/apps/inline_config/DATABRICKS_BUNDLE_ENGINE=terraform
3:01	aws-ucws windows	TestAccept/bundle/resources/apps/inline_config/DATABRICKS_BUNDLE_ENGINE=direct
2:57	azure linux	TestAccept
2:55	aws-ucws linux	TestAccept/bundle/resources/apps/inline_config/DATABRICKS_BUNDLE_ENGINE=direct
2:54	azure-ucws windows	TestAccept/bundle/resources/apps/inline_config/DATABRICKS_BUNDLE_ENGINE=direct
2:53	gcp linux	TestAccept
2:53	azure-ucws linux	TestAccept
2:49	aws linux	TestAccept/bundle/resources/apps/inline_config/DATABRICKS_BUNDLE_ENGINE=direct
2:47	aws-ucws linux	TestAccept
2:46	azure linux	TestAccept/bundle/resources/apps/inline_config/DATABRICKS_BUNDLE_ENGINE=terraform
2:40	azure-ucws linux	TestAccept/bundle/resources/apps/inline_config/DATABRICKS_BUNDLE_ENGINE=terraform
2:40	aws-ucws windows	TestAccept/bundle/resources/apps/inline_config/DATABRICKS_BUNDLE_ENGINE=terraform
2:37	aws linux	TestAccept/bundle/resources/apps/inline_config/DATABRICKS_BUNDLE_ENGINE=terraform
2:35	aws-ucws linux	TestAccept/bundle/resources/apps/inline_config/DATABRICKS_BUNDLE_ENGINE=terraform
2:29	azure linux	TestAccept/bundle/resources/apps/inline_config/DATABRICKS_BUNDLE_ENGINE=direct
2:24	azure windows	TestAccept/bundle/resources/apps/inline_config/DATABRICKS_BUNDLE_ENGINE=direct
2:24	aws linux	TestSecretsPutSecretStringValue
2:21	azure-ucws linux	TestAccept/bundle/resources/apps/inline_config/DATABRICKS_BUNDLE_ENGINE=direct
2:21	aws windows	TestAccept/bundle/resources/apps/inline_config/DATABRICKS_BUNDLE_ENGINE=direct

shreyas-goenka temporarily deployed to test-trigger-is May 4, 2026 08:31 — with GitHub Actions Inactive

shreyas-goenka force-pushed the shreyas-goenka/import-api branch from 0a9923f to e86aeba Compare May 6, 2026 10:38

shreyas-goenka temporarily deployed to test-trigger-is May 6, 2026 10:38 — with GitHub Actions Inactive

shreyas-goenka temporarily deployed to test-trigger-is May 6, 2026 11:11 — with GitHub Actions Inactive

shreyas-goenka force-pushed the shreyas-goenka/import-api branch from 4db3fb6 to 6907378 Compare May 6, 2026 11:26

shreyas-goenka temporarily deployed to test-trigger-is May 6, 2026 11:27 — with GitHub Actions Inactive

shreyas-goenka temporarily deployed to test-trigger-is May 6, 2026 12:55 — with GitHub Actions Inactive

shreyas-goenka force-pushed the shreyas-goenka/import-api branch 2 times, most recently from 37a1400 to 1e2fe83 Compare May 6, 2026 14:30

shreyas-goenka temporarily deployed to test-trigger-is May 6, 2026 14:31 — with GitHub Actions Inactive

shreyas-goenka temporarily deployed to test-trigger-is May 6, 2026 15:45 — with GitHub Actions Inactive

shreyas-goenka temporarily deployed to test-trigger-is May 7, 2026 10:00 — with GitHub Actions Inactive

shreyas-goenka temporarily deployed to test-trigger-is May 8, 2026 06:17 — with GitHub Actions Inactive

shreyas-goenka temporarily deployed to test-trigger-is May 8, 2026 06:22 — with GitHub Actions Inactive

shreyas-goenka temporarily deployed to test-trigger-is May 8, 2026 06:27 — with GitHub Actions Inactive

shreyas-goenka temporarily deployed to test-trigger-is May 8, 2026 06:36 — with GitHub Actions Inactive

shreyas-goenka commented May 8, 2026

View reviewed changes

shreyas-goenka added 5 commits May 8, 2026 08:56

shreyas-goenka force-pushed the shreyas-goenka/import-api branch from 5dd5be7 to d4e962c Compare May 8, 2026 07:11

shreyas-goenka temporarily deployed to test-trigger-is May 8, 2026 07:12 — with GitHub Actions Inactive

shreyas-goenka temporarily deployed to test-trigger-is May 8, 2026 07:37 — with GitHub Actions Inactive

shreyas-goenka temporarily deployed to test-trigger-is May 8, 2026 08:05 — with GitHub Actions Inactive

shreyas-goenka marked this pull request as ready for review May 8, 2026 08:06

shreyas-goenka requested review from andrewnester, denik and pietern May 8, 2026 08:06

shreyas-goenka added 5 commits June 12, 2026 17:20

shreyas-goenka temporarily deployed to test-trigger-is June 12, 2026 22:57 — with GitHub Actions Inactive

Merge origin/main into import-api branch

5a38b31

Co-authored-by: Isaac

shreyas-goenka temporarily deployed to test-trigger-is June 12, 2026 23:02 — with GitHub Actions Inactive

Provide feedback

Saved searches

Use saved searches to filter your results more quickly

filer: switch workspace upload from import-file to /workspace/import#5165

filer: switch workspace upload from import-file to /workspace/import#5165
shreyas-goenka wants to merge 20 commits into
mainfrom
shreyas-goenka/import-api

shreyas-goenka commented May 4, 2026 •

edited

Loading

Uh oh!

shreyas-goenka May 8, 2026

Uh oh!

shreyas-goenka commented May 8, 2026 •

edited

Loading

Uh oh!

github-actions Bot commented May 8, 2026 •

edited

Loading

Uh oh!

eng-dev-ecosystem-bot commented Jun 12, 2026

Uh oh!

Reviewers

Assignees

Labels

Projects

Milestone

Development

Uh oh!

2 participants

Conversation

shreyas-goenka commented May 4, 2026 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Summary

Why

Why multipart

End-to-end bundle deploy benchmark (aws-prod-ucws, direct engine)

File-level throughput (aws-prod-ucws, 10 workers, 4 KiB files)

Size caps (verified)

Error mapping

Test plan

Uh oh!

shreyas-goenka May 8, 2026

Choose a reason for hiding this comment

Uh oh!

shreyas-goenka commented May 8, 2026 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Uh oh!

github-actions Bot commented May 8, 2026 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Approval status: pending

/acceptance/bundle/ - needs approval

/libs/filer/ - needs approval

General files (require maintainer)

Uh oh!

eng-dev-ecosystem-bot commented Jun 12, 2026

Integration test report

Uh oh!

Reviewers

Assignees

Labels

Projects

Milestone

Development

Uh oh!

2 participants

shreyas-goenka commented May 4, 2026 •

edited

Loading

End-to-end `bundle deploy` benchmark (aws-prod-ucws, direct engine)

shreyas-goenka commented May 8, 2026 •

edited

Loading

github-actions Bot commented May 8, 2026 •

edited

Loading

`/acceptance/bundle/` - needs approval

`/libs/filer/` - needs approval