Repository navigation
Add kubernetes integration tests - #9581
Draft
parshimers wants to merge 5 commits into
Draft
parshimers wants to merge 5 commits into
parshimers wants to merge 5 commits into
Conversation
Creating a per-user warehouse registers it in Lakekeeper with the webserver's own S3 endpoint and credentials. The chart never set them on the webserver, so it sent the localhost default, Lakekeeper's storage validation failed against it, and POST /api/warehouse answered 502 on every Kubernetes deployment. Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
The webserver, file-service, workflow-compiling-service and the computing-unit manager had no readiness probe, so Kubernetes marked them ready as soon as the container started and the gateway routed requests to a JVM that was not listening yet, answering 503. Probe /api/healthcheck as config-service and access-control-service already do. Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Nothing exercised the chart as a whole, so defects that only appear once the services talk to each other in a cluster shipped unnoticed; the two preceding fixes were found this way. The test deploys published images rather than building them, which keeps it to a few minutes; it runs after the nightly image build, on chart changes, and on demand. Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Main added the observability env include at the end of each service's env list, where this branch adds a readiness probe; keep both, with the include first so it stays inside env. Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Contributor
Automated Reviewer SuggestionsBased on the
|
Codecov Report✅ All modified and coverable lines are covered by tests. Additional details and impacted files@@ Coverage Diff @@
## main #9581 +/- ##
============================================
- Coverage 91.57% 91.56% -0.01%
Complexity 5158 5158
============================================
Files 1300 1300
Lines 58092 58092
Branches 7615 7615
============================================
- Hits 53195 53192 -3
- Misses 2959 2960 +1
- Partials 1938 1940 +2
☔ View full report in Codecov by Harness. 🚀 New features to boost your workflow:
|
Contributor
|
| config | throughput | MB/s | latency | max Δ latest / 7d | |
|---|---|---|---|---|---|
| 🔴 | bs=10 sw=10 sl=64 | 420 | 0.256 | 23,036/29,340/29,340 us | 🟢 -21.7% / 🔴 +127.3% |
| 🟢 | bs=100 sw=10 sl=64 | 867 | 0.529 | 113,875/131,330/131,330 us | 🟢 -28.2% / 🔴 +50.9% |
| 🔴 | bs=1000 sw=10 sl=64 | 966 | 0.59 | 1,032,596/1,152,136/1,152,136 us | 🔴 +9.8% / 🔴 +41.5% |
Baseline details
Latest main 7f9bdf3 from same runner
| config | metric | PR | latest main | 7d avg | Δ latest | Δ 7d |
|---|---|---|---|---|---|---|
| bs=10 sw=10 sl=64 | throughput | 420 tuples/sec | 449 tuples/sec | 1,019 tuples/sec | -6.5% | -58.8% |
| bs=10 sw=10 sl=64 | MB/s | 0.256 MB/s | 0.274 MB/s | 0.622 MB/s | -6.6% | -58.9% |
| bs=10 sw=10 sl=64 | p50 | 23,036 us | 20,783 us | 10,136 us | +10.8% | +127.3% |
| bs=10 sw=10 sl=64 | p95 | 29,340 us | 37,483 us | 13,860 us | -21.7% | +111.7% |
| bs=10 sw=10 sl=64 | p99 | 29,340 us | 37,483 us | 16,484 us | -21.7% | +78.0% |
| bs=100 sw=10 sl=64 | throughput | 867 tuples/sec | 835 tuples/sec | 1,363 tuples/sec | +3.8% | -36.4% |
| bs=100 sw=10 sl=64 | MB/s | 0.529 MB/s | 0.509 MB/s | 0.832 MB/s | +3.9% | -36.4% |
| bs=100 sw=10 sl=64 | p50 | 113,875 us | 111,129 us | 79,194 us | +2.5% | +43.8% |
| bs=100 sw=10 sl=64 | p95 | 131,330 us | 182,878 us | 87,022 us | -28.2% | +50.9% |
| bs=100 sw=10 sl=64 | p99 | 131,330 us | 182,878 us | 106,894 us | -28.2% | +22.9% |
| bs=1000 sw=10 sl=64 | throughput | 966 tuples/sec | 986 tuples/sec | 1,407 tuples/sec | -2.0% | -31.3% |
| bs=1000 sw=10 sl=64 | MB/s | 0.59 MB/s | 0.602 MB/s | 0.859 MB/s | -2.0% | -31.3% |
| bs=1000 sw=10 sl=64 | p50 | 1,032,596 us | 1,016,235 us | 776,422 us | +1.6% | +33.0% |
| bs=1000 sw=10 sl=64 | p95 | 1,152,136 us | 1,048,940 us | 814,091 us | +9.8% | +41.5% |
| bs=1000 sw=10 sl=64 | p99 | 1,152,136 us | 1,048,940 us | 860,577 us | +9.8% | +33.9% |
Raw CSV
config_idx,batch_size,schema_width,string_len,num_batches,total_ms,total_tuples,total_bytes,tuples_per_sec,mb_per_sec,lat_p50_us,lat_p95_us,lat_p99_us
0,10,10,64,20,476.42,200,128000,420,0.256,23036.44,29340.44,29340.44
1,100,10,64,20,2305.95,2000,1280000,867,0.529,113874.76,131330.32,131330.32
2,1000,10,64,20,20705.78,20000,12800000,966,0.590,1032596.39,1152136.20,1152136.20This branch has not been deployed
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
What changes were proposed in this PR?
This adds some integration tests for the helm charts we have, using k3d.
Before this, there wouldn't really be any way to test a change that might affect this area.
That's not ideal, because it is the only officially supported way to deploy Texera
Any related issues, documentation, discussions?
Not yet. Probably somewhere.
How was this PR tested?
It itself is a test, so it either works or it doesn't.
Was this PR authored or co-authored using generative AI tooling?
Yes, Claude Opus 5.5