JustRelax.jl

Public

https://github.com/PTsolvers/JustRelax.jl

Merge branch 'main' into adm/sinkingballs

#1182
/
adm/sinkingballs/64af94f
Failed in 10h 35m
:buildkite: Pipeline upload
🏃 Dynamically launch Pipelines
CUDA Julia 1.10
CUDA Julia 1.11
AMDGPU Julia 1.10
AMDGPU Julia 1.11
Pascal Aellig
Pascal Aellig
Created
Triggered from Webhook
Parallel Workload (0dt deploy) succeeded with known error logs, main history: :bk-status-passed::bk-status-passed::bk-status-passed::bk-status-passed::bk-status-passed: 
parallel-workload-materialized2-1    | 2025-04-02T23:48:26.014846Z  thread 'coordinator' panicked at src/storage-controller/src/lib.rs:973:17: dependency since has advanced past dependent (u172) upper
Test details & reproducer Runs a randomized parallel workload stressing all parts of Materialize, can mostly find panics and unexpected errors. See zippy for a sequential randomized tests which can verify correctness. 
bin/mzcompose --find parallel-workload run default --runtime=1500 --scenario=0dt-deploy --threads=16

SQLsmith explain failed, main history: :bk-status-failed::bk-status-passed::bk-status-passed::bk-status-passed::bk-status-passed:

sqlsmith-mz_1-1  | 2025-04-02T23:43:45.853702Z  thread 'tokio:work-37' panicked at src/expr/src/scalar/func.rs:2154:31: Datum::unwrap_string called on Float64(10.0)

sqlsmith-mz_1-1  | 2025-04-02T23:52:22.761235Z  thread 'tokio:work-6' panicked at src/expr/src/scalar/func.rs:2154:31: Datum::unwrap_string called on Float64(1e-307)

sqlsmith-mz_1-1  | 2025-04-02T23:52:56.697187Z  thread 'tokio:work-6' panicked at src/expr/src/scalar/func.rs:2154:31: Datum::unwrap_string called on True

sqlsmith-mz_2-1  | 2025-04-02T23:55:38.895692Z  thread 'tokio:work-1' panicked at src/expr/src/scalar/func.rs:2154:31: Datum::unwrap_string called on UInt16(0)

sqlsmith-mz_2-1  | 2025-04-03T00:02:32.869962Z  thread 'tokio:work-2' panicked at src/expr/src/scalar/func.rs:2154:31: Datum::unwrap_string called on False

sqlsmith-mz_1-1  | 2025-04-03T00:04:24.551997Z  thread 'tokio:work-17' panicked at src/expr/src/scalar/func.rs:2154:31: Datum::unwrap_string called on Float32(1.0)

sqlsmith-mz_1-1  | 2025-04-03T00:04:55.346727Z  thread 'tokio:work-7' panicked at src/expr/src/scalar/func.rs:2154:31: Datum::unwrap_string called on False
Test details & reproducer Use SQLsmith to generate random queries (AST/code based) and run them against Materialize: https://github.com/MaterializeInc/sqlsmith The queries can be complex, but we can't verify correctness or performance. 
bin/mzcompose --find sqlsmith run default --max-joins=15 --explain-only --runtime=1500

Canary Deploy in Staging Cloud failed, main history: :bk-status-failed::bk-status-failed::bk-status-failed::bk-status-failed::bk-status-failed:

builtins.ValueError: Redpanda API call failed: 400 {"code":"INVALID_ARGUMENT","message":"throughput tier tier-1-aws-v2-arm is not available in region us-east-1 for cluster type TYPE_DEDICATED","details":[{"@type":"google.rpc.ErrorInfo","reason":"REASON_THROUGHPUT_TIER_NOT_AVAILABLE_IN_REGION","domain":"redpanda.com/controlplane","metadata":{"cluster_type":"TYPE_DEDICATED","region":"us-east-1","throughput_tier_name":"tier-1-aws-v2-arm"}}]}
Test details & reproducer Deploy the current version on a real Staging Cloud, and run some basic verifications, like ingesting data from Kafka and Redpanda Cloud using AWS Privatelink. Runs only on main and release branches. 
bin/mzcompose --find cloud-canary run default

💡 SQL logic tests 8 failed, main history: :bk-status-passed::bk-status-passed::bk-status-passed::bk-status-passed::bk-status-passed:

OutputFailure:test/sqllogictest/cardinality.slt:433
        expected: Values(["Explained Query:\n  Project (#0{x}, #1{y}, #0{x}, #3{y}, #0{x}, #5{y}, #0{x}, #7{y}, #0{x}, #9{y}, #0{x}, #11{y}, #0{x}, #13{y}, #0{x}, #15{y}, #0{x}, #17{y}, #0{x}, #19{y})\n    Join on=(#0{x} = #2{x} = #4{x} = #6{x} = #8{x} = #10{x} = #12{x} = #14{x} = #16{x} = #18{x}) type=delta\n      implementation\n        %0:t » %1:t2[#0]KA|10000| » %2:t3[#0]K|169| » %9:t10[#0]K|260| » %8:t9[#0]K|273| » %5:t6[#0]K|299| » %6:t7[#0]K|299| » %7:t8[#0]K|299| » %4:t5[#0]K|494| » %3:t4[#0]K|611|\n        %1:t2 » %0:t[#0]KA|4| » %2:t3[#0]K|169| » %9:t10[#0]K|260| » %8:t9[#0]K|273| » %5:t6[#0]K|299| » %6:t7[#0]K|299| » %7:t8[#0]K|299| » %4:t5[#0]K|494| » %3:t4[#0]K|611|\n        %2:t3 » %0:t[#0]KA|4| » %1:t2[#0]KA|10000| » %9:t10[#0]K|260| » %8:t9[#0]K|273| » %5:t6[#0]K|299| » %6:t7[#0]K|299| » %7:t8[#0]K|299| » %4:t5[#0]K|494| » %3:t4[#0]K|611|\n        %3:t4 » %0:t[#0]KA|4| » %1:t2[#0]KA|10000| » %2:t3[#0]K|169| » %9:t10[#0]K|260| » %8:t9[#0]K|273| » %5:t6[#0]K|299| » %6:t7[#0]K|299| » %7:t8[#0]K|299| » %4:t5[#0]K|494|\n        %4:t5 » %0:t[#0]KA|4| » %1:t2[#0]KA|10000| » %2:t3[#0]K|169| » %9:t10[#0]K|260| » %8:t9[#0]K|273| » %5:t6[#0]K|299| » %6:t7[#0]K|299| » %7:t8[#0]K|299| » %3:t4[#0]K|611|\n        %5:t6 » %0:t[#0]KA|4| » %1:t2[#0]KA|10000| » %2:t3[#0]K|169| » %9:t10[#0]K|260| » %8:t9[#0]K|273| » %6:t7[#0]K|299| » %7:t8[#0]K|299| » %4:t5[#0]K|494| » %3:t4[#0]K|611|\n        %6:t7 » %0:t[#0]KA|4| » %1:t2[#0]KA|10000| » %2:t3[#0]K|169| » %9:t10[#0]K|260| » %8:t9[#0]K|273| » %5:t6[#0]K|299| » %7:t8[#0]K|299| » %4:t5[#0]K|494| » %3:t4[#0]K|611|\n        %7:t8 » %0:t[#0]KA|4| » %1:t2[#0]KA|10000| » %2:t3[#0]K|169| » %9:t10[#0]K|260| » %8:t9[#0]K|273| » %5:t6[#0]K|299| » %6:t7[#0]K|299| » %4:t5[#0]K|494| » %3:t4[#0]K|611|\n        %8:t9 » %0:t[#0]KA|4| » %1:t2[#0]KA|10000| » %2:t3[#0]K|169| » %9:t10[#0]K|260| » %5:t6[#0]K|299| » %6:t7[#0]K|299| » %7:t8[#0]K|299| » %4:t5[#0]K|494| » %3:t4[#0]K|611|\n        %9:t10 » %0:t[#0]KA|4| » %1:t2[#0]KA|10000| » %2:t3[#0]K|169| » %8:t9[#0]K|273| » %5:t6[#0]K|299| » %6:t7[#0]K|299| » %7:t8[#0]K|299| » %4:t5[#0]K|494| » %3:t4[#0]K|611|\n      ArrangeBy keys=[[#0{x}]]\n        ReadIndex on=t t_x=[delta join 1st input (full scan)]\n      ArrangeBy keys=[[#0{x}]]\n        ReadIndex on=t2 tt_x=[delta join lookup]\n      ArrangeBy keys=[[#0{x}]]\n        ReadStorage materialize.public.t3\n      ArrangeBy keys=[[#0{x}]]\n        ReadStorage materialize.public.t4\n      ArrangeBy keys=[[#0{x}]]\n        ReadStorage materialize.public.t5\n      ArrangeBy keys=[[#0{x}]]\n        ReadStorage materialize.public.t6\n      ArrangeBy keys=[[#0{x}]]\n        ReadStorage materialize.public.t7\n      ArrangeBy keys=[[#0{x}]]\n        ReadStorage materialize.public.t8\n      ArrangeBy keys=[[#0{x}]]\n        ReadStorage materialize.public.t9\n      ArrangeBy keys=[[#0{x}]]\n        ReadStorage materialize.public.t10\n\nSource materialize.public.t3\nSource materialize.public.t4\nSource materialize.public.t5\nSource materialize.public.t6\nSource materialize.public.t7\nSource materialize.public.t8\nSource materialize.public.t9\nSource materialize.public.t10\n\nUsed Indexes:\n  - materialize.public.t_x (delta join 1st input (full scan))\n  - materialize.public.tt_x (delta join lookup)\n\nTarget cluster: quickstart\n"])
        actually: Values(["Explained Query:\n  Project (#0{x}, #1{y}, #0{x}, #3{y}, #0{x}, #5{y}, #0{x}, #7{y}, #0{x}, #9{y}, #0{x}, #11{y}, #0{x}, #13{y}, #0{x}, #15{y}, #0{x}, #17{y}, #0{x}, #19{y})\n    Join on=(#0{x} = #2{x} = #4{x} = #6{x} = #8{x} = #10{x} = #12{x} = #14{x} = #16{x} = #18{x}) type=delta\n      implementation\n        %0:t » %1:t2[#0]KA » %2:t3[#0]K » %3:t4[#0]K » %4:t5[#0]K » %5:t6[#0]K » %6:t7[#0]K » %7:t8[#0]K » %8:t9[#0]K » %9:t10[#0]K\n        %1:t2 » %0:t[#0]KA » %2:t3[#0]K » %3:t4[#0]K » %4:t5[#0]K » %5:t6[#0]K » %6:t7[#0]K » %7:t8[#0]K » %8:t9[#0]K » %9:t10[#0]K\n        %2:t3 » %0:t[#0]KA » %1:t2[#0]KA » %3:t4[#0]K » %4:t5[#0]K » %5:t6[#0]K » %6:t7[#0]K » %7:t8[#0]K » %8:t9[#0]K » %9:t10[#0]K\n        %3:t4 » %0:t[#0]KA » %1:t2[#0]KA » %2:t3[#0]K » %4:t5[#0]K » %5:t6[#0]K » %6:t7[#0]K » %7:t8[#0]K » %8:t9[#0]K » %9:t10[#0]K\n        %4:t5 » %0:t[#0]KA » %1:t2[#0]KA » %2:t3[#0]K » %3:t4[#0]K » %5:t6[#0]K » %6:t7[#0]K » %7:t8[#0]K » %8:t9[#0]K » %9:t10[#0]K\n        %5:t6 » %0:t[#0]KA » %1:t2[#0]KA » %2:t3[#0]K » %3:t4[#0]K » %4:t5[#0]K » %6:t7[#0]K » %7:t8[#0]K » %8:t9[#0]K » %9:t10[#0]K\n        %6:t7 » %0:t[#0]KA » %1:t2[#0]KA » %2:t3[#0]K » %3:t4[#0]K » %4:t5[#0]K » %5:t6[#0]K » %7:t8[#0]K » %8:t9[#0]K » %9:t10[#0]K\n        %7:t8 » %0:t[#0]KA » %1:t2[#0]KA » %2:t3[#0]K » %3:t4[#0]K » %4:t5[#0]K » %5:t6[#0]K » %6:t7[#0]K » %8:t9[#0]K » %9:t10[#0]K\n        %8:t9 » %0:t[#0]KA » %1:t2[#0]KA » %2:t3[#0]K » %3:t4[#0]K » %4:t5[#0]K » %5:t6[#0]K » %6:t7[#0]K » %7:t8[#0]K » %9:t10[#0]K\n        %9:t10 » %0:t[#0]KA » %1:t2[#0]KA » %2:t3[#0]K » %3:t4[#0]K » %4:t5[#0]K » %5:t6[#0]K » %6:t7[#0]K » %7:t8[#0]K » %8:t9[#0]K\n      ArrangeBy keys=[[#0{x}]]\n        ReadIndex on=t t_x=[delta join 1st input (full scan)]\n      ArrangeBy keys=[[#0{x}]]\n        ReadIndex on=t2 tt_x=[delta join lookup]\n      ArrangeBy keys=[[#0{x}]]\n        ReadStorage materialize.public.t3\n      ArrangeBy keys=[[#0{x}]]\n        ReadStorage materialize.public.t4\n      ArrangeBy keys=[[#0{x}]]\n        ReadStorage materialize.public.t5\n      ArrangeBy keys=[[#0{x}]]\n        ReadStorage materialize.public.t6\n      ArrangeBy keys=[[#0{x}]]\n        ReadStorage materialize.public.t7\n      ArrangeBy keys=[[#0{x}]]\n        ReadStorage materialize.public.t8\n      ArrangeBy keys=[[#0{x}]]\n        ReadStorage materialize.public.t9\n      ArrangeBy keys=[[#0{x}]]\n        ReadStorage materialize.public.t10\n\nSource materialize.public.t3\nSource materialize.public.t4\nSource materialize.public.t5\nSource materialize.public.t6\nSource materialize.public.t7\nSource materialize.public.t8\nSource materialize.public.t9\nSource materialize.public.t10\n\nUsed Indexes:\n  - materialize.public.t_x (delta join 1st input (full scan))\n  - materialize.public.tt_x (delta join lookup)\n\nTarget cluster: quickstart\n"])
        actual raw: [Row { columns: [Column { name: "Optimized Plan", table_oid: None, column_id: None, type: Text }] }]
Test details & reproducer Run SQL tests using an instance of Mz that is embedded in the sqllogic binary itself. Good for basic SQL tests, but can't interact with sources like MySQL/Kafka, see Testdrive for that. 
BUILDKITE_PARALLEL_JOB=7 BUILDKITE_PARALLEL_JOB_COUNT=10 bin/mzcompose --find sqllogictest run slow-tests
Matrix
CUDA Julia 1.10julia -e 'println("--- :julia: Instantiating project") && using Pkg && Pkg.develop(; path=pwd())' || exit 3 && julia -e 'println("+++ :julia: Running tests") && using Pkg && Pkg.test("JustRelax"; test_args=["--backend=CUDA"], coverage=true)'
Waited 10s
·
Ran in 29m 55s
Agent
Matrix
CUDA Julia 1.11julia -e 'println("--- :julia: Instantiating project") && using Pkg && Pkg.develop(; path=pwd())' || exit 3 && julia -e 'println("+++ :julia: Running tests") && using Pkg && Pkg.test("JustRelax"; test_args=["--backend=CUDA"], coverage=true)'
Waited 6s
·
Ran in 43m 29s
Agent
Matrix
AMDGPU Julia 1.10julia -e 'println("--- :julia: Instantiating project") && using Pkg && Pkg.develop(; path=pwd())' || exit 3 && julia -e 'println("+++ :julia: Running tests") && using Pkg && Pkg.test("JustRelax"; test_args=["--backend=AMDGPU"], coverage=true)'
Waited 2h 54m
·
Ran in 13m 14s
Agent
Matrix
AMDGPU Julia 1.11julia -e 'println("--- :julia: Instantiating project") && using Pkg && Pkg.develop(; path=pwd())' || exit 3 && julia -e 'println("+++ :julia: Running tests") && using Pkg && Pkg.test("JustRelax"; test_args=["--backend=AMDGPU"], coverage=true)'
Waited 9h 57m
·
Ran in 38m 10s
Agent
Total Job Run Time: 2h 4m