search

timescale / tsbs

Time Series Benchmark Suite, a tool for comparing and evaluating databases for time series data
MIT License
1.29k stars 300 forks source link

Cassandra tsbs_run_queries_cassandra error: can not unmarshal bigint into *float64 #127

Open seybi87 opened 4 years ago

seybi87 commented 4 years ago

Hi guys,

after fixing https://github.com/timescale/tsbs/issues/124 I run into another issue with executing TSBS against a Cassandra instance.

As outlined in https://github.com/timescale/tsbs/issues/124, the data is loaded in Cassandra and the following dummy queries are generated:

tsbs_generate_queries --use-case "cpu-only" --seed "123" --scale "1000" --timestamp-start "2016-01-01T00:00:00Z" --timestamp-end "2016-01-01T01:00:01Z" --queries "10" --query-type "single-groupby-1-1-1" --format "cassandra" --file "/tmp/cassandra-queries"

These queries are executed with the following command: tsbs_run_queries_cassandra --aggregation-plan "server" --db-name "benchmark" --workers "1" --host "1.2.3.4" --file "/tmp/cassandra-queries" --debug "1" --print-responses

This ends in the following error message:

[hlqe] Do: HumanLabel: Cassandra 1 cpu metric(s), random    1 hosts, random 1h0m0s by 1m, HumanDescription: Cassandra 1 cpu metric(s), random    1 hosts, random 1h0m0s by 1m: 2016-01-01T00:00:00Z, MeasurementName: cpu, AggregationType: max, TimeStart: 2016-01-01 00:00:00.646325489 +0000 UTC, TimeEnd: 2016-01-01 01:00:00.646325489 +0000 UTC, GroupByDuration: 1m0s, TagSets: [[hostname=host_249]]
[hlqe] query planning took 6.979200ms
[qpsa] query with server aggregation plan has 61 CQLQuery objects
panic: can not unmarshal bigint into *float64

goroutine 2603 [running]:
github.com/timescale/tsbs/query.(*BenchmarkRunner).processorHandler(0xc0001044d0, 0xc00019e450, 0xc00061c2d0, 0xf9fba0, 0xc33d00, 0xc00004c3e0, 0x0)
    tsbs/query/benchmarker.go:196 +0x530
created by github.com/timescale/tsbs/query.(*BenchmarkRunner).Run
    tsbs/query/benchmarker.go:156 +0x2a2

Any help is much appreciated!

seybi87 commented 4 years ago

I debugged the code the calls a bit and it looks like the following line causes the error when closing the Iterator : https://github.com/timescale/tsbs/blob/master/cmd/tsbs_run_queries_cassandra/query_plan.go#L74

Yet, I do not see why closing the iterator causes this issue.

Without closing the Iterator the query execution works (for the simple example posted above) but will this cause any additional issues later on?