Gemini is an automatic random testing tool for Scylla.
Usage:
gemini [flags]
Flags:
--async-objects-stabilization-attempts int Maximum number of attempts to validate result sets from MV and SI (default 10)
--async-objects-stabilization-backoff duration Duration between attempts to validate result sets from MV and SI for example 10ms or 1s (default 10ms)
-b, --bind string Specify the interface and port which to bind prometheus metrics on. Default is ':2112' (default ":2112")
-c, --concurrency uint Number of threads per table to run concurrently (default 10)
--consistency string Specify the desired consistency as ANY|ONE|TWO|THREE|QUORUM|LOCAL_QUORUM|EACH_QUORUM|LOCAL_ONE (default "QUORUM")
--cql-features string Specify the type of cql features to use, basic|normal|all (default "basic")
--dataset-size string Specify the type of dataset size to use, small|large (default "large")
-d, --drop-schema Drop schema before starting tests run
--duration duration (default 30s)
-f, --fail-fast Stop on the first failure
-h, --help help for gemini
--level string Specify the logging level, debug|info|warn|error|dpanic|panic|fatal (default "info")
--max-clustering-keys int Maximum number of generated clustering keys (default 4)
--max-columns int Maximum number of generated columns (default 16)
--max-mutation-retries int Maximum number of attempts to apply a mutation (default 2)
--max-mutation-retries-backoff duration Duration between attempts to apply a mutation for example 10ms or 1s (default 10ms)
--max-partition-keys int Maximum number of generated partition keys (default 6)
--max-tables int Maximum number of generated tables (default 1)
--min-clustering-keys int Minimum number of generated clustering keys (default 2)
--min-columns int Minimum number of generated columns (default 8)
--min-partition-keys int Minimum number of generated partition keys (default 2)
-m, --mode string Query operation mode. Mode options: write, read, mixed (default) (default "mixed")
--non-interactive Run in non-interactive mode (disable progress indicator)
--normal-dist-mean float Mean of the normal distribution (default 9.223372036854776e+18)
--normal-dist-sigma float Sigma of the normal distribution, defaults to one standard deviation ~0.341 (default 6.290339729134958e+18)
-o, --oracle-cluster strings Host names or IPs of the oracle cluster that provides correct answers. If omitted no oracle will be used
--oracle-replication-strategy string Specify the desired replication strategy of the oracle cluster as either the coded short hand simple|network to get the default for each type or provide the entire specification in the form {'class':'....'} (default "simple")
--outfile string Specify the name of the file where the results should go
--partition-key-buffer-reuse-size uint Number of reused buffered partition keys (default 100)
--partition-key-distribution string Specify the distribution from which to draw partition keys, supported values are currently uniform|normal|zipf (default "uniform")
--replication-strategy string Specify the desired replication strategy as either the coded short hand simple|network to get the default for each type or provide the entire specification in the form {'class':'....'} (default "simple")
--schema string Schema JSON config file
-s, --seed uint PRNG seed value (default 1)
--table-options stringArray Repeatable argument to set table options to be added to the created tables
-t, --test-cluster strings Host names or IPs of the test cluster that is system under test
--token-range-slices uint Number of slices to divide the token space into (default 10000)
--tracing-outfile string Specify the file to which tracing information gets written. Two magic names are available, 'stdout' and 'stderr'. By default tracing is disabled.
--use-counters Ensure that at least one table is a counter table
-v, --verbose Verbose output during test run
--version version for gemini
--warmup duration Specify the warmup perid as a duration for example 30s or 10h (default 30s)