...
Used to check all the features work correctly after any changes.
Case | Covered by |
---|---|
Any |
Used to check important features behavior
Case | Covered by | ||||||||||
---|---|---|---|---|---|---|---|---|---|---|---|
control.sh | --activate: Done --baseline: Done --deactivate: Done --tx:
--snapshot: --restore: | ||||||||||
Thin JDBC driver coverage |
| ||||||||||
ODBC driver coverage | |||||||||||
Metrics coverage | |||||||||||
JMX coverage | |||||||||||
Thin Client API coverage | |||||||||||
Cellular affinity distribution is cellular | Done |
Used to check integrations with other projects (eg. Zookeeper, Kafka, Spark, etc).
Case | Covered by |
---|---|
Used to check migrations from one version/vendor to another.
Case | Covered by |
---|---|
Persistence based Rolling Upgrade | |
Thin client compatibility (backward and forward) |
Used to check some concepts or to automate MVP demonstration.
Case | Covered by |
---|---|
Any | Covered on demand |
Used to check data consistency during the typical load.
Case | Covered by |
---|---|
Any | Jepsen? |
Used to check results correctness during mixed load.
Case | Covered by |
---|---|
Used to check system behavior on crashes.
Case | Covered by | ||||||||||
---|---|---|---|---|---|---|---|---|---|---|---|
Latency degradation on Switch on node left/fail | PME: PME-free: Done Cellular:
| ||||||||||
Failure detection duration | TCPDiscovery: Zookeeper: | ||||||||||
Various rebalance modes (historical, regular, etc) | |||||||||||
Cluster coordinator crash | |||||||||||
Transaction coordinator crash |
Used to check measure duration of important precesses (eg. PME, Snapshot, Rebalance, Node start, etc)
Case | Covered by |
---|---|
Rebalance duration | |
Snapshot duration |
Used to check customer's profile performance on some environment.
Case | Covered by |
---|---|
Any | Covered on demand |
Used to compare overall performance from version to version and from commit to commit
Case | Covered by |
---|---|
The operation's through/latency | Yardstick |