8037f43d00
* Don't block on vschema channel in case of heavy vschema changes and vstream load
Signed-off-by: Rohit Nayak <rohit@planetscale.com>
* Prevent VStreamer engine deadlocks during state transitions
The VStreamer engine is somewhat unusual in two ways:
1. It is open and running on replica tablets rather than only
running on primary tablets.
2. It has no controllers so the main engine mutex is widely shared.
Because of this, when a tablet has open vstreams (direct binary log
streams) performing work and a state transition starts, it can
deadlock with the tabletmanager's state lock when checking if
the engine is open or not.
Signed-off-by: Matt Lord <mattalord@gmail.com>
* Address review comments
Signed-off-by: Rohit Nayak <rohit@planetscale.com>
Signed-off-by: Rohit Nayak <rohit@planetscale.com>
Signed-off-by: Matt Lord <mattalord@gmail.com>
Co-authored-by: Matt Lord <mattalord@gmail.com>
|
||
|---|---|---|
| .. | ||
| bin | ||
| client | ||
| templates | ||
| vttest_schema | ||
| README.md | ||
| TestingStrategy.md | ||
| ci_workflow_gen.go | ||
| client_java.sh | ||
| client_jdbc.sh | ||
| client_test.sh | ||
| config.json | ||
| config_partial_keyspace.json | ||
| local_example.sh | ||
| region_example.sh | ||
| vthook-make_mycnf | ||
| vthook-test.sh | ||
| vthook-test_backup_error | ||
| vthook-test_backup_transform | ||
| vtop_example.sh | ||
README.md
##Github CI Workflows
This document has a short outline of how tests are run in CI, how to add new tests and where these are configured.
Adding a new test
Unit tests are run by the unit test runner, one per platform, currently mysql57, mysql80, mariadb101, mariadb102, mariadb103.
The workflow first installs the required database server before calling make unit_test.
To add a new end-to-end (e2e) test (also called cluster end to end tests):
- Add a new object to test/config.json
- If you are creating a new test shard:
- update
clusterListinci_workflow_gen.go make generate_ci_workflows
- update
- If you are adding a new database platform, update the
templates\unit_test.tplto add the platform specific packages and updateunitTestDatabases
Vitess test runner
The .github/workflows directory contains one yaml file per workflow. e2e tests are run using the test.go script
in the repository root.
This script invokes the vitess e2e test framework using a json configuration file test/config.json which has one object per test.
Each test is of the form:
"vtgate": {
"File": "unused.go",
"Args": ["vitess.io/vitess/go/test/endtoend/vtgate"],
"Command": [],
"Manual": false,
"Shard": 17,
"RetryMax": 0,
"Tags": []
},
The important parameters here are Args which define the arguments to go test and the Shard which says
which Test VM should run this test. All tests which have a common Shard value are run in the same test vm.
Known Issue
- Each VM does not seem to be able to create a lot of vttablets. For this reason we have had to split a few VReplication e2e tests across Shards. We need to identify and if possible fix this limitation so that we can reduce the number of test Shards