CherryPick(vitessio#13799): Endtoend: stress tests for VTGate FOREIGN…

… KEY support (vitessio#3199)

* cherry pick of 13799

* resolved conflict

Signed-off-by: Shlomi Noach <2607934+shlomi-noach@users.noreply.github.com>

* make generate_ci_workflows

Signed-off-by: Shlomi Noach <2607934+shlomi-noach@users.noreply.github.com>

---------

Signed-off-by: Shlomi Noach <2607934+shlomi-noach@users.noreply.github.com>
Co-authored-by: Shlomi Noach <2607934+shlomi-noach@users.noreply.github.com>

.github/workflows/cluster_endtoend_vtgate_foreignkey_stress.yml

@@ -0,0 +1,82 @@
+# DO NOT MODIFY: THIS FILE IS GENERATED USING "make generate_ci_workflows"
+
+name: Cluster (vtgate_foreignkey_stress)
+on: [push]
+
+jobs:
+ build:
+ name: Run endtoend tests on Cluster (vtgate_foreignkey_stress)
+ runs-on: ubuntu-shr
+ timeout-minutes: 60
+
+ steps:
+ - name: Configure git private repo access
+ env:
+ GITHUB_TOKEN: ${{ secrets.PLANETSCALE_ACTIONS_BOT_TOKEN }}
+ run: |
+ git config --global --add url."https://${GITHUB_TOKEN}@github.com/".insteadOf "https://github.com/"
+
+ - name: Set up Go
+ uses: actions/setup-go@v4
+ with:
+ go-version: 1.21.1
+
+ - name: Set up python
+ uses: actions/setup-python@v4
+
+ - name: Tune the OS
+ run: |
+ sudo sysctl -w net.ipv4.ip_local_port_range="22768 65535"
+ # Increase the asynchronous non-blocking I/O. More information at https://dev.mysql.com/doc/refman/5.7/en/innodb-parameters.html#sysvar_innodb_use_native_aio
+ echo "fs.aio-max-nr = 1048576" | sudo tee -a /etc/sysctl.conf
+ sudo sysctl -p /etc/sysctl.conf
+
+ - name: Check out code
+ uses: actions/checkout@v3
+
+ - name: Get dependencies
+ env: # Or as an environment variable
+ AWS_ACCESS_KEY_ID: ${{ secrets.BUILDKITE_S3_ACCESS_KEY_ID }}
+ AWS_SECRET_ACCESS_KEY: ${{ secrets.BUILDKITE_S3_SECRET_ACCESS_KEY }}
+ AWS_DEFAULT_REGION: us-east-1
+ run: |
+ sudo apt-get update
+ # stop any existing running instance of mysql
+ sudo service mysql stop
+ sudo ln -s /etc/apparmor.d/usr.sbin.mysqld /etc/apparmor.d/disable/
+ sudo apparmor_parser -R /etc/apparmor.d/usr.sbin.mysqld
+
+ # Uninstall any previously installed MySQL first
+ sudo DEBIAN_FRONTEND="noninteractive" apt-get remove -y --purge mysql-server mysql-client mysql-common
+ sudo apt-get -y autoremove
+ sudo apt-get -y autoclean
+ sudo deluser mysql
+ sudo rm -rf /var/lib/mysql
+ sudo rm -rf /etc/mysql
+
+ # install necessary tools
+ sudo apt-get install -y make unzip g++ etcd curl git wget awscli eatmydata xz-utils libncurses5
+ sudo service etcd stop
+
+ # Get latest version of mysql from s3 bucket
+ LATEST_BUILD=$(aws s3api list-objects-v2 --bucket "planetscale-mysql-server-private-ci-artifacts" --prefix mysql/main/dist --query 'reverse(sort_by(Contents[?contains(Key, `jammy`)], &LastModified))[:1].Key' --output=text)
+ echo "latest build is $LATEST_BUILD"
+ # Pin this to 8.0.34
+ LAST_BUILD="mysql/main/dist/mysql-8.0.34.20230901-ps-4dc8f581f5f-jammy-linux-x86_64.tar.gz"
+ echo "installing psdb mysql $LAST_BUILD"
+ aws s3 cp "s3://planetscale-mysql-server-private-ci-artifacts/${LAST_BUILD}" .
+ sudo tar xf $(basename $LAST_BUILD) -v -C /usr --strip-components=1
+
+ go mod download
+
+ - name: Run cluster endtoend test
+ timeout-minutes: 45
+ run: |
+ # We set the VTDATAROOT to the /tmp folder to reduce the file path of mysql.sock file
+ # which musn't be more than 107 characters long.
+ export VTDATAROOT="/tmp/"
+ source build.env
+
+ set -exo pipefail
+
+ eatmydata -- go run test.go -docker=false -follow -shard vtgate_foreignkey_stress

go.mod

@@ -127,6 +127,7 @@ require (
  go.uber.org/goleak v1.2.1
  go.uber.org/multierr v1.11.0
  go.uber.org/zap v1.24.0
+ golang.org/x/exp v0.0.0-20230817173708-d852ddb80c63
  golang.org/x/sync v0.3.0
  gopkg.in/src-d/go-errors.v1 v1.0.0
  modernc.org/mathutil v1.5.0

go.sum

@@ -723,6 +723,8 @@ golang.org/x/exp v0.0.0-20191227195350-da58074b4299/go.mod h1:2RIsYlXP63K8oxa1u0
 golang.org/x/exp v0.0.0-20200119233911-0405dc783f0a/go.mod h1:2RIsYlXP63K8oxa1u096TMicItID8zy7Y6sNkU49FU4=
 golang.org/x/exp v0.0.0-20200207192155-f17229e696bd/go.mod h1:J/WKrq2StrnmMY6+EHIKF9dgMWnmCNThgcyBT1FY9mM=
 golang.org/x/exp v0.0.0-20200224162631-6cc2880d07d6/go.mod h1:3jZMyOhIsHpP37uCMkUooju7aAi5cS1Q23tOzKc+0MU=
+golang.org/x/exp v0.0.0-20230817173708-d852ddb80c63 h1:m64FZMko/V45gv0bNmrNYoDEq8U5YUhetc9cBWKS1TQ=
+golang.org/x/exp v0.0.0-20230817173708-d852ddb80c63/go.mod h1:0v4NqG35kSWCMzLaMeX+IQrlSnVE/bqGSyC2cz/9Le8=
 golang.org/x/image v0.0.0-20190227222117-0694c2d4d067/go.mod h1:kZ7UVZpmo3dzQBMxlp+ypCbDeSB+sBbTgSJuh5dn5js=
 golang.org/x/image v0.0.0-20190802002840-cff245a6509b/go.mod h1:FeLwcggjj3mMvU+oOTbSwawSJRM1uh48EjtB4UJZlP0=
 golang.org/x/lint v0.0.0-20181026193005-c67002cb31c3/go.mod h1:UVdnD1Gm6xHRNCYTkRU2/jEulfH38KcIWyp/GAMgvoE=

go/test/endtoend/cluster/cluster_util.go

@@ -223,26 +223,53 @@ func filterResultWhenRunsForCoverage(input string) string {
  return result
 }
 
+func ValidateReplicationIsHealthy(t *testing.T, tablet *Vttablet) bool {
+ query := "show replica status"
+ rs, err := tablet.VttabletProcess.QueryTablet(query, "", true)
+ assert.NoError(t, err)
+ row := rs.Named().Row()
+ require.NotNil(t, row)
+
+ ioRunning := row.AsString("Replica_IO_Running", "")
+ require.NotEmpty(t, ioRunning)
+ ioHealthy := assert.Equalf(t, "Yes", ioRunning, "Replication is broken. Replication status: %v", row)
+ sqlRunning := row.AsString("Replica_SQL_Running", "")
+ require.NotEmpty(t, sqlRunning)
+ sqlHealthy := assert.Equalf(t, "Yes", sqlRunning, "Replication is broken. Replication status: %v", row)
+
+ return ioHealthy && sqlHealthy
+}
+
 // WaitForReplicationPos will wait for replication position to catch-up
-func WaitForReplicationPos(t *testing.T, tabletA *Vttablet, tabletB *Vttablet, hostname string, timeout float64) {
+func WaitForReplicationPos(t *testing.T, tabletA *Vttablet, tabletB *Vttablet, validateReplication bool, timeout time.Duration) {
+ hostname := "localhost"
+ ctx, cancel := context.WithTimeout(context.Background(), timeout)
+ defer cancel()
+ ticker := time.NewTicker(10 * time.Millisecond)
+ defer ticker.Stop()
+
  replicationPosA, _ := GetPrimaryPosition(t, *tabletA, hostname)
  for {
+ if validateReplication {
+ if !ValidateReplicationIsHealthy(t, tabletB) {
+ assert.FailNowf(t, "Replication broken on tablet %v. Will not wait for position", tabletB.Alias)
+ }
+ if t.Failed() {
+ return
+ }
+ }
  replicationPosB, _ := GetPrimaryPosition(t, *tabletB, hostname)
  if positionAtLeast(t, tabletA, replicationPosB, replicationPosA) {
- break
+ return
  }
  msg := fmt.Sprintf("%s's replication position to catch up to %s's;currently at: %s, waiting to catch up to: %s", tabletB.Alias, tabletA.Alias, replicationPosB, replicationPosA)
- waitStep(t, msg, timeout, 0.01)
- }
-}
-
-func waitStep(t *testing.T, msg string, timeout float64, sleepTime float64) float64 {
- timeout = timeout - sleepTime
- if timeout < 0.0 {
- t.Errorf("timeout waiting for condition '%s'", msg)
+ select {
+ case <-ctx.Done():
+ assert.FailNowf(t, "Timeout waiting for condition '%s'", msg)
+ return
+ case <-ticker.C:
+ }
  }
- time.Sleep(time.Duration(sleepTime) * time.Second)
- return timeout
 }
 
 func positionAtLeast(t *testing.T, tablet *Vttablet, a string, b string) bool {

go/test/endtoend/onlineddl/vtgate_util.go

@@ -207,7 +207,7 @@ func CheckLaunchAllMigrations(t *testing.T, vtParams *mysql.ConnParams, expectCo
 }
 
 // CheckMigrationStatus verifies that the migration indicated by given UUID has the given expected status
-func CheckMigrationStatus(t *testing.T, vtParams *mysql.ConnParams, shards []cluster.Shard, uuid string, expectStatuses ...schema.OnlineDDLStatus) {
+func CheckMigrationStatus(t *testing.T, vtParams *mysql.ConnParams, shards []cluster.Shard, uuid string, expectStatuses ...schema.OnlineDDLStatus) bool {
  query, err := sqlparser.ParseAndBind("show vitess_migrations like %a",
  sqltypes.StringBindVariable(uuid),
  )
@@ -229,7 +229,7 @@ func CheckMigrationStatus(t *testing.T, vtParams *mysql.ConnParams, shards []clu
  }
  }
  }
- assert.Equal(t, len(shards), count)
+ return assert.Equal(t, len(shards), count)
 }
 
 // WaitForMigrationStatus waits for a migration to reach either provided statuses (returns immediately), or eventually time out

go/test/endtoend/tabletgateway/buffer/reparent/failover_buffer_test.go

@@ -51,7 +51,7 @@ func failoverExternalReparenting(t *testing.T, clusterInstance *cluster.LocalPro
  primary.VttabletProcess.QueryTablet(demoteQuery, keyspaceUnshardedName, true)
 
  // Wait for replica to catch up to primary.
- cluster.WaitForReplicationPos(t, primary, replica, "localhost", 60.0)
+ cluster.WaitForReplicationPos(t, primary, replica, false, time.Minute)
 
  duration := time.Since(start)
  minUnavailabilityInS := 1.0