Timothy Messier
33037b4887
When sessions are deleted, a trigger is used to insert records into the session_deleted table. This table is utilized by the sessions list endpoint when using a refresh token to inform a client of the sessions that have been deleted since the last request. We delete sessions in bulk via a controller job to delete sessions that have been terminated over an hour ago, which results in the trigger running a large number of separate insert statements while processing the delete statement. This changes the trigger to run once for the delete statement, instead of for each row, resulting in a single bulk insert statement to the session_deleted table. This new trigger function also avoids the use of `on conflict`. When testing this function, while the single statement was still faster than running multiple inserts, the `on conflict` still added significant overhead, even when there were no conflicts. It should be safe to perform the insert without the `on conflict`, since the same ID should never be deleted more than once if it is successfully deleted. (cherry picked from commit 3569b36c6eb2bee7a4e36e7584fcc6bbf1124f95)
db package
Usage
Just some high-level usage highlights to get you started. Read the godocs for a complete list of capabilities and their documentation.
dialect = postgres.New(postgres.Config{
DSN: connectionUrl},
)
conn, _ := gorm.Open(dialect, &gorm.Config{})
// Db implements both the Reader and Writer interfaces
rw := Db{Tx: conn}
// There are writer methods like: Create, Update and Delete
// that will write Gorm struct to the db. These writer methods
// all support options for writing Oplog entries for the
// caller: WithOplog(yourWrapper, yourMetadata)
// the caller is responsible for the transaction life cycle of the writer
// and if an error is returned the caller must decide what to do with
// the transaction, which is almost always a rollback for the caller.
err = rw.Create(context.Background(), user)
// There are reader methods like: LookupByPublicId,
// LookupByName, SearchBy, LookupBy, etc
// which will lookup resources for you and scan them into your Gorm struct
err = rw.LookupByPublicId(context.Background(), foundUser)
// There's reader ScanRows that facilitates scanning rows from
// a query into your Gorm struct
where := "select * from test_users where name in ($1, $2)"
rows, err := rw.Query(context.Background(), where, []interface{}{"alice", "bob"})
defer rows.Close()
for rows.Next() {
user := db_test.NewTestUser()
// scan the row into your Gorm struct
if err := rw.ScanRows(rows, &user); err != nil {
return err
}
// Do something with the Gorm user struct
}
if err := rows.Err(); err != nil {
// do something with the err
}
// DoTx is a writer function that wraps a TxHandler
// in a retryable transaction. You simply implement a
// TxHandler that does your writes and hand the handler
// to DoTx, which will wrap the writes in a retryable
// transaction with the retry attempts and backoff
// strategy you specify via options.
_, err = w.DoTx(
context.Background(),
10, // ten retries
ExpBackoff{}, // exponential backoff
func(w Writer) error {
// the TxHandler updates the user's friendly name
return w.Update(context.Background(), user, []string{"FriendlyName"},
// write oplogs for this update
WithOplog(
InitTestWrapper(t),
oplog.Metadata{
"key-only": nil,
"deployment": []string{"amex"},
"project": []string{"central-info-systems", "local-info-systems"},
},
),
)
})