2023-10-19 12:19:10 +02:00
|
|
|
package setup
|
|
|
|
|
|
|
|
import (
|
|
|
|
"context"
|
|
|
|
"database/sql"
|
|
|
|
"embed"
|
|
|
|
"errors"
|
|
|
|
"strings"
|
|
|
|
|
2024-03-27 14:48:22 +01:00
|
|
|
"github.com/jackc/pgx/v5/pgconn"
|
2023-10-19 12:19:10 +02:00
|
|
|
"github.com/zitadel/logging"
|
|
|
|
|
|
|
|
"github.com/zitadel/zitadel/internal/database"
|
2024-01-25 17:28:20 +01:00
|
|
|
"github.com/zitadel/zitadel/internal/eventstore"
|
2023-10-19 12:19:10 +02:00
|
|
|
)
|
|
|
|
|
|
|
|
var (
|
|
|
|
//go:embed 14/cockroach/*.sql
|
|
|
|
//go:embed 14/postgres/*.sql
|
|
|
|
newEventsTable embed.FS
|
|
|
|
)
|
|
|
|
|
|
|
|
type NewEventsTable struct {
|
|
|
|
dbClient *database.DB
|
|
|
|
}
|
|
|
|
|
2024-01-25 17:28:20 +01:00
|
|
|
func (mig *NewEventsTable) Execute(ctx context.Context, _ eventstore.Event) error {
|
2023-10-19 12:19:10 +02:00
|
|
|
// if events already exists events2 is created during a setup job
|
|
|
|
var count int
|
perf(oidc): nest position clause for session terminated query (#8738)
# Which Problems Are Solved
Optimize the query that checks for terminated sessions in the access
token verifier. The verifier is used in auth middleware, userinfo and
introspection.
# How the Problems Are Solved
The previous implementation built a query for certain events and then
appended a single `PositionAfter` clause. This caused the postgreSQL
planner to use indexes only for the instance ID, aggregate IDs,
aggregate types and event types. Followed by an expensive sequential
scan for the position. This resulting in internal over-fetching of rows
before the final filter was applied.
![Screenshot_20241007_105803](https://github.com/user-attachments/assets/f2d91976-be87-428b-b604-a211399b821c)
Furthermore, the query was searching for events which are not always
applicable. For example, there was always a session ID search and if
there was a user ID, we would also search for a browser fingerprint in
event payload (expensive). Even if those argument string would be empty.
This PR changes:
1. Nest the position query, so that a full `instance_id, aggregate_id,
aggregate_type, event_type, "position"` index can be matched.
2. Redefine the `es_wm` index to include the `position` column.
3. Only search for events for the IDs that actually have a value. Do not
search (noop) if none of session ID, user ID or fingerpint ID are set.
New query plan:
![Screenshot_20241007_110648](https://github.com/user-attachments/assets/c3234c33-1b76-4b33-a4a9-796f69f3d775)
# Additional Changes
- cleanup how we load multi-statement migrations and make that a bit
more reusable.
# Additional Context
- Related to https://github.com/zitadel/zitadel/issues/7639
2024-10-07 15:49:55 +03:00
|
|
|
err := mig.dbClient.QueryRowContext(ctx,
|
2023-10-19 12:19:10 +02:00
|
|
|
func(row *sql.Row) error {
|
perf(oidc): nest position clause for session terminated query (#8738)
# Which Problems Are Solved
Optimize the query that checks for terminated sessions in the access
token verifier. The verifier is used in auth middleware, userinfo and
introspection.
# How the Problems Are Solved
The previous implementation built a query for certain events and then
appended a single `PositionAfter` clause. This caused the postgreSQL
planner to use indexes only for the instance ID, aggregate IDs,
aggregate types and event types. Followed by an expensive sequential
scan for the position. This resulting in internal over-fetching of rows
before the final filter was applied.
![Screenshot_20241007_105803](https://github.com/user-attachments/assets/f2d91976-be87-428b-b604-a211399b821c)
Furthermore, the query was searching for events which are not always
applicable. For example, there was always a session ID search and if
there was a user ID, we would also search for a browser fingerprint in
event payload (expensive). Even if those argument string would be empty.
This PR changes:
1. Nest the position query, so that a full `instance_id, aggregate_id,
aggregate_type, event_type, "position"` index can be matched.
2. Redefine the `es_wm` index to include the `position` column.
3. Only search for events for the IDs that actually have a value. Do not
search (noop) if none of session ID, user ID or fingerpint ID are set.
New query plan:
![Screenshot_20241007_110648](https://github.com/user-attachments/assets/c3234c33-1b76-4b33-a4a9-796f69f3d775)
# Additional Changes
- cleanup how we load multi-statement migrations and make that a bit
more reusable.
# Additional Context
- Related to https://github.com/zitadel/zitadel/issues/7639
2024-10-07 15:49:55 +03:00
|
|
|
if err := row.Scan(&count); err != nil {
|
2023-10-19 12:19:10 +02:00
|
|
|
return err
|
|
|
|
}
|
|
|
|
return row.Err()
|
|
|
|
},
|
|
|
|
"SELECT count(*) FROM information_schema.tables WHERE table_schema = 'eventstore' AND table_name like 'events2'",
|
|
|
|
)
|
|
|
|
if err != nil || count == 1 {
|
|
|
|
return err
|
|
|
|
}
|
|
|
|
|
perf(oidc): nest position clause for session terminated query (#8738)
# Which Problems Are Solved
Optimize the query that checks for terminated sessions in the access
token verifier. The verifier is used in auth middleware, userinfo and
introspection.
# How the Problems Are Solved
The previous implementation built a query for certain events and then
appended a single `PositionAfter` clause. This caused the postgreSQL
planner to use indexes only for the instance ID, aggregate IDs,
aggregate types and event types. Followed by an expensive sequential
scan for the position. This resulting in internal over-fetching of rows
before the final filter was applied.
![Screenshot_20241007_105803](https://github.com/user-attachments/assets/f2d91976-be87-428b-b604-a211399b821c)
Furthermore, the query was searching for events which are not always
applicable. For example, there was always a session ID search and if
there was a user ID, we would also search for a browser fingerprint in
event payload (expensive). Even if those argument string would be empty.
This PR changes:
1. Nest the position query, so that a full `instance_id, aggregate_id,
aggregate_type, event_type, "position"` index can be matched.
2. Redefine the `es_wm` index to include the `position` column.
3. Only search for events for the IDs that actually have a value. Do not
search (noop) if none of session ID, user ID or fingerpint ID are set.
New query plan:
![Screenshot_20241007_110648](https://github.com/user-attachments/assets/c3234c33-1b76-4b33-a4a9-796f69f3d775)
# Additional Changes
- cleanup how we load multi-statement migrations and make that a bit
more reusable.
# Additional Context
- Related to https://github.com/zitadel/zitadel/issues/7639
2024-10-07 15:49:55 +03:00
|
|
|
statements, err := readStatements(newEventsTable, "14", mig.dbClient.Type())
|
|
|
|
if err != nil {
|
|
|
|
return err
|
|
|
|
}
|
|
|
|
for _, stmt := range statements {
|
|
|
|
stmt.query = strings.ReplaceAll(stmt.query, "{{.username}}", mig.dbClient.Username())
|
|
|
|
logging.WithFields("file", stmt.file, "migration", mig.String()).Info("execute statement")
|
|
|
|
_, err = mig.dbClient.ExecContext(ctx, stmt.query)
|
2023-10-19 12:19:10 +02:00
|
|
|
if err != nil {
|
|
|
|
return err
|
|
|
|
}
|
|
|
|
}
|
|
|
|
return nil
|
|
|
|
}
|
|
|
|
|
|
|
|
func (mig *NewEventsTable) String() string {
|
|
|
|
return "14_events_push"
|
|
|
|
}
|
|
|
|
|
|
|
|
func (mig *NewEventsTable) ContinueOnErr(err error) bool {
|
|
|
|
pgErr := new(pgconn.PgError)
|
|
|
|
if errors.As(err, &pgErr) {
|
|
|
|
return pgErr.Code == "42P01"
|
|
|
|
}
|
|
|
|
return false
|
|
|
|
}
|